issues in testing business english - cambridgeenglish.org · an empirical investigation of the...
TRANSCRIPT
Issues in testing businessEnglish
The revision of the Cambridge BusinessEnglish Certificates
Also in this series:
An investigation into the comparability of two tests of English as a Foreign Language: TheCambridge-TOEFL comparability studyLyle F. Bachman, F. Davidson, K. Ryan, I.-C. Choi
Test taker characteristics and performance: A structural modeling approachAntony John Kunnan
Performance testing, cognition and assessment: Selected papers from the 15th Language TestingResearch Colloquium, Cambridge and ArnhemMichael Milanovic, Nick Saville
The development of IELTS: A study of the effect of background knowledge on readingcomprehensionCaroline Margaret Clapham
Verbal protocol analysis in language testing research: A handbookAlison Green
A multilingual glossary of language testing termsPrepared by ALTE members
Dictionary of language testingAlan Davies, Annie Brown, Cathie Elder, Kathryn Hill, Tom Lumley, Tim McNamara
Learner strategy use and performance on language tests: A structural equation modellingapproachJames Enos Purpura
Fairness and validation in language assessment: Selected papers from the 19th Language TestingResearch Colloquium, Orlando, FloridaAntony John Kunnan
Issues in computer-adaptive testing of reading proficiencyMicheline Chalhoub-Deville
Experimenting with uncertainty: Essays in honour of Alan DaviesA. Brown, C. Elder, N. Iwashita, E. Grove, K. Hill, T. Lumley, K. O’Loughlin, T. McNamara
An empirical investigation of the componentiality of L2 reading in English for academic purposesCyril Weir
The equivalence of direct and semi-direct speaking testsKieran O’Loughlin
A qualitative approach to the validation of oral language testsAnne Lazaraton
Continuity and innovation: Revising the Cambridge Proficiency in English Examination1913–2002Edited by Cyril Weir and Michael Milanovic
European language testing in a global contextEdited by Cyril Weir and Michael Milanovic
A modular approach to testing English language skills: The development of the Certificates in English Language Skills (CELS) examinationsRoger Hawkey
Changing language teaching through language testing: A washback studyLiying Cheng
UnpublishedThe impact of high-stakes examinations on classroom teaching: A case study using insights from testing and innovation theoryDianne Wall
Impact theory and practice: Studies of the IELTS test and Progetto Lingue 2000Roger Hawkey
Issues in testing businessEnglish
The revision of the Cambridge BusinessEnglish Certificates
Barry O’Sullivan
iv
C A M B R I D G E U N I V E R S I T Y P R E S S
Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo
Cambridge University PressThe Edinburgh Building, Cambridge CB2 2RU, UK
www.cambridge.orgInformation on this title: www.cambridge.org/9780521013307
© UCLES 2006
This publication is in copyright. Subject to statutory exceptionand to the provisions of relevant collective licensing agreements,no reproduction of any part may take place without the written permission of Cambridge University Press.
First published 2006
Printed in the United Kingdom at the University Press, Cambridge
A catalogue record for this publication is available from the British Library
ISBN-13 978-0-521-01330-7 paperbackISBN-10 0-521-01330-5 paperback
The author is grateful to the copyright holders for permission to use the copyrightmaterial reproduced in this book. Every effort has been made to trace the copyrightholders. Cambridge University Press apologises for any unintentional omissions andwould be pleased, in such cases, to add an acknowledgement in further editions.
Reprinted 2006
To Maura
v
vii
Contents
Acknowledgements ix
Series Editors’ note x
Abbreviations xii
Chapter 1Introduction to the testing of language for business purposes 1
Chapter 2The revision of BEC 82
Chapter 3Major changes to the suite 119
Chapter 4Changes in the BEC papers 130
Chapter 5Conclusions and the way forward 174
AppendicesAppendix 1.1: JOCT Evaluation Criteria 196Appendix 1.2: CEFLS Pilot Test 197Appendix 1.3: CEIBT – Test of Reading and Writing –
June and November 1992 205Appendix 1.4: BULATS – Standard Test English 225Appendix 1.5: BULATS – Speaking Test 241Appendix 1.6: BULATS – Standard Test German 244Appendix 2.1: ALTE Work Typical Abilities 260Appendix 3.1: BEC 1 Sample Paper 261Appendix 3.2: BEC 2 Sample Paper 277Appendix 3.3: BEC 3 Sample Paper 291Appendix 4.1: BEC Preliminary Sample Paper 302Appendix 4.2: BEC Vantage Sample Paper 328Appendix 4.3: BEC Higher Sample Paper 353
viii
Contents
Additional information on tests of language for business purposes 378
References 381
Index 393
Acknowledgements
I would like to thank the many people at Cambridge ESOL who contributed tothis volume. I was lucky enough to be able to interview many of the individualswho were involved in the development and administration of their tests ofEnglish for business purposes, particularly the BULATS and BEC groups. Inparticular, I am indebted to Hugh Bateman, without whom the book could nothave been completed. Others I would like to single out include Mike Milanovicand Nick Saville, who provided historical information and access to internalreports, and Neil Jones, who provided documentation on the grading proceduresfor BEC. I would also like to thank David Thighe who clarified a number ofgrading-related issues, and who made valuable comments on the later drafts ofthis book. Finally, I would like to thank Rowena Akinyemi for all her work inensuring that this book made it to press.
I would also like to thank the following individuals and institutions forproviding information on their tests and for giving their permission for itemsfrom the tests to be included in the book:
Certificazione della conoscenza dell’italiano commerciale (Certificate in Italianfor Commerce – CIC): Professor Giuliana Grego Bolli and Francesca Parizzi ofthe Università per Stranieri di Perugia.JETRO tests: Professor Kiyokata Katoh of Tokyo Gakugei University andReiko Kimura of the Japan External Trade Organization.Pitman tests: Glyn Jones of City and Guilds, London.Cambridge ESOL tests: Dr Mike Milanovic of Cambridge ESOL.Table on page 101: reproduced by permission of Oxford University Press fromOxford Applied Linguistics: Fundamental Considerations in Language Testingby Lyle F Bachman © Lyle F Bachman 1990.
Finally, I would like to thank Professor Cyril Weir who read various parts of thebook and offered invaluable critical comments and advice.
ix
x
Series Editors’ note
The language testing world has flirted with the testing of English for specificpurposes for many years. In small scale testing contexts there have been andcontinue to be numerous specific assessments tailored to particular needs but inthe context of large scale international language testing, specific purpose assess-ments have been far less common.
Cambridge ESOL started testing English in 1913. In some ways you mightconsider the original Certificate of Proficiency in English (CPE) a specificpurpose examination designed to meet the needs of those teaching Englishalthough over the years it has become far more general in its emphasis.Cambridge ESOL also collaborated with the British Council on the devel-opment of the English Language Testing Service (ELTS) in the 1970s which hadsix subject specific modules. IELTS, which evolved from ELTS in 1990, saw areduction in the academic modules to 3 and the 1995 revision of IELTS led to thesingle academic module alongside a general training variant.
There are many reasons for this retreat by IELTS, both practical andtheoretical. Developing numerous multiple modules in the quantities requiredwas a far from easy task. Ensuring that candidates took the right module wasdifficult. Equating modules proved technically very demanding. Ensuringcontent appropriateness required access to experts in a number of fields and soon. However, the need to extend the remit of General English assessmentremains. The context of Cambridge ESOL English language assessmentcurrently falls into four broad categories. Academic English (IELTS), BusinessEnglish (BEC and BULATS), Young Learners’ English (YLE) and GeneralEnglish (KET, PET, FCE, CAE, CPE and CELS) and in 2006 Cambridge ESOLwill launch the International Legal English Certificate. This test seeks to addressmore specifically English in the legal domain of use.
In Issues in testing business English, Barry O’Sullivan provides a frameworkfor classifying and understanding specific purpose language assessment. Thefirst part of the volume provides the reader with a comprehensive review ofnumerous business English tests as well as business language tests in otherlanguages. Some of the tests described no longer exist so the volume also servesas a useful historical record. This is followed by a detailed look at the revision ofthe Cambridge Business English Certificates (BEC).
Chapter 1 considers the relationship between general English and English forspecific purposes and the definition of a business English construct. O’Sullivanpresents a continuum ranging from an unspecified purpose to one that is highly
xi
specified. To this he adds construct, test method, skills coverage measurementqualities, degree of specificity/authenticity, non language factors and thereporting of test performance. This provides him with a framework forcomparison and he proceeds to evaluate a series of business language tests onthis basis. Particular attention is paid to some very widely used tests such asTOEIC, BULATS and BEC although the coverage of less widely known assess-ments is comprehensive. The text is illustrated with numerous examples of testitem types which make interesting reading.
Having provided a detailed context against which to understand BEC, subse-quent chapters consider BEC’s revision and look in detail at each of the threeBEC levels. The discussion of development methodology is interesting as is thefocus on test reliability. It is gratifying to note that an examination like BEC,operating on a truncated sample of the test taking population at each of its threelevels, demonstrating very good construct and content validity features andusing a good variety of realistic material with an authentic orientation, cannonetheless achieve respectably high reliability estimates. Throughout thisvolume readers are referred to Volume 15 (Weir, Cyril and Milanovic, Michael(Eds) (2003) Continuity and innovation: Revising the Cambridge Proficiency inEnglish Examination 1913 – 2002 ) in the same series which gives an even moredetailed account of the principles that underline the Cambridge approach to testdevelopment and validation. The appendix has a comprehensive set of BECmaterials but is complemented by a focus on two other tests, the Certificate inEnglish as a Foreign Language for Secretaries and the Certificates in English forInternational Business and Trade which informed the development of BEC butare no longer available.
The final chapter is particularly important as it discusses in some detail theissue of authenticity and its relationship to the specificity continuum linking theargument in with Weir’s validation framework (Cyril J. Weir (2004) LanguageTesting and Validation: An Evidence-Based Approach, Palgrave Macmillan).
O’Sullivan presents a multicomponential view of specificity and is able toclearly distinguish between different tests and tasks using his approach. Thevolume concludes with a focus on future research suggestions, part of whichwas arrived at collaboratively with staff at Cambridge ESOL.
Issues in testing business English is the third volume in this series (the othertwo being volumes 15 and 16) to document both a historical perspective and astudy of test revision with a focus on the implications this has. A volume onacademic English assessment authored by Alan Davies is forthcoming. Thisvolume documents the history of the assessment of English for academicpurposes from the 1950s to the present with a particular focus on the devel-opment and validation of IELTS.
Michael MilanovicCyril Weir
Cambridge 2005
Series Editors’ note
xii
Abbreviations
ALTE Association of Language Testers in EuropeAPA American Psychological AssociationBEC Business English CertificateBULATS Business Language Testing SystemCAE Certificate in Advanced EnglishCAL Center for Applied LinguisticsCAT Computer Adaptive TestCBT Computer-Based TestCEF Common European Framework CEFLS Certificate in English as a Foreign Language for SecretariesCEIBT Certificate in English for International Business and TradeCIC Certificate in Italian for Commerce (Certificazione della
conoscenza dell’italiano commerciale)CPE Certificate of Proficiency in EnglishDIF Differential Item FunctioningEAP English for Academic PurposesEBC English for Business CommunicationEFL English as a Foreign LanguageELT English Language TeachingEOS English for Office SkillsESOL English for Speakers of Other LanguagesESP English for Specified PurposesETS Educational Testing ServiceFCE First Certificate in EnglishGIMS General Impression Mark SchemeGQ General QuestionnaireIATEFL International Association of Teachers of English as a Foreign
LanguageIELTS International English Language Testing SystemIRT Item Response TheoryJETRO Japan External Trade OrganizationJOCT JETRO Oral Communication TestJRLT JETRO Reading and Listening Comprehension TestKCQ Key Contacts QuestionnaireKET Key English TestLCCIEB London Chamber of Commerce and Industry Examinations Board
LSP Language for Specific PurposesMCQ Multiple-Choice QuestionOET Occupational English TestOIBEC Oxford International Business English CertificateUODLE University of Oxford Delegacy of Local ExaminationsPET Preliminary English TestPLAB General Medical Council’s Professional and Linguistic
Assessments Board (test of overseas doctors’ language proficiency)
QCA Qualifications and Curriculum AuthorityRITCME Recruitment, Induction, Training, Co-ordination, Monitoring,
EvaluationRSA Royal Society of ArtsSAQ Short Answer QuestionTAAS Texas Assessment of Academic SkillsTEEP Test of English for Educational PurposesTFI Test de français internationalTOEFL Test of English as a Foreign LanguageTOEIC Test of English for International CommunicationUCLES University of Cambridge Local Examinations SyndicateVRIP Validity, Reliability, Impact and Practicality
xiii
Abbreviations
1
Introduction to the testing oflanguage for businesspurposes
A brief historical introductionThough there have been formal tests of general proficiency around for manyyears – see Weir (2003a) for an interesting and informative historicalperspective on the Certificate of Proficiency in English (CPE) since its intro-duction in 1913 – interest in language for specific purposes has a far shorterhistory, emerging, according to Swales (1984:11) with Barber’s (1962) SomeMeasurable Characteristics of Modern Scientific Prose. This is not to say thatthere has been an awareness of the use of language for specific purposes only inrecent times. Schröder reminds us:
. . . when new counting house regulations were issued for the London Salhofin 1554, these stated amongst other things that young apprentices fromGermany would have to spend one year with a clothmaker in the country, sothat they might get a proper command of everyday English and the morespecific technical terms . . . (1981:43).
Much of the early work in the area was driven by research which focused onthe identification of unique instances of language use in specific contexts(Hüllen 1981a, 1981b, Johns 1980, Lackstrom, Selinker and Trimble 1973,Selinker and Douglas 1985, Swales 1971, to list but a few), the issue of authen-ticity in the use of materials for teaching (e.g. Carver 1983) and the central placeof needs analysis in identifying the specific language needs of learners in givencontexts (Alwright and Alwright 1977, Brindley 1984, Gledhill 2000, Hawkey1978, Hutchinson and Walters 1987, Kennedy and Bolitho 1984, LCCIEB1972, Robinson 1980, 1985, Thurstun and Candlin 1998, West 1994). As can beseen from the dates of these publications, much of the English for SpecificPurposes (ESP) debate was conducted almost twenty years ago, yet many of thesame questions continue to be asked today.
Hawkey (2004) outlines the changes in theories of language learning andteaching that lead to the development of a clearly defined ESP methodology,and led to an awareness of the need to establish a set of clearly rationalised testing procedures. In the case of the testing of language for
1
1 Introduction to the testing of language for business purposes
2
business purposes, the first test to emerge was the Test of English for Interna-tional Communication (TOEIC). It was developed by Educational TestingServices (ETS) in the USA and introduced in 1979. The test, originally devisedfor the Japanese market, was based firmly on psychometric–structuralist theory(Spolsky 1995) and represents one of the few remaining (though highlysuccessful from a commercial perspective) examples of a multiple-choiceformat, standardised, international language test.
While the TOEIC looked backwards for its theoretical underpinning, othertests of business language, particularly those developed in the UK, werebeginning to look to a more communicative model. Theorists on communicativecompetence, particularly Canale and Swain (1980), Hymes (1972) and practi-tioners like Munby (1978) had a profound influence on the practice of languageteaching and testing. One major influence was the facilitation of a movementaway from the psychometric–structuralist methodology, based on the teachingand testing of discrete aspects of language, to the psycholinguistic–sociolinguistic era, where language teaching and testing were seen from aholistic or integrated perspective. The shift in emphasis in language teachingfrom language knowledge to language use paved the way for a testing method-ology which reflected the same ideas. Hawkey (2004) traces the historical devel-opment of the theoretical movements of this period and provides acontextualisation for the emerging interest in the teaching and later testing ofESP. With the exception of the TOEIC, the tests described in the followingsections have an essentially performance-based orientation in which emphasisis placed on the contextualisation of the tasks and predicted linguistic responseswithin the business setting.
In the mid-1980s the move to the testing of language for business purposes inthe UK began in earnest with the development by the Royal Society of Arts(RSA) of the Certificate in English as a Foreign Language for Secretaries(CEFLS) – which was later administered as the Certificate in English for Inter-national Business and Trade (CEIBT) – and a corresponding move by theLondon Chamber of Commerce and Industry Examinations Board (LCCIEB)and Pitman (now part of the City and Guilds Examinations Board) to createlanguage tests with a business focus. When the RSA was subsumed into theUniversity of Cambridge Local Examinations Syndicate (UCLES) in 1988 theRSA test was administered by UCLES, establishing its portfolio of languagetests for business.
In the early 1990s two new examinations, the Business English Certificate(BEC) and Business Language Testing System (BULATS) were developed byUCLES. It is the former of these tests that forms the basis for the latter part of thisbook, in which the procedures used by Cambridge ESOL in the BusinessEnglish Certificate (BEC) suite revision are outlined and exemplified.
During the mid- to late-1990s a number of tests of other languages forbusiness emerged. These included JETRO (Japanese), Test de français interna-
Theoretical perspectives
3
tional (TFI) from the makers of TOEIC, the Certificate in Italian for Commerce(CIC) and the tests in the BULATS series (French, German and Spanish inaddition to the English version).
There is clearly a growing interest in the area of testing language for businesspurposes, particularly with the internationalisation of business and the need foremployees to interact in more than just a single language. The move towards a‘business language’ testing genre is reflected in the tests mentioned above anddescribed in the latter part of this chapter.
Theoretical perspectivesIn the only serious attempt to date to build a theoretical rationale for the testingof language for specific purposes, Douglas (2000) argues that a theoreticalframework can be built around two principal theoretical foundations. The first ofthese is based on the assumption that language performance varies with thecontext of that performance. This assumption is supported by a well establishedliterature in the area of sociolinguistics – see for example Labov’s (1963) classicstudy of vowel change on Martha’s Vineyard – in addition to research in theareas of second language acquisition (Dickerson 1975, Ellis 1989, Schmidt1980, Smith 1989, Tarone 1985, 1988) and language testing (Berry, 1996, 1997,Brown 1995, 1998, Brown and Lumley 1997, O’Sullivan 1995, 2000a, 2000b,2002a, Porter 1991a, 1991b, Porter and Shen, 1991). This fits well with thegrowing interest in a socio-cognitive approach to language test developmentwhere performance conditions are seen to have a symbiotic relationship with thecognitive processing involved in task completion (introduced by O’Sullivan2000a and discussed in detail by Weir 2004).
In the case of the second foundation, Douglas sees specific purpose languagetests as being ‘precise’ in that they will have lexical, semantic, syntactic andphonological characteristics that distinguish them from the language of more‘general purpose’ contexts. This aspect of Douglas’s position is also supportedby an ever increasing literature, most notably in the area of corpus-based studiesof language in specific contexts (Beeching 1997, Biber et al 1998, Dudley-Evans and St John 1996, Gledhill 2000, Thurstun and Candlin 1998).
When it came to an actual definition of specific purpose tests, Douglas placesthese two foundations within a single overriding concept, that of authenticity,defining a test of specific purposes as:
One in which test content and methods are derived from an analysis of aspecific purpose target language use situation, so that test tasks and contentare authentically representative of tasks in the target situation, allowing foran interaction between the test taker’s language ability and specific purposecontent knowledge, on one hand, and the test tasks on the other. Such a testallows us to make inferences about a test taker’s capacity to use language inthe specific purpose domain (2000:19).
1 Introduction to the testing of language for business purposes
4
This definition highlights the core element of Douglas’s view of LSP tests; thatof authenticity. Douglas does not see this as being a simple matter of replicatingspecific purpose tasks in a testing context, but of addressing authenticity fromtwo perspectives. The first perspective is that of situational authenticity, whereLSP test tasks are seen as being ‘authentic’ in that they are derived from ananalysis of the language use domain with which they are associated. The secondperspective is interactional authenticity, which relates to the actual processingthat takes place in task performance, what Weir (2004) refers to as theory-basedvalidity.
This definition has not remained unquestioned. In fact, Douglas (2001)himself acknowledges that there are a number of issues left unanswered by hisdefinition, an argument also made by Elder (2001). This criticism focuses onwhat Elder (2001) sees as the three principal problematic areas identified in thework of Douglas, namely, the distinguishability of distinct ‘specific purpose’contexts; authenticity; and the impact (and interaction) of non-language factors.
By non-language factors one of two things is meant. The first relates to theelements of communication not associated with language – in everyday commu-nication, transferral of message is achieved through a combination of language,cues, signals and symbols. There is a broad literature in psychology on thisphenomenon (see for example Brown, Palmeta and Moore 2003, Vargo 1994).The second way of looking at this is the impact of background knowledge, in thiscase of the business domain, on an individual’s ability to perform a particulartask, in this case related to an aspect of business communication.
The first of these two perspectives is common across all tests of languageproduction, not solely Language for Specific Purposes (LSP) tests. It is not justrelated to tests of speaking, where variables such as physical appearance, dress,gestures and posture have all been shown to have an effect on interlocutorperceptions of performance (see for example the work in the area of job inter-views of Bordeaux 2002, Chia et al 1998, and Straus, Miles and Levesque 2001),but is also to be seen in tests of writing where handwriting and general presen-tation skills impact on how writing is evaluated by examiners (see for exampleSprouse and Webb 1994, Sweedler-Brown 1992). This aspect of performanceassessment is certainly a potential threat to test validity, and is typically dealtwith in the development of assessment scales or, more likely, throughrater/examiner training.
The latter perspective, the extent to which candidates’ backgroundknowledge impacts on his/her test performance is again not associated solelywith LSP tests. A test of language for specific purposes is situated, by its verynature, in a specific context, and, also by its very nature, expects (if notdemands) of its candidates a knowledge of that context. The literature has shownthat background knowledge has a significant and apparently systematic effecton LSP test performance (see for example Alderson and Urquhart 1984, 1985,1988, Clapham 1996, Steffensen and Joag-Dev 1984). It also appears that as a
Distinguishing LSP from general English
5
test becomes more highly specific this effect becomes more acute and it wouldseem that it is at this extreme that the difficulty in teasing apart languageperformance and task completion occurs – in other words, in a highly specifictest, success on a task is dependent on a successful interplay of language andnon-language elements. This feature of highly specific tests at one time led toinnovations such as in the General Medical Council’s Professional andLinguistic Assessments Board (PLAB) oral test where medics assessed themedical content of ESP tasks and the language examiner commented on thelanguage performance (both informal with patients and formal with profes-sional colleagues, on a generic ELT scale) though specialist lexis etc. remainedthe domain of the subject specialists.
It can be argued that a test of language for a specific purpose should not eventry to avoid the background knowledge issue, as it is this that defines the test.How we deal with the situation will depend on the degree of specificity of thetest and the inferences we intend to draw from performance on the test.
Turning to the remaining criticisms of an ESP approach to testing, we can seethat there are basically two questions that should be addressed. These are:
1. Distinguishing LSP from general language – is it possible and/or feasible?2. Authenticity – can LSP tests be made both situationally and interactionally
authentic?
Distinguishing LSP from general EnglishThere is a considerable body of work over the last thirty years which has quiteclearly demonstrated the distinguishability of language use in specific contexts.We can point to the work on the definition of language needs and usage inspecific contexts of needs analysis researchers and theorists. Among theinfluential early work were studies undertaken by Hawkey (1978), who offereda practical demonstration of how needs analysis can lead to a specific purposecurriculum, and Alwright and Alwright’s (1977) practical advice on anapproach to the teaching of medical English.
In the area of testing language for specific purposes, perhaps the mostimportant undertaking was that of the London Chamber of Commerce andIndustry Examinations Board (LCCIEB) in 1972. The LCCIEB had beenproviding business-related qualifications around the world for almost a hundredyears when, in 1972, its language section undertook a major analysis of ‘foreign’language use involving over 11,500 employees of almost six hundred interna-tional firms. This analysis, and the replications undertaken in the FederalRepublic of Germany, France, Greece and Spain between 1982 and 1985, wereto prove influential in the development of teaching and testing practice in the UKduring the 1970s and 1980s.
In a series of seminal articles in the 1980s, Alderson and Urquhart (1984,
1 Introduction to the testing of language for business purposes
6
1985, 1988) found that ‘academic background can play an important’ thoughnot consistent ‘role in test performance’ (Alderson and Urquhart 1985:201) andthat ‘particular groups of students may be disadvantaged by being tested onareas outside their academic field’ (Alderson and Urquhart 1988:182). Theyalso suggested that their studies ‘demonstrated the need to take account of otherfactors, such as linguistic proficiency’ (Alderson and Urquhart 1985:201). Atabout the same time Steffensen and Joag-Dev (1984) demonstrated thesignificant impact on comprehension of a reader’s cultural background. Thepicture that is developing here is that background knowledge is a significantfactor in specific purpose language testing, a point that was made by Clapham(1996) with reference to highly specific tests.
In fact, Clapham’s (1996) study provided quite a few answers, or at leastdirections in which to look for answers, to many of the questions asked about theimpact of background knowledge on performance in LSP tests. While looking atperformance on a test of English for academic purposes (International EnglishLanguage Testing System IELTS), Clapham’s interpretation of the results ofher in-depth and complex study have direct consequences for the testing oflanguage for any specific purpose. It is therefore worth looking back overClapham’s work. Among other things, Clapham reports that:
• . . . students achieved significantly higher scores on the module in their ownsubject area than on the module outside it (1996:188) … [though] the resultsdepend on the specificity of the tests (1996:189)
• . . . it is possible to identify some of the characteristics which lead topassages being more or less specific, but that these characteristics are notalways immediately obvious (1996:191) . . . [though] it was the rhetoricalfunction of the passages rather than the sources of the texts which affectedtheir specificity (1996:191)
• it is not always easy to classify candidates into simply defined subgroups, asthe evidence from Clapham indicates that her participants were widely readoutside of their own area of study (1996:192–3)
• it seems likely that as the modules became more subject specific,background knowledge had a proportionally stronger effect on test scores(1996:193). In addition, subject area familiarity made a significantcontribution to test scores, whereas topic familiarity did not . . . [this]suggests that knowledge of a subject area might have a greater effect thantopic familiarity on the subject specificity of a reading passage (1996:193)
• there seemed to be a threshold below which students did not make use of this[background] knowledge, and above which they did (1996:194).
The implications of the work referred to earlier in the chapter (e.g. Barber1962, Hüllen 1981a, 1981b, Johns 1980, Lackstrom, Selinker and Trimble1973, LCCIEB 1972, Schröder 1981, Selinker and Douglas 1985, Swales 1971,1984, Weir 1983) when seen in light of these findings suggest that there is a
Authenticity
7
clearly definable language of business (and of other areas of specific interestsuch as science, technology etc.) and that where tests are devised with a deliber-ately high level of specificity towards an explicit area, then candidates whosebackground is grounded in that area can be expected to outperform candidatesfrom a different background, given similar linguistic competence.
There is still a problem, however, in defining the boundaries of specificcontext areas (Cumming 2001, Davies 2001, Elder 2001). It appears to be thecase that while we can identify particular aspects of language use as beingspecific to a given context (such as vocabulary, syntax, rhetorical organisation),we cannot readily identify exact limits to the language that is used in thatcontext. This is because there are no ‘exact limits’. Business language, likescientific or medical language is situated within and interacts with the generallanguage domain, a domain that cannot, by its very nature, be rigidly defined.
AuthenticityThough Douglas (2000) built his definition of what makes a test ‘specific’around the notions of situational and interactional authenticity, he later(Douglas 2001) pointed to some difficulties in operationalising such adefinition. The notion of situational authenticity is relatively easy to conceptu-alise. Situational authenticity refers to the accurate reflection in the test design ofthe conditions of linguistic performance from the language use domain – Weir’s(2004) text and task demands. Tests such as that for air traffic controllersdescribed by Teasdale (1994), where candidates were tested in a situation thatclosely replicated the specific purpose domain, are as close as we can get to acompletely situationally authentic test. The mere fact that the event is being usedas a test lessens the authenticity – though I’m sure that few readers would expectthat the ability of air traffic controllers to cope linguistically with the demands oftheir work should be tested in a truly authentic situation! The opposite to thiswould be the relative situational inauthenticity of the MATHSPEAK test, thespecific purpose version of the SPEAK (the institutional form of the Test ofSpoken English, the TSE) referred to by Elder (2001), where there is no attemptmade to replicate the teaching context it is designed to be generalised to.
However, in the case of interactional authenticity there is a lesser degree ofcertainty in that, to the present time, it has not been clearly conceptualised, letalone operationalised. Though the common view (that the test should result in aninteraction between the task and the relevant language ability) is clear enough,to my knowledge there has not been a significant contribution to its operational-isation – that is, insufficient work has been done to link context-based validityelements to theory-based processing. Test developers and researchers tend torely on anecdotal evidence or ‘expert’ judgements to make decisions on theinteractional authenticity of a test task – in the review of a range of businesslanguage tests that comes later in this chapter, I fall foul of the same tendency.
1 Introduction to the testing of language for business purposes
8
So, critics of an LSP approach to language testing have raised genuineconcerns regarding the distinguishability of distinct ‘specific purpose’ contexts,authenticity, and the impact on test performance of non-language factors – notjust for LSP testing but for language testing in general. I do not believe that theseare insurmountable and I will return to the matter in the final chapter of thisbook.
Assessing performanceWhile the above issues have focused on the test content and on the theoreticaljustification for utilising a particular test task, there are other issues in LSPtesting that have not really been addressed. Like any test, the reliability(stability, consistency and accuracy) of LSP tests is central to the test’s value. Inthe section devoted to reliability in the context of the BEC suite (Chapter 2) Ilook in some detail at this issue, so I will not spend time or space here in anextended discussion, except to say that the way we estimate and report the relia-bility of tests such as the BEC suite is in need of re-appraisal as the statisticalapproaches taken to date offer us only a limited understanding of the true relia-bility of these tests.
A related issue is the way in which we evaluate or assess writing and speakingtest performances, in that it is associated with the creation of the test score,which is central to any test.
There are a number of issues here:
• the scale criteria• the level represented by the scale• the use of the scale (who, how etc.).
The scale criteria
Though the literature abounds with scales that do not seem to have been derivedfrom any particular theoretical or empirical base, the movement in the 1990stowards more supportable scale development means that the current ratingscales which reflect best practice in the area tend to have a sound basis (see North1996, North and Schneider 1998). While the whole area of rating scale devel-opment is far too complex to be dealt with adequately in this short section, it isimportant to point to the need for any rating scale to be based on the same modelor perception of language as drives the rest of the test development process. Agood example of this are the rating scales used in the Cambridge ESOL MainSuite examinations (Hawkey 2001).
In their response to the criticisms voiced by Foot (1999), Saville andHargreaves (1999) present a model of communicative ability upon which theCambridge ESOL Main Suite speaking examinations are based (see Figure 1.1).
Assessing performance
9
This model is based on the earlier work of Canale and Swain (1980) andBachman (1990), as well as on the Council of Europe specifications for theWaystage and Threshold levels of competence (Saville and Hargreaves 1999:46).
We can see that language competence is described in terms of Bachman(1990:84–98) and Bachman and Palmer’s (1996:67) organisational (grammarand discourse), pragmatic and strategic competences.
Figure 1.1 Communicative language ability
Source: Saville and Hargreaves (1999:45)
The rating scales used in the Cambridge Main Suite Speaking paper examinations consist of four criteria, grammar and vocabulary, discoursemanagement, pronunciation and interactive communication, each of which isawarded a score in the range of 0–5. Though it is not clear from Saville andHargreaves exactly how the scale is meant to reflect the model of competencethey quote, it would appear that it is meant to operate as represented inFigure 1.2.
It is clear from this figure that the notion of pragmatic competence is notexplicitly dealt with in the scales (for convenience, only the middle score of 3 ispresented in this figure, though the descriptions offered here are similar to theother levels in terms of relevance to model criteria). The notion of pragmaticcompetence (or knowledge) is seen by Bachman and Palmer as being related tothe ability to ‘create or interpret discourse by relating utterances or sentencesand texts to their meanings’ (1996: 69). In other words, pragmatic competence isseen as being comprised of functional and sociolinguistic knowledge and assuch has been identified here with the criterion discourse management – which,though the name implies an ability to ‘manage’ the interaction (in the sense ofBygate 1987), in the context of this scale it is actually concerned with coherence,
Grammatical
SyntaxMorphologyVocabulary
Pronunciation
Rhetorical OrganisationCoherenceCohesion
e.g. Sensitivity to
illocution
Interaction skillsNon-verbal features of
interaction
Discourse Pragmatic
Spoken Language Ability
Language Competence Strategic Competence
1 Introduction to the testing of language for business purposes
10
cohesion and, if this representation is accepted, an ability to demonstratefunctional and sociolinguistic competence.
Figure 1.2 Communicative language ability and the Cambridge ESOLFCE analytic scale
When advocating a move towards an integrated language/specific areaability approach, Douglas (2000) suggests using what he refers to as‘indigenous’ scales in LSP tests. The argument being that the criteria actuallyemployed in the evaluation of specific purpose performances are specific to thecontext of that performance – a position which is seen as support for the insepa-rability of language and performance of specific purpose tasks (Douglas 2001,Elder 2001). While the case made by Douglas is strong, there are a number ofpoints which still need further consideration.
The central problem here is one of construct definition, and therefore of theinferences that are to be drawn from a particular test. In the case of the Occupa-tional English Test (OET), for instance, which is criticised by Douglas and by itsprinciple creator, McNamara (in Jacoby and McNamara 1999) for using a‘general purpose’ rating scale, rather than one devised from an analysis of thetarget language use (TLU) situation, the criticism has some basis, in that thescale used was a rather primitive adaptation of the FSI oral proficiency scale(Wilds 1975). However, the test, for whatever reason (the one suggested wasbureaucratic expedience) was meant to offer a measure of the ability of overseashealth professionals to cope with the English language demands of theirparticular medical specialisation. The inferences to be drawn from performanceon the test were therefore related to their language competence, nothing else. In
3.0
Band
Grammar is sufficiently
accurate. Usesappropriate
vocabulary indealing with the
tasks.
Grammar andVocabulary
Uses adequaterange of linguisticresources to dealsufficiently wellwith the tasks.
Contributions mayoccasionally belimited or lack
coherence
DiscourseManagement
Produces individualsounds and
prosodic featuressufficiently well to
be understood. L1 accent may
cause occasionaldifficulty.
Pronunciation
Has sufficient interactive ability tocarry out the tasks. Maintains flowof language when carrying out the
tasks although may occasionally lack sensitivity to turn
taking and hesitation may occurwhile searching for
language. Does not require majorassistance or prompting to carry
out the tasks.
Interactive Communication
Grammatical Discourse Pragmatic
Strategic CompetenceLanguage Competence
Assessing performance
11
this respect, the OET appears to have been a successful test. If it were to becomea ‘true’ performance measure (in that it should offer a measure of the test taker’sability to perform the particular medical duties under scrutiny) then clearly adifferent approach to the evaluation of the performance would be needed. It maybe, for instance, that the same role-played performance could be used as alanguage measure and, when subjected to scrutiny using ‘indigenous’ criteria(which might include an aspect of language), serve to offer evidence of medicalability (see the reference on page 5 to the PLAB test in the UK).
The level represented by the scale
Scales can be designed to represent a whole range of ability levels, for examplesee the sample band descriptors for the Test of English for Educational Purposes(TEEP) from the University of Reading – Figure 1.3 (O’Sullivan 1999). As wecan see from this figure, this scale ranges from a level of non-language to that ofvery high competence in the language and is obviously designed to be usedacross the whole ability range.
When a test is designed to measure language ability on or around a particularproficiency level – for example if we are planning to design a test of writing forcandidates at the Common European Framework (CEF) level B2 (Vantage) weare faced with a bit of a conundrum. If we decide to create a scale to describeability across all levels (see Figure 1.4), with only the portion corresponding toB2 in use for this particular test, we are faced with either making simpletrichotomous decisions (the candidate is below this level, at this level or abovethis level), or describing multiple levels of ability within each of the six abilitylevels. This would make the scale both extremely difficult to develop andvalidate and also very difficult if not impossible to use, as raters would be facedwith the same problem they met in trying to use the scales devised by Fulcher(1996) where the ‘thick’ description of typical performance at each scale levelwas so detailed that the scale became unusable.
Another option is to create a single scale, which is then interpreted atwhichever ability level it is to be used at (say C1 or A2). With this type of scale,there is increased pressure on the developer to ensure that the scale is sufficientlyclear so as to ensure that users can easily distinguish the different levels ofperformance within the scale, but sufficiently general to allow the scale to beinterpreted at the different levels of ability. While of great practical use, this typeof scale is not easy to develop and validate and depends on examiner/ratertraining and monitoring if it is to be successfully used.
The most commonly used method is to create individual scales for use at eachlevel. In order to ensure that the scales are identifying appropriate levels ofachievement at each level they must be linked in some way. This processinvolves a major investment in resources – and the resulting scale is still
1 Introduction to the testing of language for business purposes
12
dependent on rater training (though not to the extent of the option suggestedabove). However, evidence of improvements in inter- and intra-rater reliabilitysuggest that this approach is viable (see Hawkey 2001 and Hawkey and Barker2004).
The use of the scale (who, how etc.)
The remaining issues associated with the rating scale relate to who should beinvolved in the development and application of the scale. The first of theseissues centres around the content of the scale – to what extent can we define a setof criteria that will offer a valid framework through which a test performance
Figure 1.3 The overall impression scale from the Test of English forEducational Purposes (TEEP)
Source: O’Sullivan (1999)
Overall Impression Score
The writing is completely satisfactory. 9
8
The writing is satisfactory and generally communicates fluently with only occasional 7lapses of organisation and structure. Clear well argued position taken.
The writing is mainly satisfactory and communicates with some degree of fluency. 6Although there is occasional strain for the reader, control of organisational patterns and devices is evident. Clear argument, though the writer’s point of view is not obvious.
The writing sometimes causes strain for the reader. While the reader is aware of an 5overall lack of fluency, there is a sense of an answer which has an underlying coherence. Somewhat poor control of the language and little evidence of the writer’s point of view. May contain occasional direct ‘lifting’ of the text from the input or inappropriate use of quotations or references.
4
The seriousness of the problems in writing prevents meaning from coming through more 3than spasmodically. Evidence of systematic plagiarism or excessive use of quotationsor referencing.
2
• a virtual non-writer; contains no assessable pieces of English writing 1• wholly, or almost wholly copied from the source materials
• less than approximately 50 words
Candidate did not attend or attempt the question in any way. 0
Assessing performance
13
can be assessed – while the second point refers to the notion of who is qualifiedto make decisions (based on the scale) in an LSP performance test.
Douglas (2001) argues that the criteria included in a rating scale shouldemerge from the same needs analysis that is used to define the language usedomain, and that these criteria should then be augmented and supported by ourcurrently-used theoretically-based approaches (see Weir 1983). He goes on tosuggest a ‘weaker’ indigenous scale hypothesis:
. . . in which the indigenous criteria may be used first to supplement linguistically-oriented criteria in line with the construct definition, and,secondly, to help guide our interpretations of language performances inspecific purpose tests (Douglas, 2001:183).
What Douglas seems to be saying is that we should attempt to discover thelinguistic criteria relevant to making judgements of performance in a particularTLU domain and try to ‘square’ these with what we know of existing languageability theory. The problem again lies in the area of boundary definition. Howcan we decide where to draw the line between creating a scale that is very muchfocused on the task in question and creating a scale that can be used to generalisebeyond a specific event? It appears that we cannot easily do this. A scale canallow us to draw one type of inference from our test but not both.
Another problem lies in the fact that in performance tests the rating scale is alink between task performance and test score, so it must be theoretically sound(in that it is tied to our construct definition and allows for meaningful inferencesto be drawn from test performance) as well as practically usable. Though there issome evidence to suggest that raters can use rating scales in a similar wayirrespective of their background (Lumley 1998, Lumley and McNamara 1995),this is really only an issue where the decision being reached is specific to aparticular area and where the test is representative of the ‘strong’ view ofperformance testing.
It is important to remember at all times that the purpose of an LSP test is tohelp us draw inferences on the ability of a candidate to use the language of aspecified domain in the context of that domain and in a manner that is
Figure 1.4 Practicality problem with a single scale across all levels
1
2
3
4
5
6
C2
C1
B2
B1
A2
A1
unused section
‘useful’ section
unused section
‘useful’ section
1 Introduction to the testing of language for business purposes
14
appropriate to that domain. Its purpose is not to allow us to draw inferencesrelated to a candidate’s ability to perform other than linguistically in the domainitself.
Towards a theoretical conceptualisation ofbusiness language testsThe main thrust of this chapter so far is that it is not helpful to take the view thattests can only be seen as being ‘specific purpose’ (SP) if they are very narrowlyfocused on a particular ‘purpose’ area and are representative of, to borrowMcNamara’s (1996) expression, a ‘strong’ view of specific purpose testing.Instead there are a number of perspectives related to ‘specific purpose’ tests thatoffer a not incompatible expansion to the definition of SP tests offered byDouglas (2000:19).
1. As all tests are in some way ‘specific’, it is best to think of all language testsas being placed somewhere on a continuum of specificity, from the broadgeneral purpose test such as the Certificate of Proficiency in English (CPE)to the highly specific test (Figure 1.5), such as the test for air trafficcontrollers described by Teasdale (1994).
Figure 1.5 A view of test specificity
2. Very highly specific tests tend to be very poor in terms of generalisability,while the opposite can be said of non-specific tests. There is not a binarychoice in operation here, and if we accept that tests can be developed along aspecificity continuum, then it logically follows that a test which appears tobe placed somewhere other than the extremes of the continuum will have thepotential to be either more or less generalisable.We could conceive of a test task that is specific only in that it is placedwithin the context of an employment/career area (in our case ‘business’),and that will be generalisable to the broader ‘general language use’ contextbecause it is essentially testing non-specific language, or it is not activatingthe same cognitive processes as a task that is more highly specific does.
3. Where a test is situated closer and closer to the more highly specific end ofthe continuum, the focus on situational authenticity also changes. That is, ahighly specific test will most closely reflect the ‘real world’ situation orcontext, while a more general, less specific test will be less likely to do so(though it is not impossible that a specific context might be exploited in a
UnspecifiedPurpose
Highly SpecifiedPurpose
Describing tests of business language
15
test of general proficiency). In other words, a highly specific test will clearlydemonstrate situational authenticity.
4. Since we are essentially focused on tests of language, the aim of any specificpurpose language test is to attempt to say something about a candidate’slanguage ability within the specific context of interest. Therefore, the extentto which a test task engages a candidate’s underlying processing andlanguage resources to the same degree as called for within the specificcontext domain indicates the degree of interactional authenticity of that testtask.
5. The degree to which non-language factors impact on a candidate’s testperformance will reflect the degree of specificity of that test. Therefore, in ahighly specific language test it may not be possible to separate the languagefrom the specific event. Where such a test is called for (i.e. a ‘strong’ form ofspecific purpose tests) this should be recognised in the definition of theconstruct and as such the only possible way to assess language performanceshould be within performance in the event, using, for example, the type of‘indigenous’ assessment rubrics or scales suggested by McNamara andJacoby (1999) and developed by Abdul-Raof (2002).
It is clear from these five points that the notion of ‘degree of specificity’ iscentral to any definition of a specific purpose language test – since the impact ofother factors will vary, depending on the positioning of a test along a specificitycontinuum. In the sections that follow, I will review a series of tests of languagefor business purposes, taking these points into account – though of course noreview would be appropriate without some reference being made to otheraspects of a test’s quality.
Describing tests of business languageIn this section, I will review a series of business language tests from thetheoretical perspective suggested above. From this review, I hope to findevidence to support such a perspective, leading to a more comprehensive under-standing of the issues involved in the testing of language for business purposesin particular and for specific purpose language testing in general.
Of course, tests should not be evaluated solely on the basis of the theoreticalconcepts described above. Those qualities that can be seen to offer morecomprehensive evidence of the test’s usefulness should also be taken intoaccount. Accordingly, the following reviews will be structured using thefollowing framework:
1. A brief introduction to the test.2. A brief description of the test.3. An outline of the construct upon which the test focuses.
1 Introduction to the testing of language for business purposes
16
4. The test method.5. Skills’ coverage.6. Measurement qualities.7. Degree of specificity/Authenticity.8. Impact of non-language factors.9. Reporting of test performance.
Test of English for International Communication(TOEIC)
1. A brief introduction to the testThe testing of language for the purpose of establishing benchmarks for partici-pants in international business or commerce in the modern era appears to havestarted with the development of the Test of English for International Communi-cation (TOEIC). The test, developed by the Educational Testing Service (ETS)in response to suggestions by the Japanese government (prompted by its largeindustrial corporations), was first administered in 1979. According to an earlytest user’s guide (ETS 1986:1), the test was designed to test two aspects oflearners’ language:
• ability to understand a business-related conversation in standard English• reading English language work manuals, correspondence, technical books
and articles.
The TOEIC was designed as a standardised test of reading and listeningcomprehension, set in the context of international trade and commerce. Itconsisted of a series of 100 multiple-choice items for each of the two skillstested. While it was originally designed for the Asian (particularly the Japanese)market, its use has now spread to other parts of the world.
2. A brief description of the testThe TOEIC is a 200-item test in which two aspects of a test taker’s language aretested, listening and reading comprehension, as mentioned above, there are 100items for each of the two aspects tested. All items in the TOEIC use a multiple-choice question (MCQ) format.
The Listening sectionThis section consists of 100 items and takes approximately 45 minutes tocomplete. Input consists of four parts
1. Statements related to a series of photographs (20 items, 4-option MCQ).2. Questions, responses required (30 items, 3-option MCQ).
TOEIC
17
3. Short conversations (30 items, 4-option MCQ).4. Short talks (20 items, 4-option MCQ).
The Listening section offers a series of activities ranging from very basiclevel identification of elements related to a set of photographs, through to under-standing the content of short conversations and talks. At no time do the testtakers listen to extended discourse, nor do they need to actually do anything withthe information received (except select either an acceptable reply to a questionor a summary of what was heard).
The Reading section This again consists of 100 items, though here 75 minutes are allowed. There arethree parts:
1. Sentence completion (40 items, 4-option MCQ).2. Error recognition (20 items, 4-option MCQ).3. Comprehension of short texts (40 items, 4-option MCQ).
The ‘comprehension’ section has been criticised (Douglas 2000:235) both forthe fact that it is non-reciprocal in nature and for the disparate sub-skills thatappear to be tested by the different items – which seem to draw on skills such asscanning for detail and making pragmatic as opposed to propositional infer-ences from a text – in other words, drawing on background knowledge. Anothercriticism is the decision to use only largely decontextualised short texts, whichat best represent fragments of texts, rather than use a variety of text types andlengths. Similar criticisms can be made of the other section; for example, the‘sentence completion’ section appears to test grammar and vocabulary, whilethe ‘error recognition’ tests sentence level grammatical and lexical awareness –so, while we may be able to say that a test taker can identify errors in a text, wecannot say that that person would be able to identify non-highlighted errors in alonger script, nor can we say that that person would be able to correct anyidentified errors unless a selection of options is offered. This problem with thelength of the texts is also clearly important, with these two sections only dealingwith single sentence input.
As mentioned above, this format has not changed since TOEIC was firstintroduced in 1979.
3. An outline of the construct upon which the test focusesFrom the standpoint of the theoretical framework of LSP suggested here, theTOEIC is problematic from a number of perspectives. The description of the testhighlights a problem with the way in which the test is specified. It seems thatDouglas’s (2000:236) criticism that ‘it is unlikely that the reading tasks engagethe test takers in genuinely communicative behaviour or in genuinely specificpurpose language use’ suggests that the test should not be considered to be a
1 Introduction to the testing of language for business purposes
18
‘genuine’ LSP test at all, and indicates that it should be placed towards the‘general’ or ‘unspecified’ end of the specificity continuum discussed above.
There is also a problem with the inferences that can be drawn from theTOEIC. According to the TOEIC Users Guide, the test:
. . . measures the everyday English skills of people working in an interna-tional environment. TOEIC test scores indicate how well people cancommunicate in English with others in the global workplace. The test doesnot require specialized knowledge or vocabulary; it only measures the kindof English used in everyday work activities (Chauncey Group 1999:4).
Taking these three assertions separately we can see that there are clear problems.The test purports to measure everyday skills in an international work
environment, yet focuses only on listening and reading – certainly skills usefulin such an environment but hardly sufficient to allow us to say anything aboutthe second assertion, i.e. the ability of people to actually communicate. TheGuide later asserts that speaking and writing are not assessed because theyrequire ‘considerable time and expense, both for administering the test and forscoring’ (Chauncey Group 1999:8), and are comparatively less reliable than thetests of the receptive skills examined. The assertions concerning the relationshipbetween performance on the TOEIC and on separate indicators of speaking andwriting ability appear to have been based, worryingly, on measures of generalproficiency in these skills, adding to the confusion as to the ‘specific’ orientationof the test. This confusion is highlighted again in the final sentence, whichsuggests that the test writers do not see the language of ‘everyday work activ-ities’ to be in any way ‘specialised’ or different from a general languageproficiency.
There are other difficulties with the descriptions used by the test developersof the underlying construct, as reflected in the claims (i.e. inferences that can bedrawn from test scores). Perhaps the most obvious of these are reflected in state-ments quoted below from two major TOEIC websites, that for Europe and forthe USA. The European site states that the TOEIC measures test takers’:
. . . English comprehension, speaking, writing and reading skills in an inter-national environment. The scores indicate how well people can commu-nicate in English with others in business, commerce and industry.
(source: http://www.toeic-europe.com/pages/eng/the_test_pres.htm accessed January2004)
On the other hand, the main (USA-based) site for the test claims that:
. . . The TOEIC test measures the everyday English skills of people workingin an international environment.
(source: http://www.toeic.com/2_2tests.htm accessed January 2004)
There is clearly some confusion as to the underlying construct of the TOEIC.
TOEIC
19
This confusion is manifested in the claims made of what inferences can bedrawn from performance on the test (at present there is very limited empiricalsupport for claims regarding language production) and in the very nature of thetest – is it a test of general proficiency or a test of language for business-relatedcommunication, or both?
4. The test methodThe TOEIC has been criticised by Douglas as representing:
. . . a good example of a well-constructed norm-referenced traditionalmultiple choice test task, with no doubt high reliability, but extremelylimited in the inferences it will allow about language knowledge(2000:236).
This criticism is not particularly surprising given that the TOEIC is a test bornof the psychometric–structuralist era (Spolsky 1995), where tests were ratio-nalised by theoretical insights from ‘associationist learning theory, structurallinguistics, contrastive analysis and psychometrics’ and a belief that the ‘phono-logical, morphological, syntactic and lexical components of language areisolable as are the four skills of listening, speaking, reading and writing’(Hawkey 1982:124). It is unlikely that any test based on these premises mightprovide evidence of the kind of communicative behaviour referred to byDouglas (2000). Douglas does have an important point to make. The TOEICwas introduced in 1979, at a time when the theoretical rationalisation uponwhich it was based had been superseded by what Spolsky (1995) called thepsycholinguistic–sociolinguistic era. Possibly the harshest criticism that cantherefore be made of the TOEIC is of the failure of its creators to respond tochanges in theoretical perspectives of language competence and related changesin language teaching that had already begun to reshape the language testingscene by the mid- to late-1970s, see Hawkey (2004) for a useful historicaloverview of the period.
The danger of relying on high stakes test instruments based on multiple-choice questions (MCQs) has been highlighted in a number of recent reviews oftest evaluation procedures in the United States (see in particular the review ofthe Texas Assessment of Academic Skills or TAAS by McNeil and Valenzuela2000). These reviews have highlighted the presence of significant bias in theperformance on such tests by minority candidates. When this criticism iscoupled with the added problem of test validity (for example independentresearch indicated that as scores in particular school districts increased on theTAAS test of reading other indicators of the candidates’ actual ability to readshowed a significant decrease), the danger is even greater. This is not to say thatsuch item types are of no real value; when used in addition to other, more directmeasures they can add to our perspectives on the ability of a test candidate (infact the reality of modern tests means that many batteries, such as the Cambridge
1 Introduction to the testing of language for business purposes
20
ESOL Main Suite and the proposed new Test of English as a Foreign Language(TOEFL), already employ a variety of item and task types).
5. Skills’coverageThe TOEIC tests reading and listening only, a situation which means that the testis seen by this writer as something of an anomaly. This is because its veryexistence can only be justified by adapting a theoretical view of language whichis in direct contradiction to the test method used. The TOEIC claims to representa measure of an individual candidate’s ability to communicate in a businessenvironment, yet it uses a methodology which pre-dates the communicative erain which language knowledge is tested as opposed to any ability to actually usethat language. The only empirical evidence that the inferences drawn fromperformance on the test can be related to ‘communication’ comes from Wilson(1989), though there is a serious question mark over the measures he used tocompare TOEIC performance with.
6. Measurement qualitiesThe relationship between the listening, reading and total TOEIC scores areshown below, Table 1.1. The fact that these correlations are really quite highmay point to a muddying of the measure and the ability being measured.
Table 1.1 Correlations between TOEIC sub-tests
Note: * p ≤ .001
Since the listening and reading scores are included in the total score, it is not atall surprising that there are very high correlations reported between thesesections and the overall. What is surprising is the fact that there is such a highcorrelation between the two sub-tests – in correlation analysis of the reading andlistening sub-sections of the Test of English for Educational Purposes (TEEP)(O’Sullivan 1999, Weir 1983) typical correlation coefficients are in the regionof 0.5 to 0.6. Very high correlations suggest that the two tests are very stronglyrelated, for example, in one of the few studies to focus on the TOEIC, thereported correlation coefficient between a direct speaking measure and theTOEIC Total score was 0.74. This was seen by the TOEIC developers asevidence that the test can accurately predict candidates’ speaking ability(Chauncey Group 1998:1–2), yet the correlation of 0.82 reported here is not seenby the developers as a problem.
The internal consistency estimates reported in the TOEIC Technical Manual(Educational Testing Services 1998:2) for what they refer to as ‘the Japanese
Listening Reading Total
Listening 1.000 0.822* 0.952*
Reading 1.000 0.957*
Total 1.000
TOEIC
21
secure administration’ (Woodford 1982:66) are shown in Table 1.2. Thesefigures are not surprising, considering the number of items and the presumablybroad range of candidates tested. It should be noted that reliability estimatessuch as the Kuder Richardson formulae and Cronbach’s Alpha are notoriouslysusceptible to test-taking population variability (so a test can have a reliability of.93 with one population and .63 with another). However, given what we knowabout the TOEIC population, these numbers appear to be quite acceptable.
Table 1.2 KR-20 Reliability Coefficients for the TOEIC test
7. Degree of specificity/authenticityFrom the overview offered here, it would appear that the test developers havenot attempted to deal with the specificity issue and in terms of the frameworksuggested by Douglas (2000), it would be very difficult to justify calling this atrue specific purposes test. With regard to the concept of situational authenticity,which is reflected in the content of the test in terms of text and task demands,there does not seem to be any evidence that the test reflects the specific languageuse domain.
Weir (1993, 2004) suggests how the demands of the content domain might bedescribed (see Table 1.3). Here, the limitations of the TOEIC are clearlyhighlighted. The sample questions from the listening paper that appear in theExaminer’s Handbook (ETS 2002:14 and 17 for example) could be from anytest of general proficiency. While this is not necessarily a bad thing in itself, thefact that the vast majority of the items (in fact all of the items included in theExaminee Handbook) would be equally comfortable in a general proficiencylistening test suggests that there are serious shortcomings across all elements ofthe text demands’ framework – for example, the focus on single word recog-nition or on listening for detail does not reflect the range of demands of thebusiness context.
Similar limitations can be pointed out for the Reading paper – for example, inthe Examinee Handbook, all reading items are based either on sentence-lengthor short paragraph-length texts and while there are items based on short notices,there is no text longer than approximately seventy words and neither is thereanything that resembles any of the more common reading texts from thebusiness context (brochures, e-mails, business letters). Similarly, the taskdemands on both papers are uniform: a set of equally weighted multiple-choiceitems, with no consistent purpose attached to task fulfilment (other thanachieving a satisfactory grade in the test), a response format that does not reflectthat of the target domain and an extremely limited number of operationsinvolved.
Listening ComprehensionReading ComprehensionTotal Test
0.920.930.96
1 Introduction to the testing of language for business purposes
22
Looking at the issue of interactional authenticity, we can only presume thatresponding to multiple-choice items alone can never engage the candidate in thekind of cognitive processing evident in listening or reading in a businessdomain.
In fact this criticism of MCQs is not new. When presenting their model of testtask response, Pollitt and Ahmed, suggest that they:
. . . had found it extremely difficult to model the process of answeringmultiple choice questions, and are inclined to think that, perhaps for thisreason alone, they are of questionable validity for educational assessment(Pollitt and Ahmed 1999:1).
Pollitt and Ahmed were essentially attempting to model the cognitivebehaviour of candidates under test conditions, a concept further developed byWeir (2004) in the ‘theory-based validity’ element of his frameworks. Thelinking of an understanding of the executive processes and resources availableto the test taker is central to the notion of interactional authenticity.
All this suggests that the TOEIC might best be placed close to the ‘non-
Skill Area Task Demands Text Demands
Listening PurposeResponse formatWeightingKnown criteriaOrder of itemsTime constraintsIntended operations
LinguisticMode/channelTypeLengthNature of informationTopic familiarityLexical rangeStructural rangeFunctional range
InterlocutorSpeech rateVariety of accentAcquaintanceshipNumber of speakersGender
Reading PurposeResponse formatWeightingKnown criteriaOrder of itemsTime constraintsIntended operations
LinguisticChannelText typeText lengthNature of informationTopic familiarityLexical rangeStructural rangeFunctional range
Writer-reader relationship
Table 1.3 Task and text demands for Listening and Reading
Source: based on Weir (1993, 2004)
Other tests of language for business purposes
23
specified purpose’ end, calling into question any claim that it might be testinglanguage for a specific ‘business’ purpose.
8. Impact of non-language factorsThere is no evidence that the non-language factors have an unexpected impact onperformance on the TOEIC. Since the previous section places the test squarely inthe category of ‘general proficiency’, it is clear that there are no (or certainly veryfew) elements within the TOEIC that might be affected (negatively or positively)by the business language use domain – for example there are no items in theExaminee Handbook where a background in business would give a candidate anadvantage over a fellow candidate without such a background. The fact that thereare no business-related texts in the reading part, for example, means that acandidate who has never seen or read a business letter (or has had to respond tosuch a letter) would be in no way affected by his or her total lack of experience inthe business world. While it might be argued that this lack of negative bias is agood thing, it seems counter intuitive that a person without such a backgroundwould be seen as capable of communicating ‘in English with others in business,commerce, and industry’ (Chauncey Group 2002:1).
9. Reporting of test performanceThe norm-referencing methodology used in the TOEIC, means that acandidate’s test performance is reported in terms of where the candidate mightbe placed relative to the population who sat for a particular administration of thetest. In a situation where a decision is to be made on a candidate’s ability toperform (in linguistic terms) in a given context, this is problematic. It might be,in an extreme example, that none of the candidates are actually capable ofperforming at the level required by an employer. Results of this sort will not tellthe employer that this is the case however, only that candidate x is better orworse than candidate y. We can then see that the way in which a test of languagefor a specific purpose, such as business, is reported is actually a vital character-istic of that test (a similar argument is made by Douglas 2000). If a test isdesigned to offer an estimation of the ability of a candidate to cope with thelinguistic challenges required of a specific business or work environment, thensome criterion level must be set below which a candidate should not fall. Thiscriterion should only be set in relation to the specific language use domain andnot in relation to the ability of other candidates.
Other tests of language for business purposesThe growing interest in specific purpose testing during the 1990s has resulted inan increased number of tests for business, both for English and other languages.This section looks at a range of such tests, starting out with a representative
1 Introduction to the testing of language for business purposes
24
sample of tests from the UK, Pitman Qualifications and the London Chamber ofCommerce and Industry Examinations Board (LCCIEB). There then followreviews of tests of other languages (French, Italian and Japanese).
Pitman qualificationsThe Pitman tests, now administered by the City and Guilds of London, at presentoffer a pair of tests specifically aimed at business English. These are the Englishfor Business Communication (EBC) and English for Office Skills (EOS).
English for Business Communication
1. A brief introduction to the testThree levels are available: Elementary, Intermediate and Advanced. Accordingto the test developers, these correspond with the Common European Frameworklevels A2–Waystage, B2–Vantage and C2 (see Figure 2.4 for a diagrammaticoutline of the levels). Unlike the other tests referred to in this chapter, these testsare available both to native speakers and to overseas candidates, provided theyhave reached a particular level of language ability as measured by other non-business oriented Pitman tests (intermediate standard in the ESOL examinationor elementary in the English examinations for the Elementary and Intermediatetests respectively). According to the Pitman website, a ‘background knowledgeof office practice and organisation is required’.
The tests are integrative in nature, with each of the three levels involving thecandidate in writing a range of answers in response to input, often handwritten.Before taking a brief look at the suite, it should be pointed out that it is not at allclear that the developers have considered the language level of candidates whoare non-native speakers of English. This is most clearly exemplified by the brief‘Contextualisation’ offered at the beginning of each test level. As can be seenfrom Figure 1.6, the language of these three is almost indistinguishable. Thisapparent lack of concern with the language of the input undermines the suite, asit is quite conceivable that candidates, particularly at the lower levels, mayexperience significant difficulties with understanding the input. This will clearlyhave a negative impact on their test performance.
2. A brief description of the testThe three levels of the test are outlined in the following table (Table 1.4). In thistable we can see that the three levels are quite similar in content, but with agreater number of tasks to be completed (in an ever increasing amount of time).It should also be noted that there is no clear substantive difference (apart fromthe increased number of tasks, which is offset by the increased time allowed) in
Pitman qualifications
25
the output required. It would be interesting to establish, through a latent traitstudy for instance, what the differences in difficulty of the three levels really are.Unfortunately, there is no publicly available documentation on how differencein level is established or maintained – this criticism can be made of the other testdevelopers referred to in this chapter.
It is not clear from the documentation how the benchmarking to the CommonEuropean Framework (CEF) was achieved – whether it was done through a
Figure 1.6 Contextualisation offered at levels 1–3 of the Pitman EBC suite
Level 1 source: Pitman Qualifications English for Business Communication, Past Paper EL–NBC (11:2)Level 2 source: Pitman Qualifications English for Business Communication, Past PaperEL–NBC (12:2)Level 3 source: Pitman Qualifications English for Business Communication, Past Paper EL–NBC (13:2)
Level 1Elementary
SITUATION
As Personal Assistant (PA) to Mr Arthur Jordan,Managing Director of Fine Finishes, a small but flourishing decorating firm, you are frequently left incharge of the office while he is away on business.
Today he has left you the following tasks.
Level 2Intermediate
Level 3Advanced
SITUATION
Light Waves Ahead is a small commercial radio company with a station in Blantyre and another inHarare. Its main source of income is from advertising onthe air by local firms. You are Personal Assistant (PA)to Mr Moses Banda, the General Manager.
SITUATION
You are Personal Assistant (PA) to Mr Joshua Banda,Managing Director (MD) of EAST AFRICA HOTELSLtd, with hotels and holiday lodges in Kenya andZimbabwe. Your Head Office is located atIndependence Way, HARARE, Zimbabwe.
1 Introduction to the testing of language for business purposes
26
qualitative comparison between the test specifications and the CEF, or whetherevidence was gathered from test candidates (as was done in the ALTE ‘Can Do’project).
3. An outline of the construct upon which the test focusesTaking the test descriptions seen in Table 1.4 as a basis, it is clear that the testlevels described are focused primarily on the writing ability of candidates. Infact, the tasks typically involve the candidate reacting to a written prompt, sothere is a genuine attempt to mirror the language use domain of the workenvironment, see Figure 1.7 for an example of how this is conceived in a taskfrom Level 2.
We can therefore say that the construct that seems to underlie the English forBusiness Communication is that of an integrated reading into writing approach.
Level 1 Level 2 Level 3
Benchmarked CEF Waystage (B2) CEF Vantage (C1) CEF Mastery (C2)
Time Allowed 90 minutes* 120 minutes* 150 minutes*
Contextualisation Brief (35 wordsapprox)
Brief (35 wordsapprox)
Brief (35 words approx)
Task 1 Writing – guided letter Writing – guided letter Writing – guided letter
Task 2 Writing – memo Writing – fax guidedby written instruction
Writing – memo guidedby written instruction
Task 3 Writing – fax guidedby written instruction
Writing – memo Writing – fax guided by written instruction
Task 4 Writing – guided letterfrom written input
Writing – guided shortreport fromcharts/tables andwritten input
Writing – notice/memoguided by writteninstruction
Task 5 Writing – guidedarticle from writteninput
Writing – press releaseguided by writteninstruction
Task 6 Writing – report fromcharts/tables and writteninput
No word limit set fortasks, all tasks 25marks
No word limit set fortasks, all tasks 20marks
No word limit set fortasks, all tasks 20 marks except tasks 3 and4 (10 each)
Table 1.4 English for Business Communications (Pitman)
* all tests have an additional 15 minutes of reading time during which candidates are allowed toread through the test paper, but not to write.
Pitman qualifications
27
Figure 1.7 Example of integrated task (reading into writing) from Level 2Pitman EBC
Source: Pitman Qualifications English Business Communication, Past Paper EL–NBC (12:6)
4. The test methodAs described in the previous section, the test is comprised of a series of writtentasks. In the task reproduced above, candidates are told that their work isassessed for quality of layout. Layout is also assessed for one ‘memorandum’and one ‘letter’ task at each of the three levels. The other criteria are ‘language’(though exactly what this means is not defined), ‘content’, ‘neatness’ and‘legibility’.
5. Skills’coverageAn interesting feature of the tests is the lack of a ‘test-like’ rubric. Instead, allinstructions are included in the input for the tasks. However, it appears that thePitman series essentially tests only writing – though there is of course a writteninput to be read in the case of each task, there is no overt examination of thereading skill it is, presumably, tested indirectly through performance on the writing. While this may well represent an accurate picture of the ‘real’ world,the fact that the skills are integrated in this way means that there is a danger ofcross-contamination, in that it will not be clear if a test taker performs poorly dueto a lack of reading ability or a lack of writing ability.
This problem is inferred in the [Pitman] Examinations Report when it isstated of Level 2, that
some candidates lose marks for content because they are so busy inventinginformation to fit their format that they ignore the real purpose of the report(2000:11).
It could be argued that they may not actually fully understand what the focusis because they have misinterpreted the input. The report goes on to describe thefact that ‘[Many] candidates fail to read the prompts carefully enough beforestarting to write’ as an ‘area of weakness’ (2000: 12). It should be pointed outthat this is a criticism that could be made of any test using an integrated format.The problem is usually addressed by carefully monitoring the language of theinput to ensure that it is written at a level that is below that of the test (so thelanguage of input for a C1 level test is usually aimed at level B2). The worry hereis that it is not clear if this monitoring has been adequately done.
6. Measurement qualitiesNo information is currently available in the public domain.
7. Degree of specificity/authenticityAt first glance, it appears that the test is more ‘specific’ in nature than theTOEIC. It may therefore be useful to see why this might be the case, so that wecan develop a clearer picture of how specificity is manifested in this type of test.So what is more specific about the test?
1 Introduction to the testing of language for business purposes
28
In order to look at the degree of specificity it is necessary to consider the testfrom the perspective of situational and interactional authenticity. In the case ofthe latter there is no actual evidence to support any claims in this respect.However, the evidence from the test descriptions and in particular from the itemtype described earlier (Figure 1.7) suggests that there is a serious attempt here torecreate a realistic domain-specific task. The above task involves the candidatein the integration of skills (reading and writing) in order to produce a businessletter. Clearly, there is a major difference here when we compare the readingtasks in the previously reviewed test, where the candidate was expected torespond at all times to an MCQ item.
In terms of situational authenticity, there is evidence that the developers havetried to recreate a ‘business’ context through the tasks (often integratinghandwritten memos and notes to written output) which all appear to have a veryclear business focus.
While all of this is positive, there is some concern over the fact that the testsare only concerned with reading into writing tasks. This very much lessens thetrue business focus of the tests in that a major element of the business languagedomain is simply ignored. This has the effect of lessening the strength of anyspecificity argument we might wish to make for these tests.
8. Impact of non-language factorsDespite the shortcomings associated with testing only a limited aspect of acandidate’s language, there is a relatively high degree of specificity in thedifferent levels of this test. With this degree of specificity, comes a potential fornon-language factors (such as background variables) to have some impact ontest performance. However, before simply accepting that this impact is neces-sarily negative, let’s consider the argument made by Elder (2001) with regard towhat she perceived as the negative impact of non-language factors in tests ofspecific purpose.
Imagine, for example, a test candidate who has had a lot of experience inwriting the sort of letter called for in Figure 1.7. It appears only natural that thisexperience should positively affect that person’s performance on the task.
Now imagine a second scenario where another candidate, this time with littlebusiness experience, but with a similar level of language ability as the firstcandidate and a lot of experience in taking MCQ-based tests, is asked tocomplete an MCQ version of our task. This person too will perform well andagain the impact comes primarily from experience of the task type.
When we consider these two situations we see that there are in fact a numberof ways of looking at non-language factors. The latter form is clearlyproblematic as it constitutes a source of context-irrelevant variance. However,the former is quite different, in that it is certainly not ‘context-irrelevant’, in factthere is a clear argument here for the inclusion of this source of variance in theconstruct definition of tests in a specific purpose domain.
Pitman qualifications
29
If we try to dilute our specific purpose tests until there is little or no danger ofcontext-related non-language impact we end up with tests that are basically non-context dependent general proficiency instruments.
9. Reporting of test performanceTest performances are reported using the criterion levels described in Table 1.5.
Table 1.5 Criterion levels for the EBC (Pitman) levels
The reported pass rates for the three levels in 2002 are seen in Table 1.6.
Table 1.6 Pass rates for the EBC (Pitman) levels – 2002
Source: Pitman (2002).
English for Office Skills
1. A brief introduction to the testThe English for Office Skills’ series offers a pair of tests (Levels 1 and 2), thestated aim of which is
. . . [To] demonstrate accuracy in the use and transcription of English, andthe ability to perform office-related tasks to spoken or written instructions(Pitman 2003:40).
Unlike the other tests reviewed in this chapter, the EOS tests are aimed at bothnative and non-native speakers of English (Pitman 2003:40) and are claimed byits developers to be aimed at ‘[People] who need to carry out tasks in Englishwhere accuracy in writing and following instructions is important’ (Pitman2003:40).
2. A brief description of the testAs with the above Pitman test, the levels of the EOS test are described in Table1.7.
1 Introduction to the testing of language for business purposes
30
Criterion Level 1 Level 2 Level 3
Fail < 60 marks < 60 marks < 60 marks
Pass 60 – 74 marks 60 – 74 marks 60 – 74 marks
First Class Pass > 74 marks > 74 marks > 74 marks
Criterion Level 1 Level 2 Level 3
Fail 21% 32% 41%
Pass 48% 48% 47%
First Class Pass 31% 20% 12%
Section A at each level focuses on spelling and listening comprehension,while Section B focuses on reading comprehension, vocabulary and accuracy.The descriptive table indicates that writing ability is not tested directly, insteadit is estimated through a candidate’s ability to identify errors in the proof-reading tasks and in the sentence completion task.
Table 1.7 English for Office Skills (Pitman)
3. An outline of the construct upon which the test focusesThese tests do not appear to have been designed to offer a measure of a non-native English-speaking test taker’s proficiency within a business context, butrepresent a more vocational measure of practical skills. The underlyingconstruct is somewhat unclear from the test description. The test seems to focuson the form of the language – identifying different aspects of linguistic accuracyas the underlying construct.
4. The test methodThe test method is essentially confined to short answer format (SAF) itemsbased either on examiner-read input (tape recordings are, as yet, not used in thePitman tests), or on written input. The fact that the listening comprehensionportion is read by the examiner is quite problematic, as there is a clear possibility
Pitman qualifications
31
Level 1 Elementary CEF A2– Waystage [Marks]
Level 2 Intermediate CEF B2– Vantage [Marks]
Section A Spelling Sentence read by examinerTarget word repeated 20 items– accuracy of spelling [20]
Sentence read by examinerTarget word repeated 20 items – accuracy of spelling [20]
Listening Comp. Short passage read twiceForm completion (written) ormessage transfer (oral) [10]
Short passage read twiceForm completion (written) ormessage transfer (oral) [10]
Section B Reading Comp. Read newspaper/magazinearticle – sentence completion[10]
Read newspaper/magazinearticle – sentence completion[10]
Syntax 15 items – proof-reading [15] 20 items – proof-reading
Vocabulary 10 items – select appropriateword (from two) [10]
10 items – select appropriateword (from two) [10]
Punctuation Proof-reading [10] Proof-reading [10]
Proof-reading A Identify error in table offigures [10]
Identify 5 errors in table offigures [5]
Proof-reading B Proof-read letter – identify 15errors (typography, spellingand/or punctuation) [15]
Proof-read letter – identify 10errors (typography, spelling orpunctuation, style) [10]
of an ‘examiner effect’ where different people will be more or less clear in theirreading aloud (thus introducing an element of construct-irrelevant variance).Apart from this criticism, the test method appears to meet the needs of the test (totest language knowledge rather than language use).
5. Skills’coverageAs was outlined above, the skills covered in the test are limited to the ability todemonstrate knowledge of the language through measures of linguisticaccuracy, though as we can see from Figure 1.8 the proof-reading task is morerelated to identifying differences in the numbers in the tables than it is to identi-fying language-related differences. The sources of input, therefore, are quiteimportant in this type of test. The listening element of the test consists of twoparts, lexical knowledge (listening and spelling) and comprehension (through ashort dictation). Here there is no real evidence that the second part measures‘comprehension’ (by which I mean understanding) though I acknowledge thatthis is not an area in which all testers will agree (see for example the argumentsof Lado (1961) and Oller (1979) who disagree on what dictation actually tests).
The reading comprehension items are built around a single passage (approx-imately 350 words at Level 1 and 500 words at Level 2). Items are based onsentence completion, which limits the skill being tested to that of searching forspecific information (and possible lexical synonyms). The remainder of bothpapers contains items related to the display of knowledge of language accuracy.
Both levels contain a proof-reading item as Task 7. In these cases the task isto compare two tables of information, one being accurate and the other said tocontain errors (10 at Level 1 and 5 at Level 2). As can be seen from the extract inFigure 1.8, this is not actually a reading task at all – it is a proofing task and is notdependent on language ability.
6. Measurement qualitiesNo information is currently available in the public domain.
7. Degree of specificity/authenticityThe degree of specificity appears to be quite low. While the tasks have been setin a business context, the tasks that are included are not necessarily related to thebusiness domain. In other words, they are more context-oriented (i.e. they are setin the context of the business domain) than context-focused (i.e. they aredesigned to test only the language of the business domain). This suggests thatthe test is more general proficiency focused and should be placed towards thenon-specific end of the continuum.
Taking just one task as an example (Figure 1.9, the reading task which beginsSection B of each test paper) it is interesting to notice how it measures up withregard to the first items on the list of task demands suggested by Weir (1993),
1 Introduction to the testing of language for business purposes
32
Pitman qualifications
33
Figure 1.8 Proof-reading task (#7) from EOS Level 1
Source: Pitman Qualifications English for Office Skills, Past Paper EL–OFFN (11:9)
purpose and response format. While we would expect to see a clear purpose forreading in any test, within a specific purpose domain this need becomes centralto the characterisation of the task. Here, there does not appear to be a clearlyspecified purpose. As for response format, we would expect in a specific purposedomain task that the response format will replicate that of some element of thedomain, again the sentence completion format is not at all relevant to thebusiness domain. This latter criticism is possibly a bit harsh, as it would appearto be very difficult to satisfy the need for business domain-like response formatsfor all skills – in particular receptive skills.
In terms of task demands, the test can be criticised from the perspective ofchannel (tape recordings are not used in these tests so the invigilator reads thelistening passages aloud), text type (while the reading text shown is based on amagazine-type article it is not typical of business-related reading material,which is more likely to be a letter, e-mail, memo or report), and text length (inthis case the text is quite long at about 450 words; however, this is not typical ofbusiness texts, which tend to be brief and to the point).
When it becomes clear that the test can be criticised for its context validity
1 Introduction to the testing of language for business purposes
34
Figure 1.9 Reading items from Pitman qualifications EOS Level 2
(Extract from approx. 450 word passage)
(items related to that section of the passage)
Source: Pitman Qualification: English for Office Skills Level 2; Past Paper EL–OFFN (12:4–5)
(where context validity is seen in terms of task and text demands) the claim ofsituational authenticity is seen as tenuous.
Unlike the other Pitman test reviewed, there is little evidence here of interac-tional authenticity. It could, of course, be argued that the test is focusing on avery discrete level of knowledge and that generalisation to the businesslanguage use domain is possible. The distance between this micro view oflanguage ability and the macro level of language use is great and such anargument is somewhat difficult to sustain.
8. Impact of non-language factorsFrom the review to date, it is clear that the inseparability of skills issue is againproblematic. This is particularly relevant with regard to Tasks 2 (listeningcomprehension) and 7 (proof-reading), where there is a real danger of areader/speaker-related effect in the former and in the latter where the proof-reading skills are focused on identifying numerical rather than lexical orsyntactic differences.
9. Reporting of test performanceTest performance is reported in the same way (and with the same cut scoreboundaries) as the English for Business Communication (EBC) test reviewedabove. This means that the Passing level is set at 60% while the First Class Passlevel is set at 75%. The pass rates for the 2002 administration are shown in Table1.8, though no data on the test population are available.
Table 1.8 Pass rates for the EOS (Pitman) levels – 2002
London Chamber of Commerce and IndustryExaminations Board (LCCIEB) tests of languagefor business and commerce
1. A brief introduction to the testAs the name of the organisation suggests, the London Chamber of Commerceand Industry Examinations Board (LCCIEB), has, as its main focus, theprovision of specialist examinations in the area of business and commerce.Among the many examinations it offers are a number which are dedicated to the
LCCIEB tests of language for business and commerce
35
Criterion Level 1 Level 2
Fail 24% 41%
Pass 34% 33%
First Class Pass 42% 26%
testing of language for business purposes. Since the approach adopted byLCCIEB is rather unique, it will be dealt with somewhat differently to the otherexamination providers.
The uniqueness of the LCCIEB examinations stems from the adherence to asingle framework, which seems to be applied regardless of the language beingtested. This effectively means that a single test specification has been used tocreate what we might call multi-language clones. To illustrate what I mean bythis, I will briefly review their ‘. . . for Business’ range of examinations.
2. A brief description of the testThe following table (Table 1.9) is a breakdown of the description of the tests ofEnglish and Spanish for Business, as described by the LCCIEB in the extendedsyllabuses for these examinations.
Table 1.9 LCCIEB tests of language for business and commerce
MCQ – Multiple-Choice questionsT/F – True/false questionsSAF – Short answer format questionsITr – Information transfer questionsExW – Extended written output required (S = Short; L = Long)Spo – Spoken output requiredCoE – Council of Europe FrameworkEOP – Effective Operational Proficiency* – Not yet available
3. An outline of the construct upon which the test focusesBoth tests focus on reading and writing skills, and are specified in exactly thesame way. While this is not in itself problematic, the fact that the task and itemtypes are essentially identical means that the test developers see no differencebetween different languages at particular levels of proficiency. While it may
1 Introduction to the testing of language for business purposes
36
Test Skills Tested Method CoEBenchmark
L S R W MCQ T/F SAF ITr ExW Spo
English for BusinessPreliminary LevelLevel 1Level 2Level 3Level 4
•••••
•••••
•
•
• •
••
••••
SSS/LS/LL
BreakthroughWaystageThresholdVantageEOP
Spanish for BusinessPreliminary LevelLevel 1Level 2Level 3Level 4
•••••
•••••
•
•
• •
••
••••
SSS/LS/LL
BreakthroughWaystageThresholdVantageEOP
seem to be relatively easy to agree that a candidate at a particular level should becapable of performing a particular task or function, the degree of linguisticsophistication needed to achieve this is not exactly the same for all languages.Including the same task regardless of target language is at least potentiallyproblematic.
Alderson (1998) pointed out the danger of adopting such an approach in thecontext of the DIALANG project in Europe. In this project a test format throughwhich a whole series of official European languages could be tested at equiv-alent levels was envisaged. The original plans involved devising a set of detailedspecifications for the English test and then cloning tests in the other languagesfrom this. The developers found that there were real difficulties in identifyingappropriate tasks (as tasks which were seen to be at an acceptable level for onelanguage were found to be more suited to a different level for another language),and in identifying what was considered acceptable performance across differentlanguages for those tasks that were considered appropriate. The solutionadopted by the DIALANG group was to allow developers from each languagebackground to interpret the specifications to create an instrument that theyconsidered appropriate. This approach was also taken by the BULATS teamswho developed tests in the same four languages as the LCCIEB (these arediscussed below).
Not only are the series of tests for each language all based on the same model,in fact, there appears to be no difference in the make-up of the tests in thedifferent languages. For example, the description in the Extended Syllabusdocuments for a particular task at Level 3 for the German (LCCIEB 2001b:3),Spanish (LCCIEB 2001c:3) and French (LCCIEB 2001d:3) tests is shown inTable 1.10.
Table 1.10 Extract from extended Syllabus for LCCIEB tests of German, Spanish and French
When this is compared to the English for Business documentation (LCCIEB2001a:3), we find that it is again exactly the same, see Table 1.11. The suspicion
LCCIEB tests of language for business and commerce
37
German Level 3 – Task 2 Spanish Level 3 – Task 2 French Level 3 – Task 2
Question 2 involves thedrafting of an internal reportbased on raw data given in theform of graphs, notes, presscuttings, charts, tables, etc.Candidates will have tounderstand, select, collateand, if necessary, supplementthis data in order to write thereport in the light of theinstructions given.
Question 2 involves thedrafting of an internal reportbased on raw data given in theform of graphs, notes, presscuttings, charts, tables, etc.Candidates will have tounderstand, select, collateand, if necessary, supplementthis data in order to write thereport in the light of theinstructions given.
Question 2 involves thedrafting of an internal reportbased on raw data given in theform of graphs, notes, presscuttings, charts, tables, etc.Candidates will have tounderstand, select, collateand, if necessary, supplementthis data in order to write thereport in the light of theinstructions given.
that all of the tests are essentially clones of the original English test, which wasthe first to be introduced, is confirmed when we see the actual tasks.
While the LCCIEB provide details of the syllabuses for each of their exami-nations, together with specimen papers, sample answers and examiner’sreport/comments, there is no evidence supplied in support of the approach theyadopt in creating these tests in the different languages. The same criticism can bemade of the test of Spoken English for Industry and Commerce (SEFIC) and theForeign Languages for Industry and Commerce (FLIC), with only a singledifference between the two sets of examinations (the addition of a translationtask at the highest – fourth – level).
4. The test methodFrom the descriptive table (Table 1.9) we can see that the test includes a range oftask and item types, with multiple-choice (MCQ), short answer format (SAF),true/false (T/F) and written production all included in the response options.
5. Skills’coverageReading and writing are tested at both levels.
6. Measurement qualitiesNo information is currently available in the public domain.
7. Degree of specificity/authenticityThese tests appear to be closer to general purpose language tests than to specificpurpose instruments, as the following tasks suggest. The reason for this canagain be traced to the task and text demands implied in the sample materialsprovided by the developers (see Figure 1.10).
It is not easy to know what these listening items are testing. The need for‘complete and grammatically correct answers’ suggests that the items may be
1 Introduction to the testing of language for business purposes
38
English Level 3 – Task 2
Question 2 involves thedrafting of an internal reportbased on raw data given in theform of graphs, notes, presscuttings, charts, tables, etc.Candidates will have tounderstand, select, collateand, if necessary, supplementthis data in order to write thereport in the light of theinstructions given.
Table 1.11 Extract from extended syllabus for LCCIEB test of English
focused on grammar (in the Bachman sense – syntax, lexis etc.), though thenature of the task with which they are associated appears to be communicative(information transfer). Other task demands which do not appear to fit with suchtasks in the business domain are:
response format – we might expect that the listener would create a messageor memo from this type of input, rather than simply respond to a series ofdiscrete items, though it appears that other listening tasks do lead on to awriting taskknown criteria – the above example suggests that candidates might haveproblems responding to the item as the criteria for achieving marks are notrelated to the apparent communicative nature of the task. For the one piece ofwriting included in the test the marks are awarded for ‘correct titles’ (20%),‘the message’ (40%) and again ‘spelling and presentation’ (40%).In terms of the text demands, it is clear that there are limitations of functional
range, nature of information (the items could quite easily be presented as part ofa general proficiency test), and text length (only very brief reading and listeningextracts used). In addition, the lack of contextualisation means that there is noeffort made to establish any meaningful interlocutor-to-listener relationship, soany speaker-related variables remain untapped – again a situation unlikely in thebusiness domain.
The fact that these tests are more focused on the general proficiency domainmeans that there is some likelihood that the test tasks will not result in the kind ofcognitive processing that typifies a business domain task performance. Thissuggests that it is unlikely that interactional authenticity can be successfullyclaimed for these tasks. However, it is not at all clear yet if it is possible to effec-tively identify typical patterns of processing associated with successful taskperformance in a specific domain.
8. Impact of non-language factorsAs we have seen in the other tests reviewed here, the indications are that where atest is situated closer to the general proficiency end of the specificity continuum,
LCCIEB tests of language for business and commerce
39
Figure 1.10 Listening Item – English for Business, Preliminary level(LCCIEB)
Source: English for Business, Preliminary, sample paper: 6
there is little danger of non-language factors that may be associated withknowledge of the test context impacting on test performance.
9. Reporting of test performanceTest performances are reported using the criterion levels described inTable 1.12.
Table 1.12 Criterion levels for the English for Business (LCCIEB) levels
No details are available on pass rates.
Tests for business purposes in languages otherthan EnglishApart from the tests of English for business, and the foreign language testsadministered by the LCCIEB, there are a growing number of tests in languagesother than English for the purpose of describing candidates’ ability to use thatlanguage in a business or commercial context. Some of these tests are describednow.
Test de français international (TFI)
1. A brief introduction to the testThe TFI is designed to evaluate the level of French of non-native speakers. Likeits sister test, the TOEIC, the test is based on a series of MCQ items focusing onreading and listening. Somewhat confusingly, the developers make quitedifferent claims of what the test aims to measure. On the link to the TFI from themain TOEIC website, it is claimed that it can be used to assess ‘a candidate’sability to understand, speak, read and write French as it is used in the interna-tional workplace and in everyday life’ (ETS 2003a). While there is some verylimited evidence that this may be the case for the TOEIC, there is no evidencewhatsoever to support a similar claim for the TFI. On the TOEIC–Europewebsite the claim is replaced with a less bold statement that the ‘test assesses acandidate’s ability to communicate in French as it is used in the internationalworkplace and in everyday life’ (ETS 2003b).
Interestingly, the claims made of the TFI and the TOEIC are not alwaysconsistent. On the USA-based website the statements made of the TOEIC reflect
1 Introduction to the testing of language for business purposes
40
Preliminary Level 1 Level 2 Level 3 Level 4
Pass 50% 50% 50% 50% 50%
Credit 60% 60% 60% 60% 60%
Distinction 75% 75% 75% 75% 75%
those more conservative statements made of the TFI on the European site, whilethe broader claims of generalisability to all four skill areas made on theEuropean site regarding the TFI are reflected on the USA website, but withregard to the TOEIC.
2. A brief description of the testAs mentioned above, the TFI appears to be a clone of the TOEIC – with the samesub-skills tested using the same item types. Table 1.13 offers an overview of thetest. At the time of writing, no information was available in the public domain onthe make-up of the test, either in the form of a specification or of publishedsupport materials.
Table 1.13 Descriptive table of the Test de français international
3. An outline of the construct upon which the test focusesAs mentioned before, the TFI focuses on the receptive skills of reading andlistening. The claim that it offers a measure of all four skills is neither supportedby the multiple-choice-based approach nor by the decision to test only thereceptive skills.
4. The test methodThe test takes a multiple-choice approach and is solidly based in the samepsychometric–structuralist approach as the TOEIC, which, as I mentionedearlier, has long been abandoned as the primary methodology in gatheringevidence of a test taker’s ability to perform specific language tasks. This is not tosay that the approach is incapable of ever providing evidence. On the contrary,when it comes to obtaining estimates of a clearly defined and realised (orrealisable) trait the theoretical foundations of the approach are as sound today asthey were when they were developed. Among the problems with using theapproach as it is manifested in the two tests here (TFI and TOEIC) is that thepurported construct, as evidenced by the inferences that developers claim can bedrawn from test performance, does not appear to be supportable.
5. Skills’coverageThe TFI consists of two sections, one devoted to listening and the other toreading. As is consistent with the approach taken (see above), the sub-sectionspresent the language in short segments (of reading and listening texts), each
Tests in languages other than English
41
Section I LISTENING (42 min) Section II READING (68 min)
I Question-Answer [40 questions]II Short Dialogues [30 questions]III Short Conversations [20 questions]
IV Error Identification [25 questions]V Incomplete Sentences [25 questions]VI Comprehension [40 questions]
designed to test a specific aspect of the language ability of the test taker.However, we can only make assumptions about the test format and content as, atthe time of writing, there is no evidence (apart from the outline provided on theTFI website) available in the public domain.
6. Measurement qualitiesNo information is currently available in the public domain.
7. Degree of specificity/authenticitySince there is very little information about the test available in the publicdomain, it is not possible to make any definitive comment on the degree ofspecificity, though if the TFI really is a clone of the TOEIC, it would appear thatit is more a measure of general language proficiency than of proficiency in aspecific context, and like the TOEIC, it is unlikely to display evidence of eithersituational or interactional authenticity.
8. Impact of non-language factorsAgain, we do not have the evidence to establish if non-language factors have anyimpact on test performance.
9. Reporting of test performanceThe scores are reported in the same way as the scores for TOEIC, so individualscores are reported for reading (on a scale of 5–495) and listening (on a similarscale) and a total score (on a scale of 10–990). There does not appear to be anattempt to indicate what these scores might mean (for example in terms of abenchmark of ability such as the Common European Framework).
Certificazione della conoscenza dell’italiano commerciale(CIC)
1. A brief introduction to the testThe Certificazione della conoscenza dell’italiano commerciale (Certificate inItalian for Commerce – CIC) was developed at the Università per Stranieri diPerugia, Italy during the late 1990s and first administered in June 2000. The testwas developed in response to a perceived demand based on the increasinginterest at that time in the domain of Italian language for business. The CIC isintended to establish a candidate’s ability to use Italian in ‘work-relatedcontexts’: travel agencies, banks, estate agencies, and industry. To date, the testpopulation has reached the level of approximately six hundred candidates peryear – non-native speakers of Italian, working in, or hoping to work in an Italianbusiness context.
1 Introduction to the testing of language for business purposes
42
At the time of writing, the CIC is the only certificated test of Italian forbusiness.
2. A brief description of the testThe CIC consists of five sub-tests: reading, listening, grammar and lexicon,writing and speaking and is offered at two levels, these are intermedio (Interme-diate) which is set at ALTE Level 2 (or CEF Level B1) and avanzado (advanced)set at ALTE Level 4 (CEF Level C1). See Table 1.14 for an outline of the twotests.
As can be seen from this table, the tests offer an extensive assessment of thelanguage level of the candidates. According to the CIC handbook, the tests aredesigned to certify ‘that the holder’s knowledge of the Italian language isadequate for that person to interact and work in business contexts’ [theiremphasis] (CIC 2003a:2) and suggests they can be used by:
• people who work or intend to work in international environments and whowant to enhance their personal curriculum
• companies and organisations selecting personnel or those who wish to checkthe qualifications of their employees
• schools/universities with economic and business courses who want tosurvey or determine the level of knowledge of the Italian language for theirown students.
3. An outline of the construct upon which the test focusesThe test appears to have been built around the model of language abilitysuggested by Bachman (1990) and is similar in design to the Cambridge ESOLmodel. The test therefore takes the same multi-skills approach as similarCambridge ESOL tests.
4. The test methodThe five components of the CIC are weighted as shown in Table 1.15 and aretested using a variety of item types: MCQ, matching, gap-filling, letters, compo-sitions, and short essays. The method is based, to a large extent, on the use ofactual business documentation. These texts are to be found in the reading andlistening components, as well as in the writing and speaking papers. Contextual-isation of test tasks is evident, particularly in the test papers focused on languageproduction – where there is a very clear description of audience, as well asreference to the required level of formality of the output.
5. Skills’coverageAs mentioned above, the CIC includes measures of five aspects of a test taker’slanguage ability – reading, listening, grammar and vocabulary, writing andspeaking.
Tests in languages other than English
43
1 Introduction to the testing of language for business purposes
44
Intermedio (Level B1) Avanzado (Level C1)
A. Reading A1. Careful global reading – 10 MCQitems based on 20 to 120 wordpassages (businessdocumentation)
A1. Careful global reading – 8 mainideas items (matching – based onbusiness documentation)
A2. Expeditious global reading(skimming) relate item to passage– 10 items
A2. Careful global reading – passagecompletion (cloze type, givenchoice from 13 phrases) – basedon business documentation
A3. Expeditious and careful globalreading – finding main ideas inpassages 8 items (businessdocumentation – matching andgap-filling)
A3. Careful global – identifyingmain ideas in passage 5 items(MCQ) – based on businessdocumentation
B. Listening B1. Careful global – listening forgeneral understanding, 10 items,short monologues – matching)
B1. Careful global – informationtransformation (note-taking),11 items (SAF)
B.2 Careful local – form/memocompletion 8 items (phonemessage – note-taking)
B.2 Careful local – identify speakersand topics (10 items – matching)
B.3 Careful global – extended text,4 items based on business-relatedmonologue or conversation (MCQ)
B.3 Careful global – extended text,4 items (MCQ)
C. GrammarandVocabulary
C.1 15 item MCQ format C.1 12 item MCQ cloze format
C.2 Cloze passage, 10 items (basedon business communication)
C.2 Cloze passage, 12 items
D. Writing D.1 Writing a formal business letteror informal business relatedemail (90–110 words)
D.1 Writing a report based on inputfrom graphs/charts – about100 words
D.2 Write an argumentative text to aspecific person related to aspecific business topic (200 to250 words)
E. Speaking E.1 Personal Information Exchange –no preparation
E.1 Personal information exchange– no preparation
E.2 Interaction – with examinerbased on read input (materialsgiven 10 minutes before test)
E.2 Interaction – with examinerbased on read input (materialsgiven 15 minutes before test)
E.3 Long turn – on known workrelated topic (materials given10 minutes before test)
E.3 Long turn – monologue ongeneral work related topic(materials given 15 minutesbefore test)
Total Time 115 minutes 225 minutes
Table 1.14 Certificate in Italian for Commerce: description
Source: CIC (2003b)
6. Measurement qualitiesNo information is available in the public domain at the time of writing.However, the test developers are involved in a large scale pan-European projectconcerning the development and validation of an item bank. This will be used tomore accurately define test levels for the CIC (as well as the other tests theycurrently administer).
7. Degree of specificity/authenticityFrom the description of the test presented above we can deduce that there hasbeen an effort on the part of the test developers to include in the CIC tasks thatare based on business documentation and that reflect the use of language in thebusiness domain. The degree to which they succeed appears to be mixed,however. As we have seen in the reviews of the other tests, there seems to be areal problem particularly with the receptive tasks. The example shown here(Figure 1.11) is interesting in that the context is clearly that of the businessdomain and the required output reflects that of the domain, but the degree ofscaffolding (in the form of the guides to what to listen for) acts to reduce its situa-tional authenticity. It is difficult to see how this situation can be resolved; afterall if we just give the candidate a blank page and tell them to listen to the messagewe are completely changing the task. In the real world the listeners will bring tothe event a great deal of background knowledge related to the particularcompany they are working for, so a schema for dealing with the call will be inplace. The function of the scaffold is to reduce the impact of this lack of schema.
The fact that the tasks are typically based on business documentation and areexplicitly benchmarked to the work-related aspect of the CEF and ALTE frame-works (CIC 2003a:5–8) can be seen as evidence of situational authenticity.However, the inclusion of tasks that are clearly not related to the domain (partic-ularly the MCQ responses), and the limitations of tests of receptive skills(implied in the above critique of the listening task) weaken the veracity of thisevidence.
From the perspective of interactional authenticity the evidence is also mixed.The variety of task types included at both levels suggests that the interactionbetween the executive resources available to the candidate and the executiveprocesses (i.e. cognitive and meta-cognitive processing) may well be facilitatedat least for some of the tasks (particularly in the tests of production). However,this same variety means that there are tasks that are very unlikely to have thesame effect. Here I am referring to those based on the receptive skills, languageknowledge display, and in particular those that rely on multiple-choice items.
As for the degree of specificity of the CIC tests, we can see that the test fitsinto the category of a business-oriented test, with some evidence that at leastsome of the papers are also business-focused. The writing test is an example ofthis; there the expected output is in the form of a contextualised business-relatedtext with clearly defined writer/reader relationship and degree of formality.
Tests in languages other than English
45
Even here, there is some question over the tests as the written performancesare awarded scores based on language-related criteria – lexical competence,competence in morphology and syntax, sociocultural competence and consis-tency (CIC 2003a:14). The absence of any task- (and therefore business-domain) focused criterion reduces the likelihood that these are highly specifictests.
8. Impact of non-language factorsThe fact that the test consists of a battery of papers, each focusing on a particularskills’ area suggests that any non-language impact will be mixed.
9. Reporting of test performanceThe criterion level for achieving a passing grade is set at 60% – averaged fromthe results on all five components through a weighting system described in Table1.15. This system tells us quite a bit about the interpretation of the construct,with a very clear emphasis on spoken language at both levels (where thiscomponent is worth 30% of the total marks available), and the perception thatwriting becomes more central to business language needs at the higherproficiency level. It goes from being worth just 10% of the total score (the least
1 Introduction to the testing of language for business purposes
46
Figure 1.11 Listening Task B2: CIC Intermediate
important sub-skill) at the intermediate level, to 20% at the advanced level, thesecond most important sub-skill.
Table 1.15 Certificate in Italian for Commerce: weighting system
Candidates are awarded a grade based on a simple addition of the scoresachieved on each part of the examination. There are three passing grades andtwo failing grades – see Table 1.16.
Table 1.16 Certificate in Italian for Commerce: reporting system
There is no information available on the rates of grade achievement for the tests.
Other tests of European languages for business purposes
There are a number of other tests of European language for business (see Table1.17). While I do not have space here to address these tests individually, it isuseful to spend just a little time on them. All of these tests are administered bymembers of the Association of Language Testers in Europe (ALTE), and allfollow similar models – a focus on testing the four skills through a multi-tasktype approach. For more information on these tests see the section at the end ofthe book where contact information is given for all currently administered testsreferred to in this chapter.
JETRO Reading and Listening Comprehension Test (JRLT)
1. A brief introduction to the testThe Hawaii based Japan-America Institute of Management Science, JAIMS,has been involved in the education (language, business and culture) of Japanese
Tests in languages other than English
47
Intermedio Avanzado
A. ReadingB. ListeningC. Grammar and vocabularyD. WritingE. SpeakingTotal Score
40 (20%)40 (20%)40 (20%)20 (10%)60 (30%)200
35 (17.5%)35 (17.5%)35 (17.5%)35 (17.5%)60 (30%)200
Grade A Grade B Grade C
Excellent GoodSatisfactory
Pass
Grade D Grade E
UnsatisfactoryVery poor
Fail
and North American graduates for almost thirty years. The organisation admin-isters a test of Japanese language for business purposes, developed by JETRO(Japan External Trade Organization), with the support of over six hundredcompanies in Japan. The test was developed during the early 1990s and firstadministered in 1995. It originally consisted of papers at three levels, thoughfrom 2003 there has been a revised format, which consists of a single paper.
2. A brief description of the testAll items in the Reading and Listening Comprehension Test (JRLT) use a four-option MCQ format. The different types of questions are outlined in Table 1.18.From this description, we can see that the test is based on an assessment of thereceptive skills of the candidate – the associated oral test is described in thesection that follows.
Table 1.18 Item types from the JRLT
3. An outline of the construct upon which the test focusesAccording to JETRO’s website:
‘The JETRO Test is designed to objectively measure and evaluate one’sproficiency in using the Japanese language for communication involving a
1 Introduction to the testing of language for business purposes
48
Levels(ALTE/CEF)
French German Spanish
Level 3/B2 ZDf B (ZertifikatDeutsch für den Berut)
CEN (Certificado deEspanol de los Negocios)
Level 4/C1 PWD (PrüfungWirtschaftsdeutschInternational)
Level 5/C2 DSEC (DiplômeSupérieur d’EtudesCommerciales)
DEN (Diploma de Espanolde los Negocios)
Table 1.17 Other tests of European languages for business purposes
Focus No. Items Time
Listening test Matching written and audio descriptionsMatching written expression to contextCareful global listening
101015
50 min.
Listening andReading test
Matching audio description to written textCareful local listening and matching to short written texts
1515
30 min.
Reading test Grammar and vocabularyCareful local reading (expressions)Careful global listening
101015
40 min.
Total 100 120
variety of situations and circumstances, targeting non-native speakersengaged primarily in business’ (JETRO:2003a).
The developers also claim that
‘The JRLT comprehensively evaluates the examinee’s skill in usingJapanese to deal with a variety of business-related tasks and problems’(JETRO:2003a).
However, since it is clear that the test is focused only on listening and reading,the construct is actually quite limited. As with other tests, this is not a problem initself, though making claims that go beyond the definition of the construct uponwhich the test is based is justifiably seen as problematic, as these claimsrepresent the inferences that the developers believe can be drawn from testperformance. The issue is therefore one of validity.
4. The test methodAs mentioned previously, the test uses MCQ format items throughout. Mosttasks involve matching audio or read input to a visual stimulus. This can be in theform of a photograph (see Figure 1.12 for an example of this task type from theListening paper), or of a piece of written text (see Figure 1.13 for an examplefrom the Reading and Listening paper).
Figure 1.12 JRLT Listening paper: sample item
Source: JETRO (2003b)
With the Reading paper, there are three item types. The first of these is actuallytesting grammar (See Figure 1.14), while the second tests ‘forms of speech’using the same MCQ format.
In the final section of the JRLT, the candidate is asked to respond to an item inwhich they are asked to identify the main point or idea in a text (Figure 1.15).
Tests in languages other than English
49
5. Skills’coverageThe test includes papers devoted to listening and reading, though there is anintegrated listening and reading element (see above).
6. Measurement qualitiesThough there are tables of candidature (size and success rate) for each year sincethe test was introduced, there are no figures available which tell us about thequalities of the test (overall/sub-test reliability, item statistics).
7. Degree of specificity/authenticityThe degree of specificity is not high here, with a clear focus on the languagerather than on the context (this can be seen from the items included above wherethe candidates focus on their knowledge of the language as displayed throughtheir responses to MCQ items). The fact is that the MCQ item format is useful interms of the testing of aspects of language (or other skills) that lend themselvesto being broken down into ‘discrete’ elements or chunks. However, the very actof decontextualising the language to this degree negates any claims of situa-tional authenticity.
The test appears to be well constructed, though there appears to be a questionmark over its situational authenticity. This is because it offers a series of taskswith little effort to create a systematic contextualisation through relating the
1 Introduction to the testing of language for business purposes
50
Figure 1.13 JRLT Reading and Listening paper: sample item
Figure 1.14 JRLT Reading paper: sample grammar item
Source: JETRO (2003b)
Source: JETRO (2003b)
tasks to the sort of demands outlined in the earlier reviews. The interactionalauthenticity is also questionable as responses to all of the tasks are evaluatedusing MCQ items, with no reference to the interlocutor/audience for exampleand little attempt to ensure that the linguistic demands of the texts reflect those oftexts in the business domain.
8. Impact of non-language factorsAs noted above, the somewhat confused description of the construct tested (asreflected in the levels’ ability statements contained on the test website) meansthat it is not possible to identify what the developers of this test are trying toachieve – though the extent to which the non-language factors actually impacton the test performance is not altogether clear.
Tests in languages other than English
51
Figure 1.15 JLRT Reading paper: identifying overall meaning
Source: JETRO (2003b)
9. Reporting of test performanceAt the time of writing, the actual reporting mechanism is not available in thepublic domain, though according to the website there appears to have been amove from a criterion-referenced (i.e. pass/fail) system to a norm-referencedsystem based on that of the TOEIC and TOEFL. This decision appears to havebeen made without regard to the basic criticism of this type of system (made hereand by Douglas 2000) that the resultant numbers relate to how well the candidateperformed compared to other candidates – it does not tell us if the person cansurvive linguistically in a business environment.
It is required that a score of 530 be reached in order to qualify to sit the JETROOral Communication Test, though students who have achieved a passing gradein the final administration of the pre-revision JRLT may also apply.
The JETRO Oral Communication Test (JOCT)
1. A brief introduction to the testThe JETRO Oral Communication Test (JOCT) is an ‘add-on’ to the JRLT,which can only be taken by candidates who have achieved a score of 530 on theJRLT. Its developers claim that it ‘comprehensively measures and evaluatesone’s proficiency in using Japanese to communicate’ (JETRO, 2003c).
2. A brief description of the testThe JOCT is described on the JETRO website as consisting of two parts (seeTable 1.19), one involving the test taker and the examiners (there are alwaystwo, one specialising in Japanese language and another with a businessbackground) in an interactive dialogue and the other a role-play. Performancesare audio and/or video recorded for later evaluation. The holistic assessmentscale used in the JOCT is included here as Appendix 1.1.
Table 1.19 Task types from the JOCT
Source: JETRO (2003c)
3. An outline of the construct upon which the test focusesThe construct appears to be jointly focused on a candidate’s ability to useJapanese in a more social situation (though it should be noted that the notion of a
1 Introduction to the testing of language for business purposes
52
JOCT Details Duration
Q & Q Conversation, led by tester, about the test taker’s job and topicalsubject related to business
15 minutes
Role Playing Role playing in imaginary business situations, includingmonologues (short speeches, etc.) to see how the test takers dealwith given tasks and situations
15 minutes
conversation between an examiner and a candidate is probably not sustainable,due to the inequalities inherent in the event) and the candidate’s ability to use thelanguage in a typical business setting (as operationalised through a role-playtask).
4. The test methodThere are two parts. In the first part the candidate interacts with a pair ofexaminers. Here, the focus appears to be on the candidate (personal informationexchange, work experience etc.), while the two available tasks which are meantto exemplify the role-play task in part 2 appear to show the candidate in twodifferent situations, suggesting that this part of the test can vary widelyfrom administration to administration. In one version (a video clip is availableon the website) the candidate is engaged in a telephone conversation with anexaminer, while in the other the candidate makes a formal speech. The problemhere is that the first task involves the candidate in an extended interactivediscourse with an examiner while the second involves an extended monologue.We know from experience (O’Sullivan, Weir and Saville 2002, for example)that these different discourse types result in different task output profiles (interms of the language functions elicited) and may well have an effect on testperformance – particularly when we consider the work of Berry (1996, 1997)who has shown that candidates with different personality profiles are affected bytask type.
5. Skills’coverageThis test is focused on speaking only, and apart from the input prompts (whichare spoken) there is no other skill involved.
6. Measurement qualitiesNo information is currently available in the public domain.
7. Degree of specificity/authenticityThis seems to be somewhat mixed, with the first (interview) task more focusedon general proficiency, while the role-play task is more specific – in that thetasks are very much situated in the context of business.
The example videos of the role-play tasks show a very formal (and nottypically business) organisation – for example in the task where the candidateinteracts with one of the examiners by telephone, the interaction actually takesplace over the phone, but both are sitting at the same desk and are facing eachother. This affects the situational authenticity – though a simple manipulation ofthe setting, to create a physical distance or barrier between the speakers, could toa large extent, negate this criticism. The degree of interactional authenticity isprobably higher, with the tasks (particularly in the latter part of the test) more
Tests in languages other than English
53
likely to result in the candidates’ cognitive processing approaching that of thebusiness language domain.
8. Impact of non-language factorsThere does not appear to be a significant impact here of non-language factors,though the ‘speech’ role-play, the example presented of which is little more thana formal self-introduction, may be more influenced by knowledge of theJapanese business domain. The formality of the language and rhetorical structureof this type of presentation and the non-verbal ‘attitude’ of the speaker (rigidlystanding to attention while speaking) are not for example what a Europeanstudent might expect. We might therefore find that background knowledge mayplay a large part in successful performance on this type of task. The question thenis whether this is a good or a bad thing. While Elder (2001) argued that this typeof non-language impact is negative, we have seen above that it is probablyunavoidable where a test is quite specific, and it could well be seen as a positiveaspect of this type of test; after all it is part of what makes a test specific.
The other point to make related to this example is that business domains fromdifferent cultures may be radically different. So, a learner who is quite proficientin the language but is relatively unfamiliar with the culture may not perform aswell as a learner with experience of the business culture but with a lower level oflanguage proficiency.
9. Reporting of test performancePerformance is reported in terms of the evaluation criteria (see Appendix 1.1).No pass/fail criterion is set, so candidates receive a grade only (A+, A, B+, B, C,D). No effort has been made (beyond the brief descriptions offered in the evalu-ation criteria document) to say what these levels might mean (for example interms of the CEF).
Summary
The tests reviewed to date differed greatly in the language skills they examined.It is interesting that few of the tests include all four skills, though we shall seebelow that this is one of the cornerstones of the tests developed by CambridgeESOL in the UK.
The above tests differed not only in the skills’ area, but also in the approach totest and item format, to how productive language was evaluated and to howoverall performance was reported. As in any testing situation, there is no bestway, though there were examples of decisions that were taken by developers (touse only MCQ; to change from criterion to norm-referencing) that have under-mined the validity of the tests. There is also evidence to show that simply sayingthat a test is ‘specific’ or not is probably not a good idea: the complexity of the
1 Introduction to the testing of language for business purposes
54
matter means that different parts of a test can be seen to be more or less situa-tionally and interactionally authentic. Clearly a more comprehensive (thoughpractical) system of dealing with this issue is needed.
Before discussing that, it is now time to look at how the UK’s most influentialdeveloper of both general and specific purpose language tests have come to testlanguage for business purposes.
The development of business English testing atCambridgeNo mention has yet been made of the tests for business with which CambridgeESOL has been associated over the past decade. I have deliberately refrainedfrom including these tests in a general description of current practice in order totake this opportunity to establish a clearer perspective on the current CambridgeESOL approach to this aspect of testing. In order to more fully appreciate theapproach, we really need to go back to the mid-1980s, before the organisationbecame involved in business language testing. The tests in the following sectionare of historical interest, but as they are no longer administered, I will notattempt to offer the same 9-point analysis as was done for the preceding tests, butwill instead describe them in terms of their contribution to the historical devel-opment of business language testing at Cambridge ESOL.
Certificate in EFL for Secretaries (CEFLS)
In the mid-1980s, the Royal Society of Arts (RSA) in the UK developed a testknown as the Certificate in English as a Foreign Language for Secretaries(CEFLS). Like the TOEIC, this test was created in response to the perceivedneed of local clients, and was piloted from 1986 to 1989. Unlike the TOEIC, theformat of the test was based on the use of materials supplied by real companies,and was designed as a criterion-referenced test.
The report of the pilot scheme for the CEFLS (RSA 1987:4) indicates that atotal of 86 test takers sat the English Oral, Reading and Writing and Listeningtests, while six, 11 and 69 test takers sat the French, German and Swedish Trans-lation tests respectively. The high pass rate was indicated by the fact thatcertificates were awarded to 80 test takers who gained a passing score on allthree of the English tests; no mention is made of those who passed the othertranslation tests.
The Oral test was based on three distinct tasks identified as being represen-tative of ‘the type of interactions a secretary would undertake in his/her normalwork’.
These were:
1. Initiating a telephone call.
Development of business English testing at Cambridge
55
2. Receiving a telephone call.3. Face-to-face interaction with an unknown participant.
The three parts of the test were linked by a common theme, designed to obviatethe necessity for test takers to adopt different personas for each part. Test takerswere initially given a role as an employee of a real company (the Parker PenCompany was used for the 1987 pilot). The results of the pilot appear to havebeen quite satisfactory, though the final task seems to have been problematic,due to the reluctance of test takers to maintain their role (and initiate utterancesfor example), and the subsequent abandonment by the assessors of the role-play.Unfortunately, no reference is made in the report to the assessment criteria used,though there appears to have been a focus on task fulfilment.
The Listening test consisted of a series of five thematically related tasks(again based around information provided by the Parker Pen Company). TheReport tells us that the input was ‘recorded at the normal rate of delivery with arange of native speakers and included non-standard speakers’ (RSA 1987:8).The tasks are outlined in Table 1.20.
Table 1.20 Listening task types from the Certificate in EFL for Secretaries
These tasks appear to have been well attempted by the pilot group, though thehigh pass level suggests that there may have been some problem with the level ofdifficulty of the tasks – the Report (RSA 1987:9) does refer to the relativeweakness of responses to Tasks 2 and 4, which required production skills,though no additional information as to why this might have been the case ispresented.
The Reading and Writing test (Table 1.21) consisted of a set of seven tasks,each designed to test a particular aspect of the test taker’s language ability.These tasks were accompanied by materials taken from sources including theFinancial Times, the Parker Pen Company’s own publicity material, ‘andinvented tasks made as authentic as possible’ (RSA 1987:10).
While the CEFLS can be criticised in hindsight for its relative naivety andlack of professional polish (the pilot test, which is included here as Appendix 1.2was quite crude in its presentation), there were a number of very interesting andinfluential aspects of the test that deserve mention. For example, the view ofauthenticity implied in the use of materials related to real or realistic companies(though adapted or even scripted to suit the test) reflects current thinking to a
1 Introduction to the testing of language for business purposes
56
Task Description
12345
Respond to customer telephone order (complete sales order form)Respond to oral input with summary (in the form of a telex)Understanding of longer [time not given] input (true/false items)Listen for specific details from two telephone messages (written summary)Three extended messages (answer-phone) MCQ items
great degree and can be seen to satisfy the situational authenticity required of anLSP test (Douglas 2000). This authenticity was maintained in the Speaking test,where a range of tasks were included, while in the Listening test a range ofspeakers of English were used. The test developers also made efforts to ensurethat there was a strong measure of interactional authenticity in the type of taskschosen, though it is not now possible to establish empirically that actualcandidate performances reflected this view.
The fact that this test was very highly specified is not at all surprising, in thatit was developed with a particular test taker in mind, and it was never considereda requirement of the test that the results might be generalised to a wider generalpurpose language context. Of course, the question of a potential impact of non-language ability arises here again and it may well have been the case that famil-iarity with the domain may have contributed to performance.
When, in 1988, the RSA Examinations Board was amalgamated into whatwas then the University of Cambridge Local Examinations Syndicate (UCLES),it was decided to broaden the candidate base and CEFLS was redesigned,initially only slightly, and renamed the Certificate in English for InternationalBusiness and Trade (CEIBT).
Certificate in English for International Business and Trade(CEIBT)
The CEIBT consisted of three papers, testing reading and writing, listening andoral interaction. The Reading and Writing paper consisted of an introductory‘Information Page’, in which the test taker was introduced to the company andtheir own position within the company (for the purposes of the test) was contex-tualised. Among the companies used in the test were Rolls Royce, JapanAirlines, McDonald’s and The Body Shop. There followed a series of six taskswhere the test taker was expected to respond to a series of authentic stimulae inthe form of letters, memos, faxes and reports – though due to the authentic natureof the materials the task formats tended to vary from administration to
Development of business English testing at Cambridge
57
Task Description Marks available
123a3b4567
Questionnaire completionProof-reading task (10 discrete items)Formal letter Less formal letter Interpreting graphs/graphics (discrete items)Formal letter Letter and report formal Telex
1 mark5 marks5 marks5 marks4 marks5 marks5 marks5 marks
Table 1.21 Reading and Writing task types from the Certificate in EFL for Secretaries
administration (one of the factors that led to a revision of the test in 1998). Testtakers were allowed a total of 150 minutes for completion of the six tasks (thisincluded 10 minutes’ reading time).
In the Listening paper, test takers undertook a series of tasks, within thecontext of the same company. On the basis of what they heard in a series of audiorecordings, featuring both native and non-native speakers of English, they wereexpected to undertake a number of tasks. This paper lasted for approximately 65minutes including 5 minutes’ reading time.
Finally, the Oral Interaction paper consisted of a role-play, where the testtaker took the role of a company employee and the examiner took the role of avisitor to the company. The paper lasted for 13–15 minutes, with a total of 15minutes’ preparation time.
While the CEIBT has been praised for its commitment to authenticity of input(see for example Douglas 2000:175), it was this very commitment that had veryreal practical consequences for the production of the examination. One conse-quence was the difficulty in implementing full pretesting of the test tasks due tothe involvement of real companies. Additional problems identified in a 1994review document included the large amount of writing required and, perhapsmore crucially, the problem of what to do with an item shown by pre-testing notto be operating as predicted – as the test was seen as an integrated unit, a non-performing item could not be replaced with a previously banked example. Thislatter difficulty also had serious implications for the application of item bankingto the test system.
It is interesting to note that what was considered the strongest point of theCEIBT, the authenticity of its input, was also its Achilles heel. Apart from theproblems with pretesting referred to above, there were other even moreimportant difficulties. Perhaps the most relevant of these was the extreme viewof authenticity illustrated by the insistence on the use of real unedited material.The review document identified the following conditions for the production ofthe examination:
• ‘importance of obtaining genuine materials [emphasis in original] fromthe company.
• reliance on the materials voluntarily supplied by the context companies,which leads to problems if the company does not oblige [two examplesare provided in the review]
• reluctance to edit material obtained from the context company in thebelief that this is tampering with ‘authenticity’ [emphasis in original]
• belief in the importance of ensuring that the tasks on a particular paperwould actually be carried out by someone working in a particulardepartment within that company
• unwillingness to consider the use of fictional companies as a setting forCEIBT or to change the names [although this was, in fact, adopted forthe revised CEIBT, introduced in 1998]’ (UCLES 1994:13–14).
1 Introduction to the testing of language for business purposes
58
The writer goes on to identify the principal area of concern with this situation: ‘itis the materials obtained from the company rather than the existing testspecifications which drive the test’ (UCLES 1994:14). An additional problem,that of task consistency, was also identified, with examples given of significantvariation in reading load and of differences in task format. This latter situation isexemplified by comparing the two CEIBT tests that were administered in 1992(Table 1.22) where the tasks are not at all similar, either in terms of input or ofexpected response. See Appendix 1.3 for examples of two CEIBT test papers.
Table 1.22 Comparison of two CEIBT examinations
(Both of these tests are included as Appendices 1.3 and 1.4.)
This table shows how difficult it is to make meaningful comparisons betweenthe different versions of the test. This problem was also reflected in the unpre-dictability of the difficulty level of the test from year to year, a situationhighlighted in the review document (UCLES 1994:16), by the differences in thepercentage of candidates achieving a passing grade. It is therefore clear that the commitment to the use of purely authentic materials was compromising thevalidity and reliability of the test.
Other difficulties with the existing CEIBT included a perceived lack of
Development of business English testing at Cambridge
59
June 1992 November 1992
Task Input Output Input Output
1 Invoice (paymentoverdue)Brief memo
Letter (complaint) Report (cover only)Brief note
Letter (informational)
2 MemoAdvertising proof(10 errors)
Corrected proofFax (instructional)
Fax (3 questions)Office files (x 3)
Fax (informational)
3 Graphic design Letter (request) Fax (approx. 90words)Memo (approx.70 words)Article (approx.400 words)
Letter (informational)
4 Article (approx.600 words)
Report (120 wordsmax.)
Memo Article (informational)
5 Letter(suspendingcontract)Table + Chart
Letter(argumentation)
Memo (handwrittenadditions)
Memo (apologies,informational)
6 Application formMemo (x 2)MessageLetter
Note (prioritising)
clarity of definition of the role and purpose of the test – as compared with, forexample, the TOEIC. This was seen to affect the marketability of the CEIBT,and to have contributed to its relatively low take up (less than 1500 per year), andthe lack of support materials (with no published textbook). As a result of thisextensive review, with the addition of feedback from test takers, administrators,and Cambridge ESOL personnel, it was decided to revise the CEIBT. Thisrevision was to take almost three years, with the new version first administeredin June 1998.
The main changes to the test were:
• each of the three papers was to become a free-standing certificated test• each test had a different company context based on [my emphasis] an
authentic source• the Reading and Writing test and the Listening test were shortened• the Oral test now included an additional ‘mini presentation’ (but overall
length did not change).
A comparison of the test outline (Table 1.23) with that of the pre-revisionversion shows that the changes to the test were actually quite major. The taskswere now less open, in terms of expected response, and while there was definitereduction in specificity, and to some extent in the situational authenticity of thetest (mostly in that ‘real’ companies were no longer used), this does not appearto have been reflected in any way in the potential of the tasks to demonstrateevidence of interactional authenticity. However, the lack of archived data makethis impossible to demonstrate empirically.
Table 1.23 Format of the revised CEIBT Reading and Writing test
Unfortunately, despite these revisions CEIBT continued to attract very smallnumbers of candidates.
This situation, when combined with the successful development and intro-duction of the BEC suite, particularly BEC3, which was aimed at a similar leveltest taker and was able to build on the success of the earlier BEC examinations,meant that the CEIBT was withdrawn.
Another test that was brought into the UCLES fold in the mid-1990s, and thathas had some influence on the Cambridge ESOL approach to the testing ofEnglish for business was the Oxford International Business English Certificate(OIBEC).
1 Introduction to the testing of language for business purposes
60
Task Main Skill Focus
12345
Business correspondenceLanguage systemsBusiness correspondenceReading for detail, global meaning and inferenceExtended business correspondence
Oxford International Business English Certificate (OIBEC)
The OIBEC examinations were developed by the University of OxfordDelegacy of Local Examinations (UODLE) during the late 1980s and first intro-duced in November 1990.
The OIBEC offered tests at two levels, First and Executive. The examina-tions were designed for people with ‘a practical knowledge of English’ whowere ‘learning to use it in a business environment’ (UODLE 1990) and were atthe pre-intermediate and higher-intermediate levels respectively (or at the levelsof 4 and 6 on the English Speaking Union Framework Chart). Both levels werebased on case studies, and included papers testing all four skills.
An interesting feature of the OIBEC examinations was the inclusion of anextensive preparation package, which was given to each candidate three daysbefore the day of the examination. This package appears to have been devised toeliminate any individual ‘background knowledge’ effect on test performance,by giving the candidates three days in which to read through and study thebackground to the topic for the test they were about to sit.
Also of interest is the fact that the Speaking test used a paired format, theearliest inclusion of this format in a large scale test, though the format seems tohave been best exploited only at the Executive level, where the candidates wereinvolved interactively in two tasks (one of which appears to have been seen onlyas an extension of an earlier task and was not awarded individual marks). SeeTable 1.24 for an outline of the test.
CommentaryIn terms of the criteria referred to in the early stages of the chapter, the threeexaminations reviewed above can all be said to have been quite clearly specifiedwithin a business language domain – and as such are quite ‘specific’ in that theylie towards that end of the continuum. The changing attitude to situationalauthenticity can be clearly seen, in that the earlier CEFLS test was devised insuch a way as to mirror as closely as possible the target language use domain, afactor which contributed to a high degree of situational authenticity, and ‘face’validity, but which meant that the test could not be replicated. The focus onsituational authenticity reached its zenith with the CEIBT, a test that was quitepopular with certain stakeholders (teachers for example) but less so with others(candidates, test users and test developers).
The contribution of these testsApart from the obvious experiential aspects of administering tests of languagefor business purposes to an international population, these tests appear to havecontributed to the current Cambridge ESOL approach to business languagetesting in a number of ways. The development of the CEFLS through to the
Development of business English testing at Cambridge
61
CEIBT in its different versions appears to have demonstrated how the organi-sation moved from an approach where attention was drawn to the genuineness ofthe tasks used (in other words the focus was on the situational authenticity of thetask) to a perception of the test task which takes into account both its situationaland interactional authenticity. The other major change was to understand thatdifferent test versions (i.e. different versions of a single test) must be replica-
1 Introduction to the testing of language for business purposes
62
First Level Executive Level
Preparationpackage
Contents 4 pages of written input – rangefrom report to letter to table andgraphic.
6/7 pages of written input – consists of adetailed contextualisation, with excerptsfrom reports, letters, balance sheets,memos etc.
Reading andWriting
Time allowed
Marks awarded
Task 1
Task 2
Task 3
Task 4
Task 5
Task 6
75 minutes
100
10 SAF – Reading comprehension,based on preparation materials
Reading – Inferencing (3 items tobe identified, SAF)
Writing – register (3 items SAF)
Writing/Reading integrated – tablecompletion/summary
Guided writing – Memo, no wordlimit
Letter writing (scaffolded usingadditional input) – no word limit
95 minutes
100
Writing – Report completion (based onPrep. Materials) – 2 pages allowed,20 marks
Writing – Guided report (based on Prep.Materials) – 2 pages allowed, 15 marks
Reading – 5 SAF items, based onadditional fax input (15 marks)
Proof-reading – 10 items in short memotext (10 marks)
Writing – briefing paper completion(2 paragraphs) 20 marks
Writing – job application letter(150–200 words) 20 marks
Listening
Marks awarded
Time allowed
Task* 1
Task* 2
Task* 3
50
20 minutes
3 items, based on graph/table (SAF)
Complete form, based on input(3 pieces of information required –all SAF)
4 items, 1 table completion,2 additional information,1 inferencing item
* All telephone messages
50
20 minutes
10 comprehension items (MCQ)15 marks
9 comprehension items (SAF) 20 marks
3 items – corrections to tables [15 marks]
Table 1.24 Task types from the Oxford International Business EnglishCertificate
tions from a clear specification if stakeholders are to make consistent inferencesbased on test scores.
Both the CEIBT and the OIBEC contributed to the current approach throughthe move along the specificity continuum, to a situation where the tests whichhad been based on a high degree of specificity (and low generalisability) were, intheir later guises, more centrally located, allowing for a greater degree of gener-alisability than their predecessors.
Business Language Testing System (BULATS)First discussed by members of the Association of Language Testers in Europe(ALTE) shortly after the formation of the association in 1990, the main thrustbehind the development of BULATS appears to have been the decision to createa series of business language tests with a multilingual dimension. To date, testshave been developed in four languages, English (which will be the main focus ofthis review), German, French and Spanish. The tests were developed andmanaged by Cambridge ESOL [English], Alliance Française [French], Goethe-Institut [German] and Universidad de Salamanca [Spanish].
One interesting feature of the BULATS tests is that performance on all of thetests is benchmarked to the ALTE and CEF frameworks, shown in Chapter 2.This allows the test end-user to make informed decisions about performance ontests of proficiency in different languages. While it can be argued that theLCCIEB tests allow for the same cross-language comparisons to be made, thereis a big difference in the tests involved. As we could see from the LCCIEB tests,they all follow the same model. The difference with the BULATS tests is thateach test is developed and administered independently by experienced test
BULATS
63
First Level Executive Level
Speaking
Marks awarded
Time allowed
Format
Task 1
Task 2
Task 3
Not specified
20 minutes
2 candidates, 1 examiner
Each candidate makes a shortpresentation (2–3 minutes) on themerits of a particularmarket/strategy. Followed by shortdiscussion. Each given prompt cardcontaining bulleted pros of ownpoint and cons of that of othercandidate. Finally, candidates mustcome to a decision on which to gofor.
80 marks
20 minutes
2 candidates, 1 examiner
Presentation – (no time suggested)choice from 5 prompts (5 minutes’preparation time)
Decision-making task – from givenprompt cards, candidate to candidateinteraction (2 minutes’ preparation time)
Joint summary of findings [not marked]
Table 1.24 Task types from the Oxford International Business EnglishCertificate (continued)
developers who are native speakers of the target language. They may beworking from the same basic specifications, but here the developers are morelikely to be aware of subtle differences in the language concerned and to takethis into account in developing tests that are more likely to represent a validindication of proficiency in that language for the particular purpose tested.
Another facet of BULATS is the fact that it offers a number of independenttests in each of the four languages currently tested. These tests are:
• the BULATS Standard test – a 110 minute test of listening, reading andgrammar/vocabulary
• the BULATS Computer test• the BULATS Speaking test• the BULATS Writing test.
Each of these tests will be briefly reviewed in the following parts of this section.
The BULATS Standard test
As mentioned above, this is a 110-minute-long test of reading, listening andgrammar/vocabulary. The test is divided into two sections, the listening partlasts for 50 minutes and the reading and language knowledge part lasts for 60minutes. From the outline of the test in Table 1.25, we can see that it represents asubstantial measure of a candidate’s proficiency in these areas (BULATSundated/a).
Table 1.25 BULATS Standard test (English): test outline
All of the examples described below are taken from the sample paper availablethrough the BULATS website – the entire sample paper for English is includedas part of Appendix 1.4 (see Appendix 1.6 for a copy of the BULATS Germanpaper).
1 Introduction to the testing of language for business purposes
64
Part Items Format Focus
Listening
1234
10121018
MCQSAFMatching MCQ
Matching audio description to visuals or short phrasesMemo/form completionIdentify speaker from list of topics/jobs etc.Listen for detail
ReadingandLanguageKnowledge
1 7665
MCQMCQMCQCloze
Reading short memos, signs etc.Selecting appropriate lexical items or chunksReading for comprehension (300–350 words)
2 755667
Matching MC ClozeClozeMCQMCQSAF
Statements to short texts (up to 60 words)Based on 100 word business communicationBased on 100 word general textSelecting appropriate lexical itemsReading for comprehension (500–600 words)Proof-reading – identify and correct errors in text
There are four parts to the Listening section. In the first part, candidates areasked to match an audio description to a set of three visuals or short phrases. Nowriting is expected of the candidates in this section. In total there are 10 items. Inthe example from the sample item below (Figure 1.16), the candidates are askedto listen to the input and to identify a specific piece of information (here, deliverydate).
Figure 1.16 Sample item: Part 1 BULATS Listening (English)
In the second part, candidates listen (just once) to a set of three conversations ortelephone messages, and must complete a series of forms, notes or memos,totalling 12 items – all short answer format (SAF). Figure 1.17 shows anexample of this task type, where the candidate listens again for specific infor-mation and responds using one or two words or numbers.
Figure 1.17 Sample item: Part 2 BULATS Listening (English)
BULATS
65
In the third part of the test, candidates listen to five people talking about aparticular topic – there is no interaction here, all input is in the form ofmonologic discourse. They should then identify the speaker (from a given list ofspeakers). In the example shown in Figure 1.18, the speakers are talking abouttheir work. Candidates listen and respond by identifying the views held by eachof the five speakers from the list provided.
Figure 1.18 Sample item: Part 3 BULATS Listening (English)
Finally, candidates listen to a series of three short interactions (see Figure 1.19)and are asked to respond to a set of six MCQ-based items for each listening text,only two of the six items have been included in Figure 1.19. In contrast to theprevious section, here the discourse is interactional in nature. It appears that thedevelopers have attempted to avoid, or at least to limit any test method effect byincluding a range of methods in this part of the test. While there is some readingto be done in order to respond to the items, this is minimal, with the possibleexception of Part 3 – where the options range from two to six words in length,though there are only three options.
The Reading and Language Knowledge section of the test is comprised oftwo parts. Within these parts there are a number of sub-sections (see Table 1.25)which focus on various aspects of reading and language knowledge. This part ofthe test is not as clearly defined as the first part, with the candidates moving fromreading short texts and notices in Section 1 to demonstrating their knowledge ofbusiness-related vocabulary in the following section. In the first part, candidates
1 Introduction to the testing of language for business purposes
66
are presented with seven MCQ items all based on short pieces of input (whichcan be notices, memos, notes, graphics or tables). As with the earlier listeningsection, the MCQ items all have three options though they are all quite long – upto 10 words (see Figure 1.20). All responses are marked directly on to acomputer readable answer sheet.
Figure 1.20 BULATS Reading and Language Knowledge: Section 1
The following section includes a series of six MCQ items based onknowledge of language use (see the example in Figure 1.21). Here the candidateis asked to identify the most appropriate word or phrase to complete a short
BULATS
67
Figure 1.19 Sample item: Part 4 BULATS Listening (English)
sentence. The focus here is on the lexicon (including lexical chunks) of thebusiness domain.
Figure 1.21 BULATS Reading and Language Knowledge: Section 2
This section then moves on to a series of reading comprehension items whichfocus on reading for detail from a text of approximately 300–350 words. Thissection is then followed by a cloze test consisting of five items which appear tobe designed to test syntax. While the previous section is quite clearly based on abusiness-oriented text, the text on which the cloze is based is less obviouslybusiness-focused.
1 Introduction to the testing of language for business purposes
68
Figure 1.22 BULATS Reading and Language Knowledge: Reading item
In the second part of the Reading and Language Knowledge paper there aresix sub-sections. Here, the focus is on a mixture of reading (Sections 1, 3 and 5),vocabulary (Sections 2 and 4) and grammar (Section 6).
Reading is tested using a variety of item types, matching, cloze and MCQ,with the focus on reading for detail throughout. Figure 1.22 is an example of anitem from Section 1 of this part of the test. In this item candidates are expected tomatch the statements to one of a series of four short texts (I’ve included only twoitems and one of the four texts here – for the whole section see Appendix 1.3).Vocabulary is tested using two different formats (cloze and MCQ), an exampleof the latter is shown here as Figure 1.23.
Figure 1.23 BULATS Reading and Language Knowledge: Vocabularyitem
The final part of the test consists of a short letter or memo, on each line ofwhich there may be an error. Test takers are expected to identify which situationapplies to each line (correct or including an error) and to indicate whatcorrection is needed where an error has been identified – see Figure 1.24 for anextract from the sample paper supplied by the developers.
Figure 1.24 BULATS Reading and Language Knowledge
The BULATS Standard test, therefore, offers a comprehensive measure of atest taker’s receptive language proficiency and their knowledge of the structure
BULATS
69
and lexicon of the language as it is used in a business context. The fact is that theStandard paper is not meant to offer a broad perspective on the language abilityof candidates; instead both it and its computer counterpart are supported byadditional papers for speaking and writing, so comment should not really bepassed on these independent units in terms of approach or construct.
The way in which the test is constructed is interesting: apart from being splitalong the listening–reading/knowledge divide, within the two parts thereappears to have been a deliberate attempt to keep shifting the focus, by movingbetween different types of item and content – particularly true of the Readingand Language Knowledge paper. This is a situation that might not please all testdevelopers or theorists, as it could be argued that the skills might be moreefficiently tested in more compact and homogenous sub-tests. On the otherhand, the constant changing may act to maintain interest in this long paper andmay actually facilitate more accurate measurement. This is an area on which thetest developers might well devise a programme of research in which the impactof the presentation style is investigated.
The BULATS Computer test
The BULATS Computer test is a computer adaptive (CAT) version of theinstrument and like the standard version contains sub-tests of listening andreading comprehension, and vocabulary/grammar tasks. The computer versiontakes advantage of the alternative item types offered by the medium.
The test includes a variety of listening item formats:
• listening to a short monologue to identify the correct response to a writtenitem
• listening to an extended dialogue to answer a series of comprehension items.
The reading items also offer a range of item types:
• reading short texts to identify the correct summary• reading an extended passage to answer a series of comprehension texts.
Finally, the vocabulary/grammar items tend to use one of two types of clozeitem:
• responses from a series of four options• responses typed directly into text boxes.
We can see, therefore, that the format of the test reflects that of the Standard test,though there are a number of different item types used. Another unique featureof BULATS is the way in which it allows the test user to make a number ofdecisions which contribute to adapting or customising the test to suit the needsof their situation. A management dialogue box allows the user to indicate whichdemographic information to include, to decide on the test-supervisor language,
1 Introduction to the testing of language for business purposes
70
the language of instruction (in this case the person might wish that the instruc-tions be given in the candidate’s mother tongue – provided it is on the list ofoptions offered – or that they will be in the target language, here English). Inaddition, this screen also permits the user to decide to allow (or not) the test takerto view their results, to print them or view feedback – depending on the contextand purpose of the test all or none of these options might be chosen. As withother CAT tests the results are available immediately upon completion of thetest.
Like the Standard version, the Computer test is available in a number oflanguages, though only the English version is reviewed here due to limitationsof space. The test is available on CD in each of the four languages (English,French, German and Spanish).
The BULATS Speaking test
This test is independent from the other BULATS tests. The Speaking test usesthe one-to-one format, with a single examiner and test taker (see Appendix 1.5for a sample paper). All tests are audio recorded and assessed by an independentassessor, as well as by the examiner who participates in the test (BULATSundated/b). Table 1.26 shows how the test is organised.
Table 1.26 Speaking test design – BULATS
Performance is assessed on accuracy and appropriacy of grammar and vocab-ulary, discourse features such as cohesion, fluency, pronunciation, interac-tiveness, and degree of accommodation required.
Some example tasks for the Speaking test are presented below. The presen-tation task (Figure 1.25) offers a guided or scaffolded task prompt, where the testtaker is given some bulleted points which should be included in the presentation(it is not clear though if there is some penalty for not including these points in the response – in other words, it is not clear if they are suggestions or explicitdirections).
BULATS
71
Part Title Timing Description Focus
1 Interview 4 mins Personal informationexchange (answer questionsabout themselves, theirwork and interests).
Ability to respond to personalquestions in a conversationalcontext.
2 Presentation 4 mins Talk on topic (choice ofthree) for one minute – oneminute preparation time.Respond to follow-upquestions.
Ability to produce extendeddiscourse and to respond toquestions on the topic.
3 Informationexchange anddiscussion
4 mins Simulation – role play fromgiven input (candidateexpected to take initiative).
Ability to take a more activepart in a conversation.
Figure 1.25 Speaking task types from BULATS (Part 2)
The information exchange task (Figure 1.26) is again scaffolded, though herethere is clearly room for the test taker to demonstrate an ability to expand on thetopic and to offer their own opinions on aspects of the topic. This has the effectof expanding the range of language functions typically observed in an interview(informational) to include both interactional and discourse managementfunctions, see O’Sullivan, Weir and Saville (2002) for a discussion of thisphenomenon.
1 Introduction to the testing of language for business purposes
72
PART 2 Presentation
INSTRUCTIONS
Please read all THREE topics below carefully.
Choose ONE which you feel you will be able to talk about for one minute.
You have one minute to read and prepare your talk.
You may take notes.
Topic A
Describe an important business meeting you attended.You should say:
where it was;what it was about;why it was important.
What were the most interesting moments?
Topic B
Describe someone you particularly enjoy working with.You should say:
what this person does;what sort of work you do with this person;why you like working with this person.
Would you change anything about this person? Give reasons for this answer.
Topic C
Describe the best workplace you have ever had.You should say:
where the workplace was;what you were doing there;why you liked to work there.
Would you change anything about it? Give reasons for your answer.
Figure 1.26 Speaking task types from BULATS (Part 3)
The inclusion of a variety of tasks, each with a different focus, marks an inter-esting attempt to extend the range of discourse type. The paper includesinformal interactive personal information exchange, formal presentation andinformation exchange tasks.
The fact that there is a choice of situations offered to the candidates in Part 2is obviously an effort to ensure that they have an opportunity to perform at theirbest by selecting a topic on which they feel they can perform well. There isalways a danger, of course, that particular topics are either inherently more orless difficult than others, or that the examiner will consider that this may be thecase. This opens up the possibility of the examiner compensating the candidatefor selecting a ‘difficult’ topic – even where the topic may not actually be moredifficult for the candidate. As with many areas of performance assessment, thisis a matter that has received scant attention (though see Lumley and McNamara,1995).
Though no empirical evidence has been published to date, it would be inter-esting to see how the final information exchange task works in actual adminis-trations of the test. This format has been found not to work well in a number oftests as the test takers are often reluctant to adapt to a role – this could be due tothe difference in power and status between the examiner (the ‘expert’) and the
BULATS
73
PART 3 Communicative Activity
CONFERENCE ARRANGEMENTS
You have one minute to read through this task.
Information Exchange
You are making the arrangements for a one-day conference at a local hotel. The Examiner is the
Conference Organiser for the hotel and is visiting you to discuss the conference.
Find out this information:
i) the size of the largest conference room
ii) the cost for that room
iii) equipment available
Do you think the hotel is offering you a good service for the price it is charging?
Discussion
Now discuss this topic with the Examiner.
What makes a successful conference?
test taker (the ‘novice’). In a test such as BULATS this position is reversed to alarge extent, through the creation of a ‘work-based’ situation, in which the testtaker is the ‘expert’ and the examiner the ‘novice’. This is just speculation at thispoint, though it is certainly worth exploring.
The BULATS Writing test
The BULATS Writing paper consists of a pair of writing tasks, described (Table1.27) and exemplified below (BULATS undated/c). Performance on the tasks isassessed by two trained and accredited examiners working independently of oneanother. The criteria used are accuracy and appropriacy of grammar and vocab-ulary, organisation of ideas, achievement of purpose. As with the otherBULATS test papers, the topic and genre of the writing tasks are contextualisedin a business setting.
As can be seen in Figure 1.27, in the first of the tasks the candidate is given ashort text, such as a letter, memo or advert, together with a set of guidelines forwriting a reply or follow-up letter. Candidates are expected to cover all of thepoints in the instructions within about sixty words – though there are nopenalties for going over that limit. The task is typical of the business domain interms of purpose, length, structure and formality of expected output. It cantherefore be seen as being appropriate in terms of both the text and task demandsof the target domain.
The second task (Figure 1.28) offers candidates a choice of either an extendedletter or a report. The same can be said of this choice as was said of the choiceoffered in the speaking test, and the developers would be well advised tomonitor these options for any unintended bias. On the other hand, both tasks arevery definitely focused on the business domain, and like the first task, the devel-opers can claim that the options represent tasks that are very strong in terms ofsituational authenticity. The tasks are also quite likely to result in interactionallyauthentic performances as they again reflect the task and text demands of thebusiness domain.
1 Introduction to the testing of language for business purposes
74
Part Title Timing Focus
1 Short message/letter(50–60 words)
15 mins Ability to write a short letter, covering (given) relevantpoints and using appropriate style and tone.
2 Extended letter orreport (180–200words)
30 mins Ability to write and structure a piece of extendedwriting, using appropriate style and tone for theintended reader.
Table 1.27 Writing test design – BULATS
CommentaryThe BULATS tests offer an interesting insight into the way language testing ingeneral and specific purpose language testing in particular began to change inthe early 1990s.
Before this period, the traditional Cambridge ESOL approach (whichtypified the ‘British’ approach) had been to focus primarily on performance-based assessment. By this I mean that the tests had been shaped over the years toreflect a current view of the learning process, see Weir’s history of the growth ofthe CPE (2003a), while the need to reflect contemporary thinking on psycho-metric aspects of language testing seemed to take second place. The BULATStests were designed at a time when the influence of psychometrics was still quitestrong, with, for example, the TOEFL/First Certificate in English (FCE) compa-rability study (Bachman et al 1995) suggesting quite strongly that the latter testlacked adequate psychometric quality and issuing dire warnings of the conse-quences of this apparently fatal flaw. The tests, far from abandoning the existingphilosophy can be seen to have moved to embrace the two, often conflicting,movements, by combining a variety of item and task types as well as a variety ofresponse types. BULATS also includes papers on all four skills in addition to a
BULATS
75
Figure 1.27 Writing task from BULATS (Part 1)
separate grammar/vocabulary paper – as we will see later, this reflects the typeof test associated with one of the main BULATS partners (Cambridge ESOL).
Of interest here is the way in which BULATS can be interpreted in terms ofthe degree of specificity issue. It is clear from the examples shown previouslythat the different papers seem to be taking somewhat different perspectives onthe candidates’ ability, with some being quite specific in their content andcontextualisation (suggesting a high degree of situational authenticity), whileothers are apparently deliberately less focused on the business context. Thisrange supports the notion that specificity is not as straightforward as we oncethought. When a test is as complex as the one described here, there will be arange of degrees of specificity within the test (see Table 1.28).
1 Introduction to the testing of language for business purposes
76
Figure 1.28 Writing task from BULATS (Part 2)
Table 1.28 BULATS – degree of specificity in the different papers
In a similar way, we can discuss the related issues of authenticity and theimpact of non-language features on performance. The variation throughout thistest is highly likely to be reflected in tests in which a similarly complex design isused, and is equally likely to result in a broader perspective on the candidate’slanguage ability within the business, or other specific, context and hence to thedrawing of more valid inferences from performance on the test as a whole.
The development of the BEC suiteThe origins of the Business English Certificates (BEC) can be traced to aseries of meetings during 1991–1992 between UCLES representatives andthe National Education Examinations Authority (NEEA) in China. At thesemeetings, the area of business English was identified by the Chinese partners asbeing in urgent need of a new, fresh approach, one designed specifically for aChinese population.
By the end of 1992, the decision to develop such a test had been made, alongwith the decision that the test should be certificated at a low level. With this inmind, a prototype was developed using the Key English Test (KET)/Prelim-inary English Test (PET) model – these represent the lowest levels of theCambridge ESOL Main Suite general proficiency tests. This prototype includeda detailed specification and sample paper.
The partners agreed that the prototype represented an appropriate design anda decision was made in early 1993 to proceed with the development of the test asa joint venture. This led to a detailed exploration of the practical issues involvedin operationalising such a project, issues such as marking, processing, cost,printing, etc. At this early stage it became apparent that the proposed speakingpaper would be problematic from the perspective of examiner recruitment (itshould be remembered that, at that time, there was a serious shortage of qualifiedand experienced English language teachers in China). For this reason, it wasdecided that only those students who had successfully completed the other testpapers would be offered a speaking component.
Development of the BEC suite
77
Paper Degree of specificity Comment
Listening Medium/High Quite a large emphasis on social language, though withclear business-oriented contextualisation
Reading Low/Medium Some focus on business-related text types
Grammar/Vocabulary
Low/Medium Some focus on business-related text types
Speaking Medium Essentially based on more social aspects of spokenlanguage use (though again set in a business context)
Writing High Very much focused on writing in a business context
The first administration of the BEC took place in China in the autumn of1993. Table 1.29 gives some idea of the scale of that first administration.
Table 1.29 Details of the first BEC administration (China)
Even before the first administration of the BEC, it was decided that the existingtest should be supplemented with another, higher level test, envisaged as beinglinked to the existing test in terms of design model, though aimed at a higherlevel. This meant that the Business English Certificates (as the suite was nowcalled) was to consist of a pair of related examinations, called BEC1 and BEC2.
The design and planning phases of the new test were completed by late 1993,and an operational test was developed during the summer of 1994. This new testwas first administered, again in China, in the autumn of 1994. Table 1.30outlines the scope of the 1994 administration.
Table 1.30 The 1994 BEC administration (China)
Following the early burst of development, there followed a hiatus in whichthe existing tests became well established in the ‘base’ market of China. Duringthis time interest in the test in other Asian countries, particularly in India, beganto grow. This growth into other markets was not seen as being problematic, asthere was nothing in the test design that might cause it to be of use only in aChinese context.
Meanwhile, changes in the demographics of the test population, both inChina and in the newer markets, resulted in an increased demand for a test at ahigher level to the existing pair, an idea that had been in existence at the time ofthe CEIBT review in 1994. Extended discussions at this time into the feasibilityor need for an addition to the BEC suite, were influenced by the existence of theCEIBT (see the discussion of its development and administration above), whichhad been designed to test language at a level comparable with the proposed test.Eventually, however, operational difficulties with the CEIBT (again see above),and the expressed preferences of BEC stakeholders for any new test to have adesign similar to that of the existing BEC examinations, led to the decision todevelop what was to become known as BEC3.
Work on the new test began, with the test going live in 1996. This new testwas planned to extend the range of the BEC suite upwards, and was bench-marked to the Certificate in Advanced English (CAE), ALTE level C1. The
1 Introduction to the testing of language for business purposes
78
Number of candidates% of candidates achieving a passing grade
321297%
BEC1 BEC2
Number of candidates% of candidates achieving a passing grade
497493%
312172%
design of the test was again based on the other BEC examinations, offering thesame range of papers. This design is summarised in Table 1.31.
Table 1.31 The BEC suite design
During the period 1998–2000, the BEC suite spread to other parts of the worldand by the end of this time the overall candidature had grown to over 45,000.Changing demographics within the BEC population, related again to changes inthe original market, and changes related to the expanding candidature set thecontext for the revision with which the next part of this book is concerned. Thefollowing chapters outline the Cambridge test development cycle in relation tothe revision of the BEC suite (Chapter 2) and the actual changes to the testpapers (Chapter 3).
Issues resulting from this review
It would appear from this review that there are a number of different approachesto the testing of language for business purposes. The tests reviewed appear to beless than highly specific, in that they are more likely to focus on language use ina particular context, than on the performance of very specific context-relatedtasks. The fact that there is a range appears to support the argument made above,that the specificity continuum exists and that tests placed at different pointsalong the continuum will differ not only in terms of specificity, but also in termsof situational and interactional authenticity, and in terms of the impact of non-language factors on the abilities being tested.
There is also evidence here of a difference in rationale for including tasks in aspecific language test. In general purpose testing, the primary reason forincluding particular tasks is to elicit samples of language which can then beevaluated by a trained rater. On those occasions when task completion may berelevant, we can usually trace the relevance to the specific purpose of thatportion of the test – an example of this in a performance test of writing would bewhere the test taker must complete a job application form, a task that goesbeyond the bounds of general language use. In tests of language for specificpurposes, the notion of task completion becomes more central.
Here, the test taker is often explicitly judged, along with other predeterminedlanguage-related criteria, on whether a particular task has been adequately, orsufficiently completed – this is where Elder’s (2001) inseparability argument ismost clearly seen. It can be argued that relevance and adequacy of response is a
Development of the BEC suite
79
BEC1 BEC2 BEC3
Paper 1 – Reading and WritingPaper 2 – ListeningPaper 3 – Speaking
•••
•••
•••
feature of all tests. However, the primary purpose of a task in general languagetests is to elicit a sample of language which will be judged on its linguistic merits(where relevance and adequacy are features of sociolinguistic and pragmaticcompetence). The issue in LSP tests is to decide to what extent the relevance andadequacy of task performance should be judged in relation to the language usecontext in addition to its linguistic merits.
Considering these issues again, together with the suggestions made inrelation to degree of specificity and generalisability above, it might be useful tore-conceptualise specific purpose tests in general, and business language tests inparticular. In order to do this, it is necessary to revisit the three core areas ofconcern suggested by Elder and Douglas.
1. Degree of specificityIf we take the continuum suggested earlier and extend it to its naturalconclusion, that is infinity (represented by the symbol ∞) in both directions(Figure 1.29), it becomes obvious that while a theoretical conceptualisation ofthe extremes is possible, a practical application of these extremes is not. This canbe seen even in the test described by Teasdale (1994) where within the languageof air traffic controllers there will of course be unique or precise aspects of thelanguage, but there will always be a proportion of non-precise language.
Figure 1.29 The degree of test specificity continuum
Since there is a clear link between the degree of specificity and the definition ofthe construct – in that changes to one will affect the other – the obvious impli-cation will be that the inferences that can be drawn from performance on a testtask will be related to the degree of specificity of that task.
2. AuthenticityIn addition to the notion of specificity, the other principal concern with businesslanguage and other LSP tests is that of authenticity. It appears from the briefreview of current practice offered previously, that a task, and in particular a taskrelated to the receptive skills, can normally be shown to have only a measure ofsituational authenticity – though for an example of a truly situationally authentictask see Abdul-Raof (2002) whose participants actually performed realconference presentations that were video recorded and later evaluated bycolleagues from the same profession, as well as by language specialists.
As for interactional authenticity, task performance is clearly affected by theparticipants in that performance, and since its presence (or absence) is therefore
1 Introduction to the testing of language for business purposes
80
Non-Specific Degree of Specificity Very Specific
oo oo
subject to factors outside of the control of the task writer/test developer itappears to be a somewhat unrealistic expectation that all administrations of a testwill be found to demonstrate interactional authenticity. Instead, it seems morereasonable to suggest that tasks can be shown to demonstrate this aspect ofauthenticity under particular operational conditions, but not necessarily underall operational conditions. In other words, if it can be shown that a typicalsuccessful test taker will be prompted by the task to demonstrate an interactionbetween their communicative competence and features of the specific targetlanguage use domain, then interactional authenticity can be claimed of the task.While this operationalisation may not be precise enough for some readers, itdoes represent a practical and measured solution to the problem.
3. Impact of non-language factorsThe interesting thing about this feature of tests of specific purpose is the fact thatthe impact appears to be most obvious where the test is more highly specific, inother words, where it is more difficult to separate the different elements of theability being tested. This was exemplified in the tests reviewed above where agreater effort had been made to situate the test more clearly in the specificpurpose domain. Here, there seems to have been a greater likelihood that theperformance might be influenced by non-language factors.
The point of interest here is that there are a number of potential sources ofimpact, and that these are not only related to business ability or knowledge. Infact, the sources are related to the task itself and to the way in which the task isassessed. The implication is that the more specific a test the more likely theimpact of non-language factors.
The more complex tests reviewed above demonstrate that ‘degree ofspecificity’ is not necessarily a notion that can be applied to a test as a whole.Instead, it is certain that these complex tests will contain papers, and evensections of papers, that have been deliberately manipulated so as to be more orless specific in focus. This suggests that the impact of non-language factors willalso vary within a test.
Development of the BEC suite
81
82
The revision of BEC
The Cambridge ESOL test development/revisionmethodologyThe Cambridge ESOL approach to test development and revision is essentiallycyclical and iterative in nature (as can be seen from the summary presented inFigure 2.1). Like all other Cambridge ESOL tests, the original BEC tests weredeveloped using this methodology, and again like other tests, in time a numberof elements combined to create a perceived need for a revision. Among theseelements were advances in test production methodology (many linked to thevarious projects described below), and changes to the test candidature. Theoriginal BEC examinations were designed primarily for the Asia–Pacificregion, particularly China, and as the candidature grew in size over the years, italso changed with the growing international interest in the suite. The decisionwas therefore made in 1998 that any revision of the test should be undertakenwith this wider candidature in mind. Other factors which influenced theperception of the developers included an expansion of our knowledge of howlanguage is used in the specific context of business (through developments incorpus linguistics for example), to a general broadening of our understanding ofthe whole area of language testing. All of these combined to impact on thedecision to instigate a revision in 1999.
The Cambridge ESOL frameworkThe following review of the Cambridge ESOL framework will begin by re-stating the general approach to testing language that informs the framework (seeSaville 2003). This approach is the main driving force behind all CambridgeESOL test development projects.
The Cambridge approach
Saville identifies ‘five main factors’ which underpin the Cambridge ESOLapproach.
These are:
2
The Cambridge ESOL framework
83
1. To assess language skills at a range of levels, each of them having aclearly defined relevance to the needs of language learners.
2. To assess skills which are directly relevant to the range of uses to whichlearners will need to apply the language they have learnt, and cover thefour language skills – listening, speaking, reading and writing.
3. To provide accurate and consistent assessment of each language skill atthe appropriate level.
4. To relate the examinations to the teaching curriculum in such a way thatthey encourage positive learning experiences and to seek to achieve apositive impact wherever possible.
5. To endeavour to be fair to all candidates, whatever their national, ethnicand linguistic background, gender or disability (Saville 2003:62).
Assessment of a wide variety of language skillsThe BEC suite examinations, like the other Cambridge examinations, includethe full range of language skills in their design. That is, all three levels of theBEC suite consist of papers devoted to the assessment of proficiency in the four
Figure 2.1 The Cambridge ESOL test development model
Source: Saville (2003:79)
LIVE Test
Planning Phase
Design Phase
Development Phase
Operational Phase
Monitoring
Review
TriallingAnalysisEvaluation Review
Initial specifications
Evaluation
Revision
Perceived need for newor revised test
Start
2 The revision of BEC
84
skills of listening, speaking, reading and writing. In addition, the differentpapers offer a wide range of response formats through the inclusion of a varietyof tasks and item types within each skills’ paper. The benefit of including avariety of task types in the Speaking paper, for example, has been demonstratedby the recent work of O’Sullivan, Weir and Saville (2002) who, when devel-oping a set of observation checklists for monitoring test task responses (in termsof language functions elicited) were able to show that the different task typesresulted in strikingly different function profiles.
A system of criterion levels The BEC suite consists of a set of three examinations, each of which has beendevised to target a distinct level of ability. Like the Cambridge ESOL MainSuite examinations these have been linked to the Common European and ALTEframeworks through a process of benchmarking candidate responses to ‘can do’questionnaires (essentially a series of self-assessment instruments developed toelicit from test candidates estimates of what they ‘can do’ within the four skills’areas in different performance contexts – social, study and work). The impact ofthis project on the BEC revision is described below, but also see Jones (2000,2001b) for a clear outline of the project.
This whole approach allows us to view the three examinations not simply asunique measures, or even as a set of linked measures covering a broad spectrumof language ability within a business context, but essentially as a single unit,with individual elements focused on particular criterion levels of proficiency.The greatest benefit of such a system is that it allows us to make comparisons oftests both vertically (in that they can be shown to measure distinct levels oflanguage proficiency) and horizontally (so that each distinct examination can beshown to represent an empirically described level of ability).
Another advantage of this criterion levels’ approach is that it permits us toview an estimate of attainment within any single test in terms of a broad multi-level range of language ability, rather than within the confines of a single level.The implications of this will be discussed in the relevant section below.
The ALTE Can Do scales were developed to provide a series of criterion-related statements at each of the levels covered by the BEC suite in relation tothe specific domains which are covered in these examinations (situatedlanguage use for social and work purposes). Together with the criterion scale,the Can Do scale provides an external benchmark through which stakeholderscan establish a meaning for reported performance levels.
Assessment for a variety of purposesThe BEC suite of tests are a good example of how Cambridge ESOL has concen-trated on the creation of a range of tests and examinations which are designed fora variety of purposes rather than relying on a single test to address many
The Cambridge ESOL framework
85
purposes. Even within these tests there is a recognition that specific purposelanguage tests are context-oriented rather than context-focused. By context-oriented we mean that the tests are set in the context of business and will includelanguage that is socially-oriented as well as business-oriented, in recognition ofthe fact that much specific purpose language combines these two areas. Context-focused refers to tests that are designed to test only business language. In fact,the LCCIEB needs analysis project (LCCIEB 1972) quite clearly demonstratedthat a context oriented approach is most likely to reflect actual practice in thebusiness language domain. The tests reviewed in Chapter 1 demonstrate that thecontext-oriented approach is typical of current practice in the area.
A commitment to quality and fairnessThe traditional conceptualisation of fairness focuses on technical aspects oftests, such as the reliability of sub-tests. However, the view of fairness that isnow more commonly accepted incorporates more wide-ranging considerationssuch as the production and validation of test materials and assessment proce-dures. Recent events in national testing systems in the UK (failures in test datamanagement systems – Scottish Qualifications Authority 2000; test security –Edexcel 2001; and in test editing Edexcel 2002) demonstrate that these aspectsof a test’s development are as relevant to test fairness as the technical aspectsreferred to above. Cambridge ESOL ensures test quality through a system oftotal quality management, where a series of quality checks are put in place at allpoints of the development and administration process, see Weir and Milanovic(2003).
An ongoing programme of test revisionOne of the great advantages of the Cambridge ESOL commitment to research(both qualitative and quantitative) throughout its different suites of examina-tions and test systems, is the way in which research findings in one area routinelyfeed into other apparently unconnected examinations. Examples of this includethe work in the early 1990s on the development and validation of the use of inter-locutor frames in tests of speaking (first envisaged as a methodology forcontrolling input in the Main Suite Speaking papers, but now used throughoutthe Cambridge ESOL examinations); the development of the observationchecklists (originally developed as part of the CPE revision project but now used– in different guises – across the Cambridge ESOL range of tests); and not leastin the development of quantitative analysis tools for equating tests in particularexamination suites – a development of particular interest when it came to therevision of the BEC suite.
The title of a presentation made by Weir (2002) at the annual IATEFLconference in York sums up the commitment of Cambridge ESOL to an ongoingprogramme of review and revision. The title (The History of the CPE,
2 The revision of BEC
86
1913–2013) demonstrates that that particular test, which had just undergone amajor revision, was already being reviewed, at the item, sub-test, and paperlevel, so that any future revision is based on an accurate longitudinal picture ofboth how it performs and how it is perceived throughout the life of this currentversion. In the same way, the BEC suite is also under constant scrutiny. Thisprocess of ongoing revision is also to be found in a number of the tests describedin Chapter 1 – indeed a number were revised during the writing of this book.Sadly, there are still tests out there that have not changed since their introduction(e.g. the TOEIC is still essentially the same test as was introduced a quarter of acentury ago – even though the way in which we understand and engage withlanguage and communication has changed radically in the intervening period).
Some of the five elements are related to what we might call ‘core’ values ofthe Cambridge ESOL organisation – the testing of multiple skills has been adefining feature of Cambridge ESOL examinations since the introduction of theCPE in 1913 (Weir 2002, 2003a), while the commitment to the creation of testinstruments and systems for use in a variety of contexts and for a variety ofpurposes is also long established. Since this book is meant to focus primarily onthe BEC revision, it would be more interesting to look at the process in terms ofhow things like criterion levels and ongoing validation/revision are dealt withand how the developers ensure that the reported grades are accurate andconsistent.
A system of criterion levelsAs mentioned above, the individual tests in the Cambridge ESOL examinationsand test systems are designed to be seen not in the context of a single level, butwithin a wider multi-level context. This concept was realised through theCambridge ESOL Framework Project (see Jones 2000, 2001b) which resultedin a practical and useful instrument which has been used by the organisation toclassify its examinations within a common system of levels.
With the formation of the Association of Language Testers in Europe(ALTE) in 1990, the work on the framework project expanded to involve collab-oration with other international organisations (such as the Council of Europeand the European Association for Quality Language Services or EAQUALS)and fellow ALTE members. This expansion also broadened the aims to includesome of the following key areas of activity:
• ALTE and Common European Framework • ALTE CAN DO project
– Development of CAN DO scales – Validation of the scales
• Linking learner-responses to their performance on examinations • Linking ALTE Can Do Statements to the CEF
• production of Multilingual Glossary of Testing Terms in 10 languages
• production of guidelines for training item writers, including the Councilof Europe Users Guide for Examiners as supplement to the CommonEuropean Framework
• development of Content analysis checklists for analysing and comparingexaminations
• an evaluation of the Council of Europe’s Vantage Level (UCLES2000:2).
One aim of this expanded view of the project was to promote what Jones referredto as ‘the transnational recognition of certification in Europe’ (2000:11). Theproject also identifies a series of distinct levels of language ability and as such isideal as a benchmark against which individual tests are measured. This facili-tated the other aim of the project, which was to link levels of language abilityacross European national boundaries to a common proficiency scale.
While a complete description of the project is clearly beyond the scope of thisbook, it may be useful at this juncture to briefly overview its central elements.Figure 2.2 outlines the project, though does not do justice to the complexity or tothe range of different sub-projects that contributed to the overall design.
Figure 2.2 The Cambridge ESOL/ALTE framework project (outline)
Though this volume is dedicated to the BEC revision, and not to the frameworkproject, it is clear that all parts of the project have had a direct impact on the BECrevision process.
The impact of the 5-level system/Common EuropeanFramework
The ‘Can Do’ project – see Jones (2000, 2001b) for an introduction – wasdevised with the principal aim of providing a comprehensive description of
A system of criterion levels
87
Descriptive System foritem-based tests
Local Item BankingSystem (LIBS)
Investigating candidatebackground and test
performance
Linking learner responsesto their test performance
Linking the UCLES/ALTE
framework to the CEF
The Framework Project
The UCLES/ALTECommon Scale
Benchmarkingspecific tests to the
scale
Development of test‘quality assurance’
structures & instruments
The ALTE ‘Can Do’Project
The Can Do Scales
Good User
Competent User
Independent User
Threshold User
Waystage User
Breakthrough
Qualitative Data
Quantitative Data
2 The revision of BEC
88
what language users can typically do with the language at a number of distinctlevels, in the various language skills and in a range of contexts. The project wascreated with the purposes of:
• helping end users to understand the meaning of exam certificates atparticular levels, and
• contributing to the development of the Framework itself by providing across-language frame of reference.
Basically, the ‘Can Do’ project was meant to offer a practical guide to the appli-cation of the framework in test development. This is summarised neatly in thefollowing figure (Figure 2.3) from Jones (2001a). Here, we can see that theproject aimed to provide a framework through which examinations for differentlanguages and contexts could be compared.
Figure 2.3 The aims of the ‘Can Do’ project
Figure 2.4 shows where the three BEC examinations are designed to fit withinthis system of criterion definition. In order to ensure that this relationship ismore than just at a superficial level, a series of research studies was carried out.These focused on the exploration of the nature of the relationship from a quali-tative perspective by using expert judgements to establish links between each ofthe three BEC tests and a relevant ALTE level. In addition to this qualitativedata, quantitative data generated by the ALTE ‘Can Do’ project providedadditional support for the equivalence claims implicit in Figure 2.4.
Description of level in termsof typical patterns of ability:What a person at this levelcan do – by skill, by area
Englishexams
Frenchexams
Etc. ...
5
4
3
2
1
A system of criterion levels
89
Figure 2.4 Benchmarking the BEC suite to the CEF and ALTEframework
This process was made somewhat more complex due to the fact that the originalBEC tests were not benchmarked to individual levels within the ALTEframework. One design feature of the original BEC suite was that BEC1 (thelowest of the three levels) was created to straddle the Waystage and Thresholdlevels – accounting for at least some of the perceived difficulties with the test.Since the decision to address this represents one of the major changes to the BECsuite it will be dealt with in Chapter 3.
The ‘Can Do’ scales currently consist of approximately four hundred state-ments (translated into thirteen languages – Catalan, French, Portuguese, Danish,German, Spanish, Dutch, Greek, Swedish, English, Italian, Finnish, Norwe-gian) which are organised into three general areas (social and tourist, work andstudy). Obviously, for this validation project the work-related scales were used.Each of the three areas are further sub-divided into a series of more specificareas, each of which in turn includes up to three scales (listening/ speaking,reading, writing). Figure 2.5 is a graphical representation of the organisation ofthe ‘work-related’ statements. As can be seen in Figure 2.5, each of the threeareas has been sub-divided into a number of more specific situations; these areseen as being related to a particular aspect of the work environment and asdrawing from a range of language skills. In the example shown, only thelistening/speaking and writing language skills are identified as being requiredfor the meetings and seminars situation. The ‘Can Do’ statements for the work-related section are therefore built around each element of Figure 2.5, so therewill be statements at up to five levels related to the two language skills’ areasidentified here within the context of meetings and seminars. The reason that
C2 Mastery
C1Effective
Proficiency
B2 Vantage
B1 Threshold
A2 Waystage
A1 Breakthrough
ALTE Level 5Good User CPE
ALTE Level 4Competent User CAE
BECHigher
ALTE Level 3Independent User FCE
BECVantage
ALTE Level 2Threshold User PET
BECPreliminary
ALTE WaystageUser KET
ALTE BreakthroughLevelA
Com
mon
Eur
opea
n Fr
amew
ork
of R
efer
ence
Cou
ncil
of E
urop
e
C
B
A
2 The revision of BEC
90
there are ‘up to’ five levels is in recognition of the fact that in some instanceseven a relatively basic level of proficiency is sufficient to successfully deal witha situation.
Another advantage to the linking of each test in the BEC suite to a single scaleis related to the way in which we think about the reliability of the judgementsmade in score/grade awarding. While the area of reliability as it relates to theBEC tests (and potentially to all tests which are designed to work only at alimited range of ability) will be discussed in the following section when thequalities of test usefulness are examined, it is useful to make a connection at thispoint between the ALTE framework and the notion of reliability.
Figure 2.6 shows how results on one examination can be situated in relationto the much wider continuum of ability – so the proficiency level of a candidatewho achieves a Grade C for BEC Vantage can be seen beyond the specific test tothe whole range of proficiency as described in the ALTE/CEF framework. Inthis example a Grade C on BEC Vantage can be seen in terms of BEC Vantage(1), the BEC suite as a unit (2), and the whole range of ability as described by theCEF/ALTE frameworks (3). Reliability, therefore, becomes a matter of theaccuracy of level assignment within the overall continuum, and implies a verydifferent perspective on how evidence of this ‘reliability’ should be reported.
Defining the construct of business EnglishThe construct of business English as operationalised in the BEC suite of tests isbased on the clear specification of the concept from a number of perspectives:
• test taker• theory-based validity• context-based validity• scoring validity.
Figure 2.5 The ALTE ‘Can Do’ work-related statements
Social & Tourist
Work
Study
Work-related Services
Meetings & Seminars
Formal Presentations & Demonstrations
Correspondence
Reports
Publicly Available Information
Instructions & Guidelines
Telephone
Listening/Speaking
Reading
Writing
1
2
3
4
5
Statements at upto 5 levels
Defining the construct of business English
91
The test taker
Cambridge ESOL routinely collects information about test takers by askingthem to complete a Candidate Information Sheet (CIS). This is done primarily toensure that there are no tasks or items that result in uncharacteristically low testperformance from a particular sub-group of the population. Another reason forgathering this information is to better understand the population so that appro-priate tasks and items are included in the test.
The candidate information collected reflects two of the three groups of testtaker characteristics’ categories suggested by O’Sullivan (2000a), namely,physical and experiential characteristics – the third group of characteristics ispsychological, which is seen as more of a research issue related to test design,see for example O’Sullivan (2000a), who investigated among other variables,the effect on performance of candidate perceptions of the personality of peercandidates in the FCE test of speaking. By collecting data on the physicalcharacteristics of the candidates, validation officers can carry out bias studies (toensure that there is no gender bias for example, or no bias that may be related insome way to the age of the candidate), while developers can ensure that accom-modations are set in place which can allow students with special needs an equal
Figure 2.6 Viewing an estimate of attainment at one level in terms of alllevels
C2
C1
B2
B1
A2
A1
ALTE Level 5
ALTE Level 4 BEC Higher
ALTE Level 3 BEC Vantage
ALTE Level 2 BEC Preliminary
ALTE Level 1
Breakthrough
C
B
A
A
B
C
D
E1
2
3
2 The revision of BEC
92
opportunity to sit for their tests, see Gutteridge (2003) and Taylor andGutteridge (2003) for a description of the approach taken by Cambridge ESOL.
By knowing more about the background of each candidate, the test developercan also investigate the degree to which particular background variables mightimpact on test performance, a particularly relevant area of research in a specificpurpose test.
Theory-based validity
In their response to the criticisms voiced by Foot (1999), Saville and Hargreaves(1999) presented a model of communicative ability, grounded in the work ofBachman (1990) and upon which the UCLES Main Suite Speaking examina-tions are based. This model also forms the basis of the BEC suite tests. It takesaccount of the executive resources available to the candidate in terms of theircommunicative language ability and also the metacognitive strategies they willneed to deploy for effective communication in the spoken mode.
The model, see Figure 1.1, is itself based on the earlier models of Bachman(1990) and Canale and Swain (1980), as well as on the Council of Europespecifications for the Waystage and Threshold levels of competence (Savilleand Hargreaves 1999:46). Though this model deals adequately with thecognitive aspects of language as communication, or what Weir (2004) refers toas theory-based validity, it does not satisfactorily address the importance ofthe context of language use on performance (Weir’s context validity). Recentdevelopments in the socio-cognitive approach to defining language proficiencyfor testing purposes (Chalhoub-Deville 2003, McNamara 1996, O’Sullivan2000a, Weir 2004) stress the necessity of looking at both the context- andtheory-based validity of tests and the interaction between these. In other words,defining the construct involves at its core a description of the test taker (in whichtheory-based validity is embedded) in the context of a particular languagedomain as mirrored in a test. In order to complete the definition, some evidenceof the scoring validity of the test is required, so that decisions or inferences basedon test scores can be shown to share the same theoretical rationale as the otherelements of the construct.
In line with this socio-cognitive development, Cambridge ESOL defines theconstruct from these multiple-validity perspectives of which communicativelanguage ability is only one aspect. These are discussed briefly below and thenin more detail in Chapter 5.
Context-based validity
The handbooks for the BEC suite provide sets of specifications for the teststhat are freely available in the public domain. These specifications outline
Defining the construct of business English
93
the language demands of the tasks and items included in the test, while alsoidentifying the conditions under which the tests are administered. More detailedspecifications are prepared for use by test writers within Cambridge ESOL inorder to ensure the compatibility of different versions of the same test in termsof what is being tested, how it is to be tested, and how the tests are to be adminis-tered.
As can be seen from the earlier sections in this chapter, the language of theBEC suite of tests has been closely linked (or benchmarked) to the CommonEuropean Framework (CEF). This ensures that the content and levels of eachtest version can be seen in terms not only of the BEC suite, but also of the rangeof ability as defined by the CEF. In addition, Ball (2002) described how thewordlists for the three BEC levels were revised based on extensive corpus-basedresearch further grounding the context validity of these tests.
In addition to looking at the language of the tasks (input and expected output),the performance conditions are designed, as far as is practicable, to reflect thoseof the business language domain – both in terms of the physical replication of thedomain and of the replication of the conditions in which aspects of languageability which can be used to define the domain are potentially present (in thelinguistic responses of successful candidates). In order to complement theseareas, Cambridge ESOL also ensures that all tests are administered in asystematic and fair way according to pre-set guidelines. These guidelines –which again attempt to reflect the business domain where possible – add to thesituational authenticity of the test event, while setting the foundations for fairand reliable scoring and interpretation of scores.
Figure 2.7 Defining the construct
Source: based on Weir (2004)
Context-Based Validity
Scoring Validity
Test Taker
Theory-Based Validity
Response
2 The revision of BEC
94
Scoring validity
The final element of the definition relates to the transformation of the candidateresponse into a meaningful score. In the past, test developers (and users) weremost interested in the area of reliability in all its guises (stability, consistencyetc.). However, it is now believed that this represents just one aspect of whatWeir (2004) calls scoring validity.
While this area is discussed in greater detail in Chapter 5 of this book, it isuseful at this juncture, to look at the relevance of scoring validity in thedefinition of the construct.
Since scoring validity is concerned with all aspects of the score awardingprocedures, all decisions made here should reflect the developer’s view oflanguage ability and approach to testing. In the BEC suite, this means that themodel of language ability should be reflected in both theory- and context-basedvalidity evidences as well as in the scoring procedures. This can be shown for theBEC suite Speaking tests for example, by linking the tasks and the rating scale tothe Saville and Hargreaves model (Figure 2.8). Each element of the model isreflected initially in the expected response by a test taker to a particular speakingtask (context-based) and in the predicted language knowledge of the test taker(theory-based). The elements are then reflected in the rating scale used to makejudgements related to the actual response on the task. This triangulation is abasic requirement for meaningful scoring of any test event.
Figure 2.8 Linking the Model to the Rating Scale
In the very brief overview offered in this section, I have attempted to give thereader some idea of the complexity of construct definition. While suggesting amodel of language ability on which tests are based is an important element ofthis definition, on its own it is clearly not enough. The approach taken byCambridge ESOL described above marks an attempt to ensure that the constructis defined from the multiple perspectives suggested by Weir (2004) which aredescribed in more detail in the final chapter of this book.
Scale Pronunciation Grammar & Discourse InteractiveVocabulary Management Communication
Model Language Competence StrategicCompetence
An on-going programme of validation and test revision
95
An ongoing programme of validation and testrevisionThis section will focus on a number of issues central to the Cambridge ESOLtest development methodology. These are related to the qualities of testusefulness as identified in the Validity, Reliability, Impact and Practicality(VRIP) system.
Examination qualities VRIP
As mentioned by Saville (2003: 65) all Cambridge ESOL examinations are builtaround four ‘essential’ qualities:
• validity• reliability• impact• practicality.
These four qualities were abbreviated by Saville to VRIP, a convention I willalso follow here. Also similar to the approach of Weir (2004) will be my consid-eration of the four qualities as being of central importance to the overallusefulness of any test. I would argue, however, that the former pair, validity andreliability, are actually two aspects of the unitary concept of validity – and thatthis view may provide us with a more useful model of test development than amodel in which the pair are separated, but more of that later.
Before discussing the impact of VRIP on the BEC revision process, I will firstbriefly summarise the concept of VRIP as outlined by Saville (2003), andsummarised in Figure 2.9.
ValidityThe view of validity, as seen by Saville (2003), is best described as ‘mainstream’in that it propounds the by now widely supported ‘unitary’ model suggested byMessick, which sees multiple sources of evidence as adding different levels ofsupport to the central issue of validity. This view places construct-relatedvalidity at the core of validation. For this reason, it is considered imperative thata test should be based on a model of communicative language ability that can beempirically supported. According to Saville and Hargreaves (1999), the modelwhich drives the Cambridge ESOL test development and revision practice (seeFigure 1.1) has been influenced by the work of Bachman (1990) and the Councilof Europe, among others.
The rationale behind collecting evidence of content-related validity has to dowith the need to demonstrate ‘the degree to which the sample of items, tasks, orquestions on an examination are representative of a defined domain of content.
2 The revision of BEC
96
It is concerned with both relevance and coverage’ Saville (2003:67). This repre-sentativeness can be specified through a model (as the model suggested bySaville and Hargreaves 1999 is used to specify the Cambridge ESOL tests) inaddition to judgements made by experts in the field. Expert judgements may alsobe used when making decisions on the relative importance of various samples.
The relevance of content validation becomes apparent when we consider oneparticular feature of a test – a feature often associated with Cambridge ESOLMain Suite examinations and of particular relevance in the BEC series – that ofauthenticity. Weir (2002) argues that ‘the relationship between the “input” andthe expected response or “output” is an important feature of content validation’.He goes on to suggest that:
The examination content must be designed to provide sufficient evidence ofthe underlying abilities (i.e. construct) through the way the test takerresponds to this input. The responses to the test input (tasks, items, etc.)occur as a result of an interaction between the test taker and the test content.The authenticity of test content and the authenticity of the candidate’s inter-action with that content are important considerations for the examinationdeveloper in achieving high validity (ibid).
This can be seen as offering evidence in support of Messick’s view of validity, inthat it is difficult, if not impossible, to draw a clear distinction between theconcepts of construct- and content-related evidence of a test’s validity.
More evidence in support of Messick can be found in the way in which theBEC series (in the same way as the other Cambridge ESOL examinations) are
Figure 2.9 The four qualities of test usefulness
PRACTICALITY
RELIABILITYVALIDITY IMPACT
Evidence
Construct-related
Content-related
Criterion-related
Evidence
Measurement
Framework-related
Quality-related
Evidence
Anecdotal
Observational
Varied Sources
An on-going programme of validation and test revision
97
benchmarked to an external criterion – in this case the ALTE framework. Thisbenchmarking is an important aspect of the design of the tests, and also hasimplications for test content. We can therefore see that all three aspects ofvalidity are interlinked to form a ‘unitary’ conceptualisation of validity.
ReliabilityThe view of reliability within the Cambridge ESOL test development frame-work is that it ‘concerns the extent to which test results are stable, consistent,and free from bias and random error’ (Saville 2003:69). The need to developinstruments that conform to this view is, of course, paramount. However, thefact that no practical consideration of how reliability decisions impact on a testcan be made without also considering the implications that these decisionsmight have on the validity of the inferences we can draw from performances onthat test means that there is a limit to the lengths to which it is possible to go inorder to achieve maximum reliability. This last statement, apparently obviousthough it is, actually highlights a real concern with the way in which we estimatethe reliability of our tests.
Problems with the existing measuresThe most critical error in the perception of reliability of many test developersand test users is the assumption that estimates of internal consistency that arebased on item variance are measures of test reliability. I would argue that theseestimates are particularly useful for certain types of test (e.g. multi-itemstandardised tests where there is clear evidence that the items are deliberatelychosen because they test a single construct) but are not suitable for a criterion-referenced test, particularly where there is a truncated test population (i.e. alimited range of proficiency is represented in the population).
The attenuation paradox, first identified by Loevinger (1954), identifies acritical deficiency in the way we measure reliability. While writers such asBrown (1996:192) and Hughes (1989:31) suggest that 1.0 represents a ‘perfect’reliability coefficient, the attenuation paradox means that, for a test to achievethis ‘perfection’ the only possible response patterns are a perfect full score or aperfect zero score. So the data set represented in Table 2.1, will result in a‘perfectly’ reliable test (i.e. it will have a reliability coefficient of 1.0).
Table 2.1 Example of the response patterns in a ‘perfectly’ reliable test
CandIDs
Item 1 Item 2 Item 3 Item 4 Item 5 Item 6 Item 7 Item 8 Item 9 Item 10
1234
1010
1010
1010
1010
1010
1010
1010
1010
1010
1010
2 The revision of BEC
98
The other difficulty with the way in which we estimate reliability lies in the factthat it is not even necessary that the response patterns should be neatly dividedinto two equal groups of candidates. In fact, if only a single candidate achieves aperfect full score while all others score zero, the test will still appear to have a‘perfect’ reliability.
One problem with this feature of how ‘reliability’ has come to be seen is thatit is very much dependent on the test population. Where a population containsexamples of extreme behaviour the likelihood is that the ‘reliability’ estimatewill be high. The implications for any test that is benchmarked to a particularlevel of performance, as is the case with the BEC suite (as well as the CambridgeESOL Main Suite examination and the Pitman and LCCIEB tests referred to inChapter 1), is quite clear. Where a test is drawing on a truncated population, inthat the ability range of the test takers is confined to a relatively narrow range ofability, the estimates of internal consistency will always be low. This type ofmeasure is therefore unsuitable for analysing the type of tests referred to here,that is level-based tests (though it may be used as a practical measure ofcomparing the internal consistency of different administrations of the same testwhich have been proven to have similar truncated candidate populations). Toput it another way, and perhaps more accurately, it is not reasonable to expectthat these tests will result in the very high measures of variance-based internalconsistency that can be achieved by tests which test across a wide range ofability.
The real difficulty lies in the fact that we have come to accept that estimatesbased on internal consistency (KR20, Chronbach’s alpha) are accurateindicators of the reliability of a test. They are not.
Saville also argues that ‘in the case of the Cambridge ESOL examinations,which employ a wide variety of task-based materials and item types . . . veryhigh internal consistency may not be an appropriate aim’ (2003:70). He goes onto suggest that the replacement of discrete point multiple-choice items by task-based exercises (which provide far greater context and authenticity, both situa-tional and interactional) means a reduction in the number of items and also of theestimated reliability using an internal consistency estimate.
Cronbach’s alpha does not divide the test according to tasks, but items, sothat both halves of the test may contain items from one task. Items from onetask are not independent of each other to the same degree as discrete items.That is to say, if a candidate has correctly answered the first item of a multi-item task they are more likely to answer the next item correctly because oftheir response to the first item. In this case Cronbach’s alpha would exaggeratethe reliability of such a test in much the same way as if the candidate’sresponse to the same item was placed in both halves of the split test (Anastasi1988).
A solution to the above would appear to be if the internal consistency of a test
An on-going programme of validation and test revision
99
is calculated by splitting the test according to task, not item. However, as we canexpect candidates to perform differently according to task type, the reliabilitycoefficient calculated in this way would be lower than for a discrete item testeven if the discrete item test contained a variety of (single item) tasks and istherefore not, strictly, comparable.
The problem is compounded for tests such as BEC. Not only is the testpopulation truncated, but also the task types tend to result in reliability measure-ments that are not comparable to those values calculated on a fully discrete itemtest. What is clear from the above is that these estimates do not tell the wholestory about a test’s reliability.
However, internal measurements of reliability, such as Cronbach’s alpha, areuseful in the test development process in providing convenient conventionalisedmeasurements of reliability between different parallel forms of the same test.This occurs in different administrations of the same test at different sessionsthroughout the year. Here the error noted above in what may be termed the‘absolute’ reliability is not as important as the insight the measurements give inmaintaining standards across different administrations.
Estimates have been systematically calculated for BEC suite tests over theyears. Based on the information contained in Tables 2.2–2.4, it is possible tomake reference to the kind of cross-administration comparisons mentionedabove. It is possible to see, for example, that the internal consistency of theReading and Listening papers varies very little over the different sessionsreported in a 2-year period. It is interesting to note also that the estimates arehigh enough to be considered acceptable in a norm referenced test for apopulation where there is a full range of ability.
Table 2.2 Reliability (Cronbach’s alpha) for BEC Preliminary, Readingand Listening components, selected sessions 2002–2003
Table 2.3 Reliability (Cronbach’s alpha) for BEC Vantage, Reading andListening components, selected sessions 2002–2003
Session Reliability Reading Reliability Listening Sample Size
May–02Nov–02May–03Dec–03
0.850.850.860.86
0.840.820.870.83
108790510641873
Session Reliability Reading Reliability Listening Sample Size
May–02Nov–02May–03Nov–03
0.820.860.840.85
0.830.780.800.81
14587541084998
2 The revision of BEC
100
Table 2.4 Reliability (Cronbach’s alpha) for BEC Higher, Reading andListening components, selected sessions 2002–2003
As with the Cambridge ESOL Main Suite examinations, reliability of the BECsuite tests of Speaking and Writing should be seen from the perspectives of theaccuracy and consistency of the ratings which are awarded, as dictated by thecurrent American Psychological Association Standards (APA 1999).
Cambridge ESOL tests have included the pair format as the standard formatfor all Main Suite speaking papers since the early 1990s and all BEC levels arebased on the format (with two candidates and two examiners, one in the role ofinterlocutor/assessor and the other in the role of observer/assessor). Since bothexaminers use slightly different scales (the interlocutor uses a holistic scalewhich is derived from the four criteria analytic scale used by the observer), thereare some problems with any simple correlations between the scores they award.However, there are similar difficulties with any correlation procedure, as theoutcomes are affected by the nature of the scale used and by the range of abilityof the test population
In addition to calculating the correlation of scores awarded by raters, it mayalso be fruitful to compare the grades each individual examiner’s score mightlead to – in other words, an examination of classification accuracy. Multi-faceted Rasch (MFR) analysis has been suggested as a possible solution tothe inter-rater reliability problem. This process produces output tables for thedifferent variables (or facets) included in the analysis. Each output tableincludes a ‘separation reliability’ estimate. Where the output table for candi-dates is concerned we would hope that the separation reliability is high(indicating that the candidates have a range of significantly different ability). Asfar as raters are concerned, we want the separation reliability to be low(indicating that they have the same severity).
Since Rasch is a probabilistic model, the expectation is that the raters will belocally independent (they will demonstrate some amount of disagreement).This suggests that MFR offers an interesting solution, though the differentscales will represent a practical (though not insurmountable) concern indesigning a study. The other difficulty with MFR is that it is ideal for experi-mental studies, whereas little work has been done to date in expanding the method into a large-scale ‘real world’ test, mainly due to the problem ofestablishing connectivity issues among the raters – though initial groundworkhas been undertaken by Myford and Wolfe (2000) in their study of minimal
Session Reliability Reading Reliability Listening Sample Size
Mar–02Nov–02May–03Nov–03
0.850.850.810.85
0.850.860.780.80
511271581359
An on-going programme of validation and test revision
101
connectivity requirements for a large-scale test administration. Weir (2004) alsosuggests that generalisability theory may offer another direction of exploration.It would appear that for any performance test (of writing or speaking) it wouldbe safer to report reliability from a number of perspectives rather than rely onany single estimate.
The above discussion essentially argues that the notion of ‘reliability’ as itexists is not useful for the type of tests I am writing about here, except as aconvenient and conventionalised means of comparing similar tests with similartruncated populations (e.g. in different administrations of the same test) as notedby Saville (2003:71). Instead, it would be more beneficial to see the true relia-bility of a test as being centred on the degree to which, to repeat Saville ‘theresults are stable, consistent, and free from bias and random error’ (2003:69).This definition essentially brings us back to the perspective suggested byBachman who sees reliability as being associated with ‘sources of error in agiven measure of communicative language ability’ (1990:160). The sources ofthreat to reliability are suggested in Figure 2.10.
Bachman (1990) argues that test method facets should be seen as beingrelated to the testing environment, the test rubric, input and expected response,and the relationship between input and response (1990:118–152) and as being‘systematic to the extent that they are uniform from one test administration toanother’ (1990:164). This notion of systematicity is also applied to the definition
Figure 2.10 Factors that affect language test scores
Source: Bachman (1990:165)
Communicativelanguage ability
Randomfactors
Personalattributes
Test methodfacets
TESTSCORE
2 The revision of BEC
102
of ‘personal attributes’ (1990:164); random factors are seen by Bachman(1990:164–5) as being unsystematic variables associated with:
• the candidates (such as mental alertness, emotional state) • the test facets (such as changes in test performance conditions)• the test administrators (‘idiosyncratic differences in the way different test
administrators carry out their responsibilities’)• incomplete language sample• scale imprecision.
When this view of reliability is considered, we can really see the limitationsin the ‘reliability as internal consistency’ perspective that currently dominateslanguage testing – certainly if we are to take what we read in journal articles andtest reports as reflecting current practice. We can also include among these‘random’ factors sources of variance implied in the Milanovic and Saville(1996) framework, see Figure 2.11. When we consider the likelihood that thevariables included in this framework are potential sources of systematic and/orunsystematic or random variance, we get some notion of the difficultiesinvolved in establishing the conditions for truly reliable testing to take place,and of the necessity of seeing true reliability as being a function of what I wouldcall test quality.
Figure 2.11 A conceptual framework for performance testing
Source: Milanovic and Saville (1996)
EXAMINATIONDEVELOPER
CANDIDATES
EXAMINERS
KNOWLEDGEAND ABILITY
SCORE
KNOWLEDGEAND ABILITY
SPECIFICATIONSAND
CONSTRUCTASSESSMENT
CRITERIA
TASKS
ASSESSMENTCONDITIONS
AND TRAINING
EXAMINATIONCONDITIONS
SAMPLE OFLANGUAGE
An on-going programme of validation and test revision
103
This then raises the question of how we demonstrate true reliability. I wouldargue that the true reliability of any test is a unitary concept, much like the waywe look at test validity, but with multiple perspectives, again much like withvalidity. Therefore, to demonstrate it we need to provide evidence of test qualityacross a whole range of perspectives, and as far as the BEC suite is concerned,one major source of evidence is the way in which the overall approach to testdevelopment, construction and administration outlined above is applied to thesuite.
ImpactSaville suggests that:
‘From a validation perspective, it is important to be able to monitor andinvestigate the educational impact that examinations have within thecontexts they are used. As a point of principle, examination developers likeCambridge ESOL should operate with the aim that their examinations willnot have a negative impact and, as far as possible, strive to achieve positiveimpact’ (2003:74).
He identifies the following issues as central to any test organisation’s validationprocedures, this in reference to an a priori perspective on Messick’s notion ofConsequential Validity:
• the development and presentation of examination specifications anddetailed syllabus designs;
• provision of professional support programmes for institutions andindividual teachers/students who use the examinations
• the identification of suitable experts within the field to work on allaspects of examination development
• the training and employment of suitable experts within the field to act asquestion/item writers in examination production
• the training and employment of suitable experts within the field to act asexaminers’ (op. cit.).
Within the context of the BEC revision, these issues have been approached in anumber of ways (Table 2.5).
In addition to the above issues, Saville argues for a similar concern with an aposteriori perspective on test impact when he suggests that procedures also needto be put into place after an examination becomes operational to collect infor-mation which allows impact to be estimated. This should involve collecting dataon the following:
• who is taking the examination (i.e. a profile of the candidates)• who is using the examination results and for what purpose• who is teaching towards the examination and under what circumstances• what kinds of courses and materials are being designed and used to prepare
candidates
2 The revision of BEC
104
• what effect the examination has on public perceptions generally (e.g.regarding educational standards)
• how the examination is viewed by those directly involved in educationalprocesses (e.g. by students, examination takers, teachers, parents, etc.)
• how the examination is viewed by members of society outside education(e.g. by politicians, businessmen etc.) (Saville 2003:75).
These issues have been addressed in the BEC suite examinations as outlinedin Table 2.6.
PracticalityThough its importance is often neglected in the language testing literature,practicality is ‘an integral part of the concept of test usefulness and affects manydifferent aspects of an examination’ (Saville 2003:76).
The section above, in which the project structure was described, is particu-larly important here as we can see that it includes many of the aspects of practi-cality suggested by Saville as being of relevance to any test:
• ‘the management structure for the development project• a clear and integrated assignment of roles and responsibilities• a means of monitoring progress in terms of development schedules and
resources• a methodology for managing the examination production process when
Issue Action
development and presentation ofexamination specifications anddetailed syllabus designs
through the dissemination of information through theBEC website (www.cambridge-efl.org/exam/business/bg_bec.htm) and the latest BEChandbooks (downloadable from the website)
provision of professional supportprogrammes for institutions andindividual teachers/students who usethe examinations
through the professional seminar programme
identification of suitable expertswithin the field to work on all aspectsof examination development
through the appointment of leading researchers andacademics to act as consultants in all aspects of therevision process
training and employment of suitableexperts within the field to act asquestion/item writers in examinationproduction
through the provision of detailed training manuals (e.g.CAMBRIDGE ESOL’s involvement in the ALTE ItemWriters Guidelines’ Project ) and the recognition ofexpertise within the organisation
training and employment of suitableexperts within the field to act asexaminers
through detailed Minimum Professional Requirements’(MPR) documents, and the setting of rigorous selectionand accreditation standards
Table 2.5 Impact issues in the BEC tests (a priori)
An on-going programme of validation and test revision
105
the examination becomes operational (item writing, vetting, moderation,pre-testing, item banking, question paper construction)’ (Saville2003:77).
For the BEC examinations (as with other Cambridge ESOL tests), practicality isa major concern, impacting on a whole range of areas of test production, admin-istration and evaluation. The areas concerned have been identified in Weir andMilanovic’s work on the CPE revision, though clearly the list is equally valid forthe BEC examination:
• design features related to format and content of the four skills approache.g. length of papers – number of items and time allowed, type of tasks
• test production featurese.g. number of items required, replicability of tasks
• availability of the examination in terms of: examinations dates and the frequency of administrationlocation and number of centres
• level of fees to be paid by test takers• central costs in terms of:
production of question papersmarking and scoringvalidation
• local costs in relation to administration at centrese.g. hire of venues, training and deployment of oral examiners, etc.
Issue Action
who is taking the examination data related to the candidates isroutinely collected through theCandidate Information Sheet (CIS)and used in test, task and item levelanalyses
who is using the examination results and for whatpurpose
These are monitored through routinesurveys of stakeholders (bothformally and informally). The datacollected are used in all majordecisions regarding the tests inquestion – particularly in makingdecisions related to review andrevision.In the BEC revision two revisionquestionnaires were developed:• general (primarily aimed at
teachers)• key contacts (for principal
stakeholders).
who is teaching towards the examination and under whatcircumstance
what kinds of courses and materials are being designedand used to prepare candidates
what effect the examination has on public perceptionsgenerally
how the examination is viewed by those directlyinvolved in educational processes
how the examination is viewed by members of societyoutside education
Table 2.6 Impact issues in the BEC tests (a posteriori)
2 The revision of BEC
106
• central administration entry procedures – exchange of data with centrescollection of feesdespatch of materialsmarking and grading proceduresissue of results
• local administration at centres• security• special circumstances
e.g. arrangements for candidates with special needs.
Recruitment, Induction, Training, Co-ordination, Monitoring,Evaluation (RITCME)One aspect of ‘test usefulness’ that has an impact on a number of the above areasof concern is that of the structure and maintenance of what is known as the TeamLeader system for examiners (this is particularly relevant to the examiners whoare involved with the assessment of the productive skills).
As the reader can imagine, the logistics of administering a test on an interna-tional level are beyond the experience of most organisations, let aloneindividuals. In tests of speaking, for example, there is evidence of an ‘inter-locutor effect’ (Lumley and O’Sullivan 2001, O’Sullivan 1995, 2000a, 2000b,2002a), an ‘observer/assessor effect’ (McNamara and Lumley, 1997 – wherethe assessors were apparently systematic in taking into account what theyperceived as the adequacy of the performance of the examiner in awarding ascore to the individual candidates), and a ‘candidate by task effect’, for exampleBerry (1997) demonstrated how candidates of different psychological make upperformed more or less well depending on the task. While the latter effect ismore related to test design (suggesting that any performance test would benefitfrom a variety of tasks), the former pair of effects highlight the need for thecareful recruitment, induction, training, co-ordination, monitoring, and evalu-ation of all examiners.
As with the other Cambridge ESOL examinations (in which over 10,000examiners participate on a regular basis worldwide), there are a number oflevels of professional responsibility within the BEC examiner system, inaddition to the Cambridge ESOL staff. These levels are summarised in Figure2.12.
At the operational level are the Examiners (both oral and written). Incountries where there are sufficient numbers of examiners to merit it, there areTeam Leaders who have responsibility for the professional supervision ofexaminers. Team Leaders typically work with anywhere from 5–30 examiners.Where there is an ample number of Team Leaders in a country, they will besupervised by a Senior Team Leader, the average ratio being 15:1. It should benoted that all of the above are actually practising examiners, so while there is a
hierarchical structure, its principal rationale is to ensure that there is a clear two-way channel of communication through the test administration system.
The set of procedures which regulates the activities of these three profes-sional levels is summarised by the acronym RITCME – Recruitment, Induction,Training, Co-ordination, Monitoring, Evaluation. Each procedure is defined bya list of minimum professional requirements, which sets down the minimumlevels and standards (for recruitment, induction, training programmes, etc.) thatmust be achieved in order to meet the professional requirements of adminis-tering the Speaking tests and sustain a fully effective Team Leader System(Taylor 2000) .
The great advantage to this set of guidelines is that it allows local exami-nation secretaries outside of the UK to ensure that their practices mirror that ofthe UK-based parent group. Cambridge ESOL itself ‘has the primary responsi-bility for the supervision and deployment’ of examiners in the UK (UCLES1999:1) .
Ensuring accuracy and consistency of gradesAn important aspect of the Cambridge ESOL approach is the concern withensuring that the final grades awarded to candidates are a consistent, accurateand a fair reflection of the levels defined in the ALTE/CEF frameworks and thatscores and grades reflect a consistent language ability over time.
A number of key areas, related to the work of the research and validation unitwithin Cambridge ESOL, are briefly discussed below but the process of
Ensuring accuracy and consistency of grades
107
Figure 2.12 The Team Leader system
Team Leader Team Leader Team Leader
Examiner
Examiner
Examiner
Examiner
Examiner
Examiner
Examiner
Examiner
Examiner
Examiner
Examiner
Examiner
Examiner
Examiner
Examiner
Senior Team Leader
ensuring construct and content validity permeates the test construction processat each stage. For example, items are commissioned from experienced ESOLprofessionals who receive ongoing training. Similarly, at the pre-editing stageitems are examined by experts for any obvious biased or culturally inappropriatematerial.
Pretesting and test construction
The construction of tests to specified content and difficulty targets is achieved bythe item-banking process employed for BEC as for other Cambridge ESOLexams. Reading and Listening items are pretested under exam conditions, on asuitably large and diverse group of learners at an appropriate level. Whereverpossible these are BEC candidates who will be taking an exam in the near future.Pretests include so-called anchor items. These are items of known difficultytaken from previous live administrations that have been selected for theiradequate facility, discrimination and difficulty and perceived lack of biastowards or against one or more groups of test takers.
Writing and Speaking tasks are trialled on a smaller but representative groupof candidates to ensure that they elicit the desired responses and that the tasks donot contain lexical items or phrases within the instructions that would beproblematic to candidates at that level.
When the response data are analysed the anchor enables the difficulties of thepretested tasks to be located on the measurement scale which underlies theCambridge/ALTE levels’ system.
Item banking
The calibrated tasks are then stored in a sophisticated item bank: LIBS (LocalItem Banking System). This is a computer-based management and analysis tooldeveloped by UCLES, not only to store calibrated items but to handle the entireproduction process. LIBS contains a large bank of materials for use in theexaminations, which have been fully edited and pretested according to theprocedures described in some detail by Saville (2003:90–95). LIBS enablescomplete test versions to be constructed to quite precise targets in terms ofcontent and difficulty. However, versions may still vary slightly in meandifficulty, because most items are embedded within tasks and thus cannotindividually be juggled to achieve an exact mean difficulty.
Item banking exploits latent trait (item response theory) techniques. Theparticular latent trait model used by Cambridge ESOL is the Rasch model,which has proved to be well-suited to the construction of a broad measurement framework capable of accommodating a suite of communicative language
2 The revision of BEC
108
proficiency exams at different levels and covering a range of skills.
Administration
The conditions under which the exam is administered are also important forensuring that all candidates are given an equal chance to perform to their best.The assessment of the Speaking component is particularly important in thisregard, and depends on the professional skills of the oral examiners, so that thetraining, standardisation and monitoring of this cadre can be seen as a vitalelement in the achievement of a common standard.
Marking, scaling, weighting
Following the administration and before grading can take place, candidates’responses must be captured and marked.
Three kinds of marking are employed:
1. Automatic, in the case of Reading and Listening multiple-choice tasks,where the candidate indicates a choice on an optically markable answersheet.
2. Clerical, in the case of short free-text responses which allow a strictlylimited set of correct answers, and which can be marked to a high degree ofreliability by clerical staff under the supervision of a co-ordinating examinerwho has analysed live responses to ensure that the key is complete andunderstood by clerical markers.
3. Examiner marking, as in the case of Writing, where trained and standardisedexaminers apply a mark scheme and their knowledge of the level to assign amark.
The next step is scaling, where the distribution of marks for the Writingcomponent is adjusted to compensate for differences in the marking patterns ofWriting examiners. Scaling is designed to ensure that markers who are morelenient or severe compared to all other markers have their individual marks forcandidates adjusted to compensate for these tendencies. In scaling the distri-bution of marks for all candidates in writing is compared to the distribution ofmarks for candidates of a particular marker. Adjustments are made to the marksof candidates at a number of points on the markers’ candidate writing scoredistribution to bring this in line with all candidates’ writing score distribution.Allowances are made for the difference in mean and standard deviation of themarkers and all candidates as observed in the Reading and Listening compo-nents.
Next marks for each component are weighted. For BEC as for otherCambridge ESOL exams the general principle is adopted that each component
Ensuring accuracy and consistency of grades
109
should contribute an equal proportion of marks to the total available for theexam. Thus in the case of BEC each of the four components contributes 30marks to the exam total of 120. In fact the number of raw marks in the Readingpaper is greater than 30. A candidate’s Reading mark is thus weighted by multi-plying it by 30/45 (for Preliminary and Vantage) or 30/52 (for Higher).
A candidate’s total mark in the exam is a simple sum of the marks gained ineach component, after the processes described above.
Grading
A passing grade at BEC locates a candidate within a broad proficiencyframework: Passes at BEC Preliminary, Vantage and Higher correspond toALTE Levels 2, 3 and 4, or to Council of Europe Levels B1, B2 and C1 respec-tively. Thus BEC can be broadly compared with other exams at the same level.
For such interpretations to be valid it is of course necessary that BEC begraded to a consistent standard. Every stage of the exam administration cycle,from test design through exam conduct to marking and grading, is relevant toensuring this consistency.
The grade thresholds should reflect a constant standard across sessions, butthe precise number of marks needed to achieve each grade will vary within anarrow range, reflecting a judgement about the difficulty of the components in aparticular session. This judgement is based on several types of quantitative andqualitative information:
1. The difficulty of each objective component. The estimate of this depends onthe calibration of items at pretesting and another, independent, calibrationgiven by live anchor tests. These are short tests of items with knowndifficulty and suitable facility and discrimination levels, administered to aproportion of candidates at the same time or shortly before the adminis-tration of the exam itself. In BEC all candidates are requested to completeanchor tests. This ensures that a representative sample of candidates sits theanchor tests and that the estimate of difficulty for items and componentscalculated using anchor tests can be checked for the effect of first languageand ability on performance in the anchor tests. The difficulty of Reading andListening components are arrived at then by examining and weighing three,independent, sources; pretest statistics, live anchor statistics, and compar-isons with performance of live candidates in the criteria-based componentsSpeaking and Writing.
2. The performance of particular ‘cohorts’ i.e. major groups of candidates,compared with their historical trends. While it is clear that cohorts mayfollow upward or downward trends, reflecting changes in the size or make-
2 The revision of BEC
110
up of the group, it is expected that these trends will be steady, and thatgrading should not result in abrupt shifts in the pass rate for many cohorts.
3. A judgement about the standard applied to the performance components(Speaking and Writing). The mark schemes for these are criterion refer-enced, and thus the marks awarded should directly reflect the standard.Mean scores at task level and feedback from chief examiners are noted toindicate whether individual tasks in writing have proved difficult. If so somecorrection will be made for this in the grading.
Estimating the internal consistency of eachadministrationWhile acknowledging the potential difficulties associated with reportinginternal consistency estimates for each BEC examination, Cambridge ESOLreports the estimates for each BEC level for Reading and Listening (see Tables2.2 to 2.4). The overall consistency of an exam such as BEC, i.e. it is comprisedof several component papers, is known as its composite reliability.
Using the Feldt and Brennan (1989) approach, the composite reliability forthe BEC suite tests has been reported by Cambridge ESOL as lying in the rangeof 0.88 to 0.91 for all sessions in 2003 (see Table 2.7). When viewed with all ofthe other procedures that are in place to ensure the accuracy of the final gradesawarded to candidates, these figures can be seen as adding significantly to theoverall reliability evidence.
Estimating internal consistency
111
Table 2.7 Composite reliability and SEM for all BEC Levels 2003
*R= Reliability**SEM= Standard Error of Measurement*** No administration
Level Session Mar 03 May 03 June 03 July 03 Nov 03 Dec 03
BEC P Comp R*Comp SEM**Sample Size
0.904.892008
0.905.331064
0.904.991956
0.905.30885
0.895.06982
0.905.281873
BEC V Comp RComp SEMSample Size
0.904.921819
0.894.841084
0.894.892926
0.914.31791
0.904.85998
0.894.611760
BEC H Comp RComp SEMSample Size
0.903.83482
0.884.08581
0.883.941565
*** 0.894.20359
0.894.01952
Additional proceduresSeveral additional procedures are followed to ensure that as far as possiblecandidates’ final grades reflect their true ability.
Grade review is a process which follows grading in which the Writing scriptsof candidates just below the passing grade are reviewed, and if necessary theWriting mark is amended.
Examination centres may ask for ‘special consideration’ on behalf of candi-dates because of personal circumstances surrounding the exam, or because theybelieve the administration of the exam was such as to disadvantage them – forexample, that the Listening component was disrupted by noise outside the examroom. Where appropriate these cases are evaluated by a panel which followsguidelines and is informed by relevant statistical information on theperformance of the candidates. This statistical information examines thediscrepancy between the marks achieved in the administration of the componentunder question and the performance of the candidates in their other components(allowing for differences in mean score and standard deviation between compo-nents) to observe if there is a significant difference between the two.
Cases of alleged malpractice are also investigated, and where proven resultsare withheld.
All the procedures outlined above are not specific to BEC but rather arestandard practice across Cambridge ESOL examinations, and are dealt with inmore detail in Weir and Milanovic (2003:88–109).
The context for the revision of BECSince the early 1990s there have been major documented revisions to a numberof the Cambridge ESOL Main Suite examinations, most notably the Certificateof Proficiency in English (CPE), see Weir (2003a). In this excellent overview ofthe (CPE) revision project, Weir outlines in great detail the development cycledevised by the organisation for these revision/development projects thatinforms the Cambridge ESOL language testing systems. In his review, Weirlooks not only at the latest revision to the CPE and the methodology thatsupported that process, but demonstrates the connections between the revisionpractice and its outcomes over the long history of the CPE (it was first adminis-tered in 1913). While such a historical perspective is clearly beyond this book,after all the testing of language for business purposes is a very newphenomenon, as can be seen in the previous chapter, it is worthwhile revisitingthe revision methodology in order to better understand how the present BECsuite has come to reflect the Cambridge ESOL language testing philosophy bothof test quality and content.
The development cycle has been outlined in detail by Saville (2003) and will
2 The revision of BEC
112
be summarised in the remainder of this chapter only in relation to the impact ithas had on the BEC revision project.
The revision processThe Planning and Design phases for the BEC revision project took placebetween 1999 and 2000 and included the production of preliminary revisedspecifications, consultation with stakeholder groups as well as some experi-mental trialling – though the lessons learned from the recent FCE and CPErevision projects were also influential.
The development phase, which included trialling and analysis of proposedchanges took place in 2000. The specifications for internal use were approved in2000 and a revised specifications’ booklet was published in early 2001.
Groups involved in the revision process
The review/revision process was, as in other Cambridge ESOL revisionprojects, initiated within a project review group at one of its regular meetings.For a very detailed and informative overview of how this process was applied ina formal revision project, see the chapters by Ashton (Chapter 3); Weighill andShaw (Chapter 4); Barratt (Chapter 5); Boroughs (Chapter 6) and ffrench(Chapter 7) in Weir and Milanovic (2003). The rationale for the BEC revisionwas less related to actual or perceived dissatisfaction with the tests, but to anawareness of the various changes (detailed above) which resulted in anexpanded and more culturally diverse candidature.
As with the CPE revision, the first stage was to set up the necessarymanagement structures to oversee the review or revision project. This meant thecreation of a Management Steering Group chaired by the Director or DeputyDirector EFL and consisting of Cambridge ESOL senior management (e.g.group managers and the project co-ordinator). This group was empowered withthe oversight of the whole process and the management of resource allocation.Among the specific duties of the group were:
• to define parameters• to initiate research and development• to make judgements• to ratify the revised specification• to allocate appropriate level of staff time for co-ordination of the project and
participation in an Internal Working Group• to create a number of Consultants’ Working Groups (one group per skill
area, each group was headed by a member of the internal working group).
The revision process
113
The internal working groups were made up of Cambridge ESOL specialist staff(including research and validation staff).
These groups were asked to:
• co-ordinate external groups• act on recommendations from the steering group• trial revised specifications• develop and propose final specifications to the steering group• report on the revision project to the steering group.
Finally, the Consultant Working Groups, which consisted of Cambridge ESOLconsultants (typically senior researchers and academics at key British univer-sities), specialist internal staff and research and validation staff were chargedwith devising revised specifications for each component of the tests.
The plan for the revision project was similar to that designed for the CPEproject, in that it focused on the main areas shown in Table 2.8.
Table 2.8 BEC revision project plan
While many of these areas are of little interest to the reader in that they arerelatively mundane and an ‘everyday’ part of any test development project, thefirst area is of particular interest in the BEC revision process as it was clearlyinfluential in all later decisions. It is therefore to this aspect of the process that wenow turn.
The consultative exercises
The internal working group first established a Project Plan, starting with a
2 The revision of BEC
114
Number Item Focus
12
Consultation exerciseIdentify areas for revision
Consultation exercisesIdentify priority areas for revision
34
Redraft specifications and item writer guidelinesTrialling
Draft specificationsTrialling
56
WordlistsValidation work
Validation projects
7891011
Information seminarsPublication of revised specificationsItem writer guidelines and trainingSample materialsRelease information to the public and centres
Finalised specificationsRelease of information
1213
Live test material productionTraining of oral examiners Training
14 Live administration
Situational Analysis. The key aim of this phase of the project was to establish aproject timeline with an anticipated end point.
The situational analysis began with a review of validation evidence, whichhad been routinely gathered since the earliest administration of the BEC tests.This evidence was related to the qualities of test usefulness identified in theCambridge ESOL approach, and discussed in the relevant section above (i.e.validity, reliability, impact and practicality).
In order to gain insights into the impact of the existing BEC tests, a survey ofthe views and attitudes of major stakeholders was conducted. There also existedan amount of formal and informal feedback collected over the years, though thiswas not considered sufficient on its own to allow for major decisions to be made.The stakeholders consulted included:
• Local Secretaries who administer the exam• the language schools/teachers preparing candidates• the senior consultants and other professionals who are employed to work on
the materials and assessment procedures (Senior Team Leaders, TeamLeaders, chairs of item writing teams, Principal Examiners etc.).
Two main groups were surveyed using a pair of questionnaires designed by theworking groups for the project. These groups were BEC centres (the question-naires were expected to be completed with teachers involved in preparing candi-dates for the existing BEC tests) and people with detailed knowledge of the BECtests (who were also experts in language testing). These two questionnaires areoutlined briefly below, while the results are reported in the relevant chaptersrelated to the different skills.
One of the questionnaires was called the Key Contacts Questionnaire (KCQ),and was designed to elicit information from major stakeholders around theworld, including local secretaries of major markets. The KCQ consisted of atotal of 60 statements in a series of sub-sections which looked at overall orgeneral comments as well as at the papers within each level (see Table 2.9).
All items offered a 5-point scale ranging from ‘strongly agree’ to ‘stronglydisagree’, with an additional ‘no knowledge’ option.
A total of 40 KCQs were distributed in early 1999 of which 21 were returned.In the resulting interim report, presented in May 1999, a number of points werehighlighted as being problematic. These were:
• the speaking paper in general• the reporting of the speaking paper (as a separate mark)• the level of BEC1• one section of the BEC1 Reading and Writing (Part 5).
At the same time as the KCQ, a second instrument, called the GeneralQuestionnaire (GQ), was distributed to 300 test centres around the world, ofwhich a total of 67 responded. The GQ consisted of a set of seven items related
The revision process
115
to BEC1, seven more for BEC2, eight for BEC3 and three items devoted tooverall impressions of the suite. The results of the GQ are summarised in Table2.10.
As can be seen from this table, there was a clearly positive feeling for the teststhough there was a suggestion that the Speaking paper at each level (with thepossible exception of BEC3) was less than satisfactory – with over 13%expressing a high degree of dissatisfaction with the paper in BEC1 and BEC2(just short of 8% expressed a similar level of dissatisfaction with the speakingpaper in BEC3).
Table 2.10 General Questionnaire – results summary
2 The revision of BEC
116
Test Paper % Satisfaction
BEC1 ReadingWritingListeningSpeakingGeneral
95% positive95%90%59%86%
BEC2 ReadingWritingListeningSpeakingGeneral
92%85%89%65%85%
BEC3 ReadingWritingListeningSpeakingGeneral
93%85%100%85%88%
Table 2.9 Key Contacts Questionnaire – design
Focus Number of Items
General Comments 7
BEC1 Reading and WritingListeningSpeakingGeneral
10643
BEC2 Reading and WritingListeningSpeaking
654
BEC3 Reading and WritingListeningSpeaking
942
In addition to the questions about the BEC papers, respondents were asked tocomment on the proposal to amalgamate the CEIBT (see Chapter 1) with BEC3– as both tests were essentially aimed at the same candidature. The response tothis item indicated that a clear majority (76%) of the respondents believed it wasa good idea.
The various consultative exercises resulted in the identification of a numberof areas which were potentially in need of revision. These were:
1. Changes to the overall structure of the tests:• more transparent names • BEC Preliminary to be refocused to the Main Suite PET level• speaking to have a stronger business focus • one overall grade (this included the issue of weighting of component
papers)• more explicit benchmarking for each level.
2. Other changes to specific test papers:• more time for Reading and Writing• Reading and Writing separate in Vantage and Higher• choice of tasks in Higher Writing (Part 2)• more attractive presentation• Speaking tests improved to generate greater range of language.
SummaryIn this chapter I reviewed the context for the revision of BEC, focusing on thetest development and revision methodology currently employed by CambridgeESOL. The revision was seen to have taken place through the setting up of taskspecific groups who were initially guided by the outcomes of an extensiveconsultative exercise – in which the impressions and observations of a range ofstakeholders were elicited.
The actual approach taken was guided by Cambridge ESOL’s five prongedapproach, which is designed to show a commitment to the assessment of a widevariety of language skills; assessment for a variety of purposes; a system ofcriterion levels; quality and fairness; and finally to an ongoing programme ofvalidation and test revision.
Another key element in the Cambridge ESOL approach is the focus on theVRIP (Validity, Reliability, Impact and Practicality) system of identifyingexamination qualities. While all of these were discussed, the focus on how relia-bility of language tests is viewed and reported was highlighted – with thesuggestion that what was called true reliability is not simply a measurementissue, but is related to all aspects of test and test system quality.
The design of the revision project was highlighted through the impact on the
Summary
117
decision-making process of the consultation exercises. The main findings ofthese exercises were described in terms of the changes suggested both within thedifferent levels and within each test. In the following chapter, these changes willbe described.
2 The revision of BEC
118
119
Major changes to the suite
Before looking at the changes to the individual papers in the BEC suite, it wouldbe useful to first identify and briefly describe some of the more significantchanges that have been made. Copies of sample papers at all three levels for theoriginal BEC suite can be found in Appendices 3.1 to 3.3, while past exami-nation papers for all three levels of the revised suits are included as Appendices4.1 to 4.3, and are published by Cambridge ESOL (2002a, 2002b, 2002c).
Reporting of results as one overall grade In the original BEC design, performance on the Speaking paper was reportedas a separate grade. This was because the original construct design did notinclude a speaking component, partially due to the perception in the marketfor which the test was originally designed that speaking was not as relevant tobusiness needs as the other skills, and partially to practical constraints. The mainconstraint was the lack of the considerable resources required to effectivelyoperate a Team Leader (TL) system (see Chapter 2). However, a Speaking paperwas made available to those candidates who had successfully completed theother papers as it was felt that the inclusion of a separate grade for these candi-dates gave a useful indication of the candidate’s language profile in the daysbefore graphical profiling was introduced.
By 1996, the TL system was in place in China and from that time all candi-dates were offered a Speaking paper. When the BEC suite came up for review in1999 graphical profiling was an established part of the Cambridge ESOLapproach to test performance reporting (i.e. in the Main Suite tests). This meantthat the conditions were in place for these innovations to be introduced.
The original system of reporting meant that each candidate received anoverall grade for performance on the three skills (Reading, Writing andListening). In addition, for those candidates who had passed at this point, aSpeaking paper was offered, performance on which was reported as:
• 1 – Higher• 2 – Minimum satisfactory• No Grade – less than satisfactory or absent.
Since the revised BEC suite examinations are designed to mirror the Main Suitetests at similar levels (see Figure 2.4), the reporting procedures reflect this. ForBEC Preliminary, results are reported as two passing grades (Pass with Merit
3
3 Major changes to the suite
120
and Pass) and two failing grades (Narrow Fail and Fail). This follows the modelused by the Cambridge ESOL Main Suite test at the equivalent level (PET).
Again following the model of the equivalent level Main Suite tests (FCE andCAE), in the BEC Vantage and BEC Higher, results are reported as threepassing grades (A, B and C) and two failing grades (D and E).
For all levels of BEC, candidates receive statements of results which, inaddition to their grades, show a graphical profile of their performance in eachskill. These are shown against the scale Exceptional – Good – Borderline –Weak, and indicate the candidate’s performance in each skill (see Figure 3.1).This scale takes account of relative differences in candidates’ performancesacross components and also if candidates have met fixed criteria in theSpeaking and Writing components. It is solely designed to provide feedbackto candidates to allow them to make considered judgements on their strengthsand weaknesses and so to allow them to adjust the focus of their languagelearning in the future.
Figure 3.1 The BEC statement of results
An additional impact of reporting performance as a single overall grade was thenotion of how the individual components were to be weighted.
Weighting of componentsOne change that has been made is the weighting of the different components ofthe tests, both within each paper, and within each BEC level. This weighting
Within each paper
121
system was designed both for improved measurement characteristics and toencourage a positive washback effect.
Within each paperThe system of scoring within each of the Reading papers is outlined in Tables3.1 – 3.3. As can be seen from these tables, there have been very minor changesto the internal weighting of the papers. The greatest single changes are to befound at the initial level, where there are an additional five items in the generalcomprehension and grammar sections.
Table 3.1 Internal weighting – Reading BEC1/BEC Preliminary
Table 3.2 Internal weighting – Reading BEC2/BEC Vantage
Table 3.3 Internal weighting – Reading BEC3/BEC Higher
Within the Writing paper the changes in weighting are to be found in Tables3.4 – 3.6. From these tables, we can see that it is at the initial level that the mostsignificant change has been made. These have the effect of making the three
Main Skill Focus BEC1 BEC Preliminary
Reading and vocabularyReading interpreting visual informationReading comprehensionGrammar
1051510
1051812
40 45
Main Skill Focus BEC2 BEC Vantage
Reading (scanning and gist)Reading comprehensionReading (gist and scanning for detail)VocabularyReading and grammar
7581510
7561512
45 45
Main Skill Focus BEC3 BEC Higher
Reading (gist and main idea)Reading (details and structure)Reading (gist and scanning for detail)VocabularyReading and grammarGrammar
866101010
866101012
50 52
3 Major changes to the suite
122
levels more similar in terms of how the tasks are presented and weighted – witheach level now consisting of a pair of tasks, the first of which is worth one-thirdof the available marks and the other worth the remaining two-thirds.
Table 3.4 Internal weighting – Writing BEC1/BEC Preliminary
Table 3.5 Internal weighting – BEC2/BEC Vantage
Table 3.6 Internal weighting – Writing BEC3/BEC Higher
Tables 3.7 – 3.9 indicate that the changes in internal weighting within theListening papers are again quite small. It is only at the earliest level that there isany change to be found. Here, there is an increase in the number of items in theinitial set of tasks – where the focus is on listening for detail, with the emphasison the second set of items reduced.
Table 3.7 Internal weighting – Listening BEC1/BEC Preliminary
Main Skill Focus BEC1 BEC Preliminary
Reading of written inputNote, message, memo or e-mail writingLetter writing
5510
Now a reading task1020
20 30
Main Skill Focus BEC2 BEC Vantage
Note, message or memo writingCorrespondence, report or proposal writing
1015
1020
25 30
Main Skill Focus BEC3 BEC Higher
Report writing – describing, comparing, inferringReport or proposal or correspondence writing
1020
1020
30 30
Main Skill Focus BEC1 BECPreliminary
Main Skill Focus
Listening for detail Listening for detail (numbers)Listening for specific informationListening for detail
84108
8778
Listening for specific informationListening for specific informationListening for specific informationListening for gist/specificinformation
30 30
Within each level
123
Table 3.8 Internal weighting – Listening BEC2/BEC Vantage
Table 3.9 Internal Weighting – BEC3/BEC Higher
Finally, while there have been major changes within the Speaking paper therehas been no change in the internal weighting. This is because the Speakingpaper, like other Cambridge ESOL Speaking papers, is scored by the awardingof a single score or set of scores (by the interlocutor and observer respectively)at the end of the test event – i.e. no distinction is made of performance on thedifferent tasks.
Within each levelIn the original BEC suite, the Reading paper was seen by the developers as beingthe most relevant skill for the candidates likely to sit the tests. For this reason theReading paper was the most heavily weighted. However, changes in the testpopulation, both within the original market and in the emerging BEC markets,meant that this situation could no longer be supported. Each paper in the revisedBEC tests is equally weighted, meaning that each skill now contributes 25% ofthe total marks available to the candidate. The effect of this is very clear. In theoriginal tests, there was a very heavy weighting on the receptive skills (over70% of the available marks not including the Speaking test). The new weightingmeans that there is a far greater emphasis on the productive skills, even at thelowest level. This change in emphasis is designed to bring the BEC tests into linewith the Cambridge ESOL approach (outlined in Chapter 2) and to promotewhat is perceived as positive washback. There was a slight difference in theweighting profile of the BEC2 papers when these are compared with BEC1.Again, the more heavily weighted components were related to the receptiveskills, with 45% of the available marks awarded for reading. A similar picturewas found in BEC3.
Main Skill Focus BEC2 BEC Vantage
Listening for detailListening to identify topic, context, function etc.Listening for specific information
12108
12108
30 30
Main Skill Focus BEC3 BEC Higher
Listening for detailListening to identify topic, context, function, opinion etc.Listening for specific information
1020
1020
30 30
3 Major changes to the suite
124
Table 3.11 Weighting at BEC2/BEC Vantage
Table 3.12 Weighting at BEC3/BEC Higher
These tables show how the weighting process has been used to radically alter theoverall distribution of focus within the BEC suite. In the original versions, the Reading paper was apparently seen as the most important, with 45% of theavailable marks available at each of the three levels. The Writing paper was leastheavily weighted at BEC1 with a systematic increase in weighting as the level ofthe test increased. This system was designed to show the greater importance ofwriting as overall proficiency level increased.
Speaking to have a stronger business focusThe original BEC Speaking papers were criticised during the consultationexercises for being too general in nature and for not really having a strong‘business’ orientation. The example of Task 2 from BEC1 (Task 1 was based onpersonal information exchange during a one-to-one interview) highlights theperceived problem, see Figure 4.16 in the next chapter. While we can see fromthe task that the topic of the information exchange task is business-related, theexpected output is at a very basic level (with little meaningful interaction or evenlanguage required to complete the task).
The revised test at this level (BEC Preliminary) has been radically changed.While the opening task remains focused on personal information exchange, thetime allowed has been much reduced. The old Task 2 has now been replacedwith a pair of tasks designed to elicit a broader range of language (see the sectionon the Speaking papers in the next chapter). In the new Task 2, candidates make
Table 3.10 Weighting at BEC1/BEC Preliminary
Level 1 Reading Writing Listening Speaking
Old system 40 (45%) 20 (22%) 30 (33%) Optional – reported on different scale
New system 45 (25%) 30 (25%) 30 (25%) 40 (25%)
Level 2 Reading Writing Listening Speaking
Old system 45 (45%) 25 (25%) 30 (30%) Optional – reported on different scale
New system 45 (25%) 30 (25%) 30 (25%) 40 (25%)
Level Reading Writing Listening Speaking
Old system 50 (45%) 30 (27%) 30 (27%) Optional – reported on different scale
New system 52 (25%) 30 (25%) 30 (25%) 40 (25%)
Speaking to have a stronger business focus
125
a mini-presentation on a business-related topic (see the example in Figure 4.17in the next chapter). In this task, there is clearly a greater emphasis on productionof language (the candidates may choose one of the two options, take one minutefor preparation, and then speak for a further minute on the topic).
In Task 3, the examiner outlines a scenario (see Figure 3.2), which the candi-dates discuss for two minutes before being asked further questions.
These examples from BEC Preliminary demonstrate how the paper at thislevel has changed, with a broader range of language potentially elicited, andmore relevantly for this section, a clearer focus on the business context. Thetasks shown here can claim a far greater degree of specificity (in terms of taskcontent and focus) and authenticity (both situational and interactional) than theoriginal BEC speaking task at the same level.
Figure 3.2 Discussion task – BEC Preliminary
Source: Cambridge ESOL BEC Handbook (2001:40)
3 Major changes to the suite
126
Other major changesIn addition to the changes described above, there are a number of relatedchanges to the BEC suite which at first sight appear cosmetic, but upon furtherinspection reveal something of how the benchmarking process (mentionedbefore in Chapter 2) impacted on the way in which the three individual testswere situated in terms of the Common European Framework.
Names
One criticism of the original BEC suite was the suggestion that the names of thetests were less than helpful to the stakeholders who either had to decide at whichlevel the tests were aimed or what performance they were based on, for examplewhat BEC2 might mean in terms of language ability.
The revised BEC exams have been renamed, partly in order to answer thiscriticism, but also to comply with the accreditation requirements of theQualifications and Curriculum Authority (QCA) in the United Kingdom – alltests in all subjects must submit documentation to the QCA in order to beaccredited for use in the UK. The exams were also renamed in order to reflect thegrowing influence of the Common European Framework (CEF) and Associ-ation of Language Testers in Europe (ALTE) framework.
Table 3.13 Changes to the BEC names
As can be seen in Table 3.13, the names of the revised exams have been changedfrom the original numbered system to one that more clearly reflects the level ofeach of the three. The most obvious of the names is BEC Vantage, named afterthe CEF level at which it is benchmarked (B2 or Vantage). The others arepossibly more obvious to the stakeholder who may be unfamiliar with theCEF/ALTE frameworks.
Another area of potential confusion with the original system was the factthat the numbers of the BEC tests did not correspond with the ALTE or CEFlevels they were designed to reflect, i.e. BEC1 was benchmarked to ALTELevels 1 and 2, BEC2 was benchmarked to ALTE Level 3 and BEC3 to ALTELevel 4.
The renamed exams were, with the exception of BEC1/BEC Preliminary,designed to replace the existing levels with an exam at the same level – though aswe shall see in the coming chapter, there were changes in the papers making up
Original name Name after revision process
BEC3BEC2BEC1
BEC HigherBEC VantageBEC Preliminary
Summary
127
the tests. Therefore we can say that the new BEC Preliminary represents themore challenging end of BEC1 (see the following section where this is exploredin more detail).
Level of BEC1/Preliminary
The results of the consultative exercise (see Chapter 2) indicated that, with theexception of the Speaking tests, the most noticeable area of concern was withBEC1. This concern was based on the fact that it essentially straddled two levelsof the CEF/ALTE framework and was dealt with in the original test by givingfour passing grades (with two each designed to reflect performance at each of thetwo levels tested within the test). Though this system was accepted and used bythe BEC stakeholders, the difficulty of adequately sampling from the broadlanguage domain covered by the test, given the constraints of test time andadministration, made the system difficult to operationalise in the longer term.
Table 3.14 Level of BEC1 and BEC Preliminary
The re-focusing of this level was achieved through the dual process of a detailedreference to the CEF/ALTE frameworks, and by making cross-comparisonswith the Main Suite tests which were representative of the same level CEF/ALTE levels.
An outline of the perceived level criteria for the revised examinationswas presented at a revision group meeting in October 1999. This documentcontained the data from which Table 3.15 has been created. In the table we cansee again that all three of the revised exams have been more deliberately bench-marked, with level descriptions, outlines of both formal language knowledgeand language use that more clearly identify the level at which each exam hasbeen aimed.
SummaryIn this short chapter I have outlined the major changes to the BEC suite. Thesechanges have been in the areas of:
• how results are reported – with a graphical representation of a performanceprofile, designed to have a diagnostic use for the candidate
• the weighting of the components both within papers and levels – this has the
Original BusinessEnglish Certificates
National QualificationsFramework Level
Council of Europe(ALTE) Level
Revised BusinessEnglish Certificates
BEC1 Entry 3Entry 2
B1 (ALTE Level 2)A2 (ALTE Level 1)
BEC Preliminary
3 Major changes to the suite
128
advantage of ensuring that the different elements of the papers and sub-skills within levels are seen as contributing equally to the candidate’scompetence
• a stronger business focus for the Speaking papers – this will be seen moreclearly in the coming chapter in which the changes to the individual papersare exemplified, and where it is clear that the major changes have come inthe BEC Speaking papers
• the naming of the papers – while these name changes are in one way
BEC Preliminary BEC Vantage BEC Higher
LevelDescription
This is a test forcandidates atCambridge/ALTE Level 2 [CEF B1]
This is a test forcandidates atCambridge/ALTE Level 3 [CEF B2]
This is a test forcandidates atCambridge/ALTE Level 4 [CEF C1]
FormalLanguageKnowledge
Learners at this level areexpected to deal with aspecified grammaticalinventory and understandand produce a restrictedvariety of structures.They should demonstrateknowledge of certainvocabulary items.
Learners at this level areexpected to be able tohandle the mainstructures of the languageand demonstrateknowledge of a widerange of vocabulary.
Learners at this level areexpected to be able tohandle complexstructures anddemonstrate knowledgeof a wide range ofvocabulary.
LanguageUse
Learners at this level can:• extract specific
information from shortspoken exchangeswithout necessarilyunderstanding everyword
• give and receivepersonal informationin a conversationalcontext
• take down informationin order to complete aform or memo
• read and understand avariety of business-related texts
• interpret charts anddiagrams
• produce a variety ofwritten texts in orderto convey specificinformation or feeling.
Learners at this level can:• understand the overall
meaning and keypoints of a non-specialist presentationor discussion
• participate in aconversation givingpersonal information,exchanginginformation andexpressing opinions
• take down informationfrom phoneconversations andpublic announcements
• read and understandgeneral businessletters, reports, articlesand leaflets
• produce letters,memos and simplereports.
Learners at this level can:• engage in extended
conversation• contribute effectively
to meetings andseminars
• take accurate notesduring meetings
• write reports and draftinstructions
• understand mostcorrespondence,articles and reportswhere information isovertly stated
• use the telephone formost purposes
• negotiate successfullyin most situations.
Table 3.15 Revised BEC level criteria
Summary
129
superficial, they do have a role to play in the way the different levels are seenboth within the British education system, and in the way they are seen withinthe context of the ALTE/CEF levels
• the level of BEC1/Preliminary – where the original paper was not clearlybenchmarked to any definite level; the revised BEC Preliminary is nowmore obviously representative of the ALTE/CEF B1 level, where theoriginal attempted to straddle the A2 and B1 levels.
130
Changes in the BEC papers
In Chapter 4, I will describe the way in which the revision process has led tochanges in the BEC suite of examination, and since the four skills of Reading,Writing, Listening and Speaking are tested in the suite (in that order) this will bereflected in the organisation of the chapter. Sample copies of the three tests in therevised BEC Suite can be found in Appendices 4.1 to 4.3.
Changes in the Reading papersIn the following section of the BEC revision overview the changes made to theReading papers at the three levels are presented. The changes are described interms of the outline of the Cambridge ESOL approach outlined in the previous
4
Part Items Main Skill Focus Focus Format Marks
1 1–5 Reading andvocabulary
Understandingintended meaning(short texts, e.g. signs)
3 option MCQ 5
2 6–10 Reading andvocabulary
Understanding basicvocabulary (frombusiness signs,adverts)
Matching (5 from 8) 5
3 11–15 Reading interpretingvisual information
Interpretinginformation frominput (e.g. charts)
Matching (5 from 8) 5
4 16–22 Readingcomprehension
Comprehension ofwritten input (e.g.report)
T/F/not included 7
5 23–26 Readingcomprehension
Comprehension ofwritten input (e.g.information sheet)
3 option MCQ andmatching
4
27–30 Readingcomprehension
Same input Select correctoptions (4 from 7)
4
6 31–40 Grammar Grammar use incontext (rationaldeletion cloze)
3 option MCQ 10
Total marks 40
Table 4.1 BEC1 Reading paper outline
Total time allowed 70 minutes (40 Reading items + 3 Writing tasks)
Changes in the Reading papers
131
chapter. As will be seen throughout the chapter, there are occasions when nomajor changes were made to individual papers; this is particularly true of theReading papers, where the review process suggested that there were no majorchanges needed.
BEC1 and BEC Preliminary
At BEC level 1, the Reading paper was originally designed to test a range ofreading-related skills including both vocabulary and grammar in context,general comprehension and scanning for detail (see Table 4.1).
As can be seen from the outline of the revised paper at this level the constructremains very much the same. The changes that have been made include theaddition of five items in the latter half of the test and the provision of additionaltime. However, as the Reading and Writing papers are presented as a single unitat this level it is not clear exactly how the candidates will use this additionaltime, see the chapter relating to the changes in the Writing paper for someadditional comments on this. The additional items have had the effect of addingto the internal consistency of the paper.
The single most important change in the BEC1/BEC Preliminary Reading
Part Items Main Skill Focus Focus Format Marks
1 1–5 Reading andvocabulary
Understandingintended meaning(short texts, e.g. signs)
3 option MCQ 5
2 6–10 Reading andvocabulary
Understanding basicvocabulary (fromshort input)
Matching (5 from 8) 5
3 11–15 Reading interpretingvisual information
Interpretinginformation frominput (e.g. charts)
Matchingstatements to chartdata
5
4 16–22 Readingcomprehension
Comprehension ofreport
T/F/not included 7
5 23–28 Readingcomprehension
Comprehension ofwritten text
3 option MCQ 6
6 29–40 Grammar (in contextof reading text)
Grammar use incontext (rationaldeletion cloze)
Cloze (3 optionMCQ)
12
7 41–45 Reading andinformation transfer
Reading for speciflcdetail from twowritten inputs
Form completion 5
Total marks 45
Table 4.2 BEC Preliminary Reading paper outline
Total time allowed 90 minutes (45 Reading items + 2 Writing tasks)
4 Changes in the BEC papers
132
paper was a tightening up of Part 5, and the introduction of a new formcompletion task as Part 7 – actually, it was originally part of the Writing paperbut was moved here to reflect more accurately the construct being tested in bothpapers. Other important changes to this paper included the addition of alter-native input sources to Part 2 – where the candidate now identifies specificelements within more ‘realistic’ sources.
In the original version of Part 5 (Figure 4.1), we can see that there is a singletext (either divided into four paragraphs or presented as four sub-texts). Based
Figure 4.1 BEC1 Part 5 Reading
Changes in the Reading papers
133
on this reading input there are four comprehension items (three option, MCQformat) that ask the reader to read for specific details from individual sub-texts,and a further four items focusing on reading for detail, but this time using amatching format. The task seems to have been intended to provide the reader
Source: Cambridge ESOL BEC Handbook (2000:21)
4 Changes in the BEC papers
134
Figure 4.2 BEC Preliminary Task 5 Reading
Source: Cambridge ESOL BEC Preliminary, Examination Report and Past Examination Papers(2002:28–29)
Changes in the Reading papers
135
with a dual focus (reading for comprehension and detail), but ultimately appearsto test the same thing within both parts – task developers also reported that it wasvery difficult to find suitable texts, a problem which ultimately led to the failureof the task as the text requirements of the different formats in the two parts weredifferent – comprehension items can quite successfully be based on shortcohesive texts while items that focus on reading for detail require longerrelatively ‘shapeless’ texts.
In the revised version of Part 5, this problem has been addressed through thedecision to create a task with a single focus (see Figure 4.2). Here we can see thatthere is a single reading text of approximately 350 words. This input is accom-panied by a series of six comprehension items, each related to a separateparagraph in the input text.
The effect of this change is to simplify the section, giving a single clear focuson how the text is to be exploited. This is in marked contrast to the originaldesign, in which the task purpose was not really made clear to the candidate orthe test observer/evaluator.
BEC2 and BEC Vantage
At BEC2, the Reading paper was again designed to test a range of reading-related skills. At this level, the reading and writing skills were, as with BEC1,tested using a single paper. The situation was changed with the revision and twoseparate papers were offered. This complicates any comparisons of the Readingpapers, though really only in that it was never clear how candidates used the timeallowed for the Reading and Writing papers at BEC2 while with the BECVantage (the revised title for the examination) the time for each paper is set. Itwas also a concern that candidates could take information from the Reading anduse it in their Writing in an inappropriate manner.
In the same way that Task 5 on BEC1 was found to be problematic, the factthat Task 5 in BEC2 (See Figure 4.3) was based on two sets of items related totwo different texts meant that it too was in need of change – more related tosimplifying the task writing process than to changing the actual content of thetask. The actual change to the overall task is small, as the activity engaged in forboth the original version and the revised version (Figure 4.4) is the same – bothinvolve identifying problematic or non-problematic lines in a short text. InBEC2 the task had two parts, with the first focusing on a possible extra word inany line and Part 2 on a possible incorrect word which had to be corrected (it waspossible in both cases that there was no error in a line).
Another problem with the task was related to the format of the second section.Here the candidate was first meant to identify a possible error and then write thecorrect word in the response boxes in their answer book. The difficulty is that thecandidate might see a problem where none exists and offer a correction, missingthe real problem. Where the correction offered actually matched the expected
4 Changes in the BEC papers
136
response the candidate would be seen by the examiner to have answeredcorrectly – this is because the candidate did not have to identify the position ofthe error. This meant that the task was very difficult to write and it was notalways certain that the candidates’ responses matched the expectations of thetask writer.
The BEC Vantage version of the task has a single text of 14 lines (the first twoof which are examples) in which the offending word is said to be ‘either
Part Items Main Skill Focus Focus Format Marks
1 1–7 Reading (scanningand gist)
Understanding intendedmeaning (short texts,e.g. signs)
Matching(7 sentences to4 texts)
7
2 8–12 Reading Understanding textstructure
Matching (sentencelevel gaps)
5
3 13–20 Reading (gist andscanning for detail)
Interpreting overallmeaning and identifyingspecific details
MCQ (4 option) 6
4 21–35 Vocabulary Recognising vocabularyuse in context
MCQ cloze(4 option)
15
5 36–45 Reading andgrammar
Proof-reading task Identify additionalunnecessary words
12
Total marks 45
Table 4.4 BEC Vantage Reading paper outline
Table 4.3 BEC 2 Reading paper outline
Total time allowed 60 minutes (45 Reading items)
Total time allowed 90 minutes (45 Reading items + 2 Writing tasks)
Part Items Main Skill Focus Focus Format Marks
1 1–7 Reading (scanningand gist)
Understanding overallmeaning (short texts)
Matching (7sentences to 4 texts)
7
2 8–12 Reading Understanding textstructure
Text completion (5gaps with 9 options)
5
3 13–20 Reading (gist andscanning for detail)
Interpreting overallmeaning and identifyingspecific details
Matching (each part4 from 7)
44
4 21–35 Vocabulary Recognising vocabularyuse in context
MCQ cloze 15
5 36–40 Reading andgrammar
Proof-reading task Identify and correcterror
5
41–45 Reading andgrammar
Proof-reading task Identify and correcterror
5
Total marks 45
Changes in the Reading papers
137
grammatically incorrect or does not fit in with the meaning of the text’. Whilethis version may not mimic a genuine proof-reading task, it does offer the testermore control over the output, making for a potentially more reliable set of items,while at the same time offering a somewhat more viable proof-reading taskwhere the candidates are required to access a wider range of linguisticknowledge in order to respond. There is, of course some question as to whethera proof-reading task represents a test of reading ability, or a test of linguisticknowledge set in a reading context.
Figure 4.3 BEC2 Task 5 Reading
Source: Cambridge ESOL BEC Handbook (2000:49)
4 Changes in the BEC papers
138
Figure 4.4 BEC Vantage Task 5 Reading
*Note: some questions have not been included here
Source: Cambridge ESOL BEC Vantage, Examination Report and Past Examination Papers (2002:28)
It is clear from the two tables (Tables 4.3 and 4.4) and the two figures (Figures4.3 and 4.4) that there are no other major changes to the Reading paper. Both thenumber of items and the general focus of the items remain the same. A review ofthe actual tasks shows that the setting of tasks in a business context remains thesame in the two versions, with the only change being that to Task 5, describedabove. (See Appendix 4.1 for examples of the Reading papers from the threeexaminations on the revised BEC suite.)
Changes in the Writing papers
139
BEC3 and BEC Higher
Tables 4.5 and 4.6 contain the outlines of the Reading papers at the highest of thethree BEC levels (BEC3 and BEC Higher respectively). There are no realchanges here, with the exception of a slight increase in the number of items forthe proof-reading section. Like the change to BEC2, BEC Higher splits theReading and Writing papers.
Table 4.5 BEC3 Reading paper outline
Total time allowed 100 minutes (50 Reading items + Writing tasks)
From this brief review of the Reading papers at the three BEC levels, we can seethat there were very few substantial changes made in the revision process. Thefeedback from the consultation exercise (reported in Chapter 2) suggested thatthe only real area of concern with the BEC papers lay in Part 5 of BEC1. Thisproblem was dealt with by eliminating the double-focus of the part so that therewas a single clear area of interest. In general, the changes, though slight, appearto have made the construct clearer. The papers are more consistent in the waythey approach the testing of reading, with the emphasis on careful reading forgist and for detail, with an additional focus on testing vocabulary and grammarin the context of reading.
Changes in the Writing papersAs we saw in the review of the changes to the Reading papers, there werechanges to the way in which the Reading and Writing papers are presented. In
Part Items Main Skill Focus Focus Format Marks
1 1–8 Reading (gist andmain idea)
Understanding intendedmeaning (short texts,e.g. signs)
Matching(8 sentences to5 texts)
8
2 9–14 Reading (detailsand structure)
Understanding ofspecific details andstructure of ‘authentic’business text
Text completion(6 gaps with8 options)
6
3 15–20 Reading (gist andscanning for detail)
Interpreting overallmeaning and identifyingspecific details
MCQ (4 options) 6
4 21–30 Vocabulary Recognising vocabularyuse in context
MCQ cloze(4 option)
10
5 31–40 Reading andgrammar
Rational deletion clozecompletion
Cloze 10
6 41–50 Reading andgrammar
Proof-reading task Identify additionalunnecessary words
10
Total marks 50
4 Changes in the BEC papers
140
the original format, the two papers were presented as a single unit, with a totaltime given to the candidates. This may have had an unintended negative effect interms of time management (and a potentially negative washback effect, wherewriting is seen as being of lesser importance than reading) on the way in whichthe Writing paper was seen by candidates, as there was a clear difference in thescores awarded for the two sections (the Writing paper offered half the marks ofthe Reading paper at BEC1, one third at BEC2 and two fifths at BEC3).
BEC1 and BEC Preliminary
Table 4.7 shows that for BEC1 there were three different tasks included in thepaper. One criticism of the paper focused on the first five items, built aroundwhat was essentially a reading and information transfer task. The latter pair offree writing tasks were rated using a relatively simple set of scales. For the firstof these tasks, candidates’ work was rated on a 5-point scale which was focusedon task completion. For the second task, a pair of scores was awarded, one fortask completion and the other for language. The latter pair of tasks were bothscaffolded using a series of bullet pointed suggestions.
In BEC Preliminary, the first task has been altered and the expected output forthe two remaining tasks has been lengthened, each by 10 words. Both of thesetasks are scored using a General Impression Mark Scheme (GIMS). In fact, thetwo tasks use somewhat different versions of the scale, the first containing a setof very basic descriptors, while the second contains a more complex set whichfocuses both on task completion and language. Both versions are 6-level (0–5)
Items Main Skill Focus Focus Format Marks
1–8 Reading (gist andmain idea)
Understanding based on‘authentic’ business text
Matching (8 sentencesto 5 texts)
8
9–14 Reading (detailsand structure)
Understanding of specificdetails and structure of‘authentic’ business text
Text completion(6 gaps with 8 options)
6
15–20 Reading (gist andscanning for detail)
Interpreting overall meaningand identifying specific details
MCQ (4 options) 6
21–30 Vocabulary Recognising vocabulary use incontext
MCQ cloze (4 option) 10
31–40 Reading andgrammar
Rational deletion clozecompletion
Cloze 10
41–52 Reading andgrammar
Proof-reading task Identify additionalunnecessary words
10
Total marks 52
Table 4.6 BEC Higher Reading paper outline
Total time allowed 60 minutes (52 Reading items)
Changes in the Writing papers
141
scales. The scores for the tasks are weighted, with the second task worth twicethe number of marks as the first task. The overall weighting of the Writing paperhas been increased, making it worth 25% of the total score for the test (all fourpapers are now equally weighted at the three BEC levels). This makes therevised paper a clearer reflection of Cambridge ESOL’s stated commitment tothe inclusion of all four skills in their language tests (Saville 2003:62).
In order to ensure that the Writing paper accurately reflects the amendedlevel of the test, both General Impression Mark Schemes are interpreted atCambridge/ALTE level 2.
Table 4.8 BEC Preliminary Writing paper outline
Total time allowed 90 minutes (45 Reading items + 2 Writing tasks)
Figure 4.5 shows the original information transfer task from BEC1. As we cansee from this task, the output required of the candidate was simply to retrieve therelevant information from the input (in the form of a very brief memo andreceipt) and complete the simple form. The amount of writing was minimal, infact the task was based on information transfer and all responses could be foundin the reading input. For this reason the task was perceived to be more related toreading and as such it was moved from its original position in the writing section(BEC1 Part 7) to the revised reading section (BEC Preliminary Part 7). Inaddition to the move, the amount of reading input has been increased – one of thevariables that has been hypothesised by Norris et al (1998), O’Sullivan & Weir(2000) and Skehan (1998) to impact on task difficulty as it relates to ‘codecomplexity’ (number and amount of linguistic input). The revised version of thistask can be seen in Figure 4.6.
Table 4.7 BEC1 Writing paper outline
Total time allowed 70 minutes (40 Reading items + 3 Writing tasks)
Part Items Main Skill Focus Focus Format Marks
7 41–45 Reading of written input Information transfer Form completion 5
8 46 Memo writing Short written output(some scaffolding)
Free writing 5
9 47 Letter writing Short written output(some scaffolding)
Free writing 10
Part Items Main Skill Focus Focus Format Marks
1 46 Note, message, memoor e-mail writing
Short written output(some scaffolding)
Free writing 10
2 47 Letter writing Short written output inresponse to written input(some scaffolding)
Free writing 20
4 Changes in the BEC papers
142
Figure 4.5 BEC1 Part 6 Writing
Source: Cambridge ESOL, BEC Handbook (2000:23)
Changes in the Writing papers
143
Figure 4.6 BEC Preliminary Task 7 Reading
Source: Cambridge ESOL BEC Preliminary, Examination Report and Past Examination Papers(2002:32–33)
4 Changes in the BEC papers
144
BEC2 and BEC Vantage
Tables 4.9 and 4.10 show the way in which the Writing paper has been changedat the next level (BEC2 and BEC Vantage).
Table 4.9 BEC 2 Writing paper outline
Total time allowed 90 minutes (45 Reading items + 2 Writing tasks)
One of the major changes to the structure of the test (described in Chapter 3) isthe decision to place more emphasis on writing. It is at BEC Vantage that thisdecision is first manifested. At this level, we see that there are now separatepapers for Reading and Writing – and the weighting system (as with BECPreliminary) now means that the Writing paper is similar to the other threepapers, in that all are worth 25% of the total score for the test.
The first of the two tasks is quite similar in terms of input and expectedresponse, though there is a nod in the direction of contemporary businesscommunication with the inclusion in the specifications of written e-mailcommunication to the existing list of response formats used in BEC2; the otheroptions are note, message or memo. The other change to this first task is that theresponse is expected to be slightly longer.
The second writing task is quite different in terms of length of expectedresponse, type of input and output format. The candidate is expected to write asignificantly longer text (120–140 words as opposed to 100–120 words at BEC2), and the input can either be written or presented as tables/graphics/charts.This change in the nature of the input may have an impact on the difficulty of thetask, though any impact is lessened by the inclusion of written notes on thegraphics in order to make interpreting them less of an issue. The potentialproblem here is the nature of the information transfer. In the original task, theletter was based on a very basic transfer of information – the fact that the inputwas read meant that language was provided, for example. The new version asksthe candidate to transform information from a chart (which must be interpreted)to a written format. While the written notes may act to negate any significanteffect on task difficulty, there is no empirical evidence that the different inputtypes result in significantly different responses. The change from a letter to areport may also be a complicating factor with this task.
Items Main Skill Focus Focus Format Marks
46 Note, message ormemo writing
Short written output (30–40words – very basic scaffolding)
Free writing 5
47 Letter writing Letter (100–120 words –respond to written inputs)
Free writing 10
Changes in the Writing papers
145
Table 4.10 BEC Vantage Writing paper outline
Total time allowed 45 minutes (2 Writing tasks)
BEC3 and BEC Higher
The Writing paper at BEC Higher has also been separated from the Reading/Writing structure of BEC3 (see Tables 4.11 and 4.12). It is at this level that themost clearly defined changes have been made to the paper.
The major changes are:
• output for task 1 has been lengthened to 120–140 words (up from100 words)
• a choice has been offered in Task 2.
The impact of increasing the required output for the first task is to make the tasksomewhat more realistic – it being unusual to find a report in the businesscontext that is just 100 words long. While the report might, in an ideal situation,be even longer than the new range, the practical limitations of the test eventmake writing a longer text impossible unless the test is reduced to a single task.In addition to anecdotal evidence in support of using multiple tasks, Bachman,Lynch and Mason (1995) have presented empirical evidence that havingadditional tasks has a greater impact on test reliability than having additionalraters, so it would be unwise to reduce the number from the present two to asingle task.
Table 4.11 BEC3 Writing paper outline
Total time allowed 100 minutes (50 Reading items + 2 Writing tasks)
The additional time allowed for the Writing paper means that the amount ofwritten output expected of the candidate is now slightly greater than in the
Items Main Skill Focus Focus Format Marks
1 Note, message, e-mailor memo writing
Short written output (40–50words – very basic scaffolding)
Free writing 5
2 Correspondence, reportor proposal writing
Written output (120–140 words– respond to written inputs)
Free writing 10
Items Main Skill Focus Focus Format Marks
51 Report writing Short written output (100 words– input from simple graphs)
Free writing 10
52 Report writing Letter (200–250 words – basedon limited written input andsome scaffolding)
Free writing 20
4 Changes in the BEC papers
146
original. The other change relates to the fact that by offering a choice of tasks inthe second part of the Writing paper, the developers are also offering a choice ofoutput type. Candidates are asked to write on one of the three options, thesebeing a report, a proposal and a piece of business correspondence. The input forall three options is very similar in terms of length and degree of scaffolding (allprovide four bullet-pointed guiding points), and all three ask for the sameamount of written output. As with any situation where a choice is offered, thereis a danger that the different tasks will result in different levels of performance.However, the fact that the input for each choice is so similar suggests that anygains will be attributable to candidate ability – thus the choice can be seen as‘testing for best’ – in that a candidate will, it is hoped, opt for the output typewhich they perceive as offering the best chance for an acceptable performance.As mentioned above, this aspect of the task should be monitored over time toensure that no unintended bias occurs.
Table 4.12 BEC Higher Writing paper outline
Total time allowed 70 minutes (2 Writing tasks)
Figures 4.7 and 4.8 show the original version of the task and the revisedversion from BEC Higher respectively. From these two examples we can seethat the actual task has not altered, in that the format of the input remains thesame in the two test versions. However, in the revised version of the task candi-dates are offered a choice of writing one of three options, a report (as in theoriginal BEC), a proposal or a letter. The decision to offer candidates a choice isnot without problems, and care must be taken to ensure that candidates are notnegatively affected by their choice of task. Analysis of trial and test data showsthat there has been no negative impact to date – with no significant differences inthe scores achieved for the different options across the test population. Ofcourse, this situation must be monitored at each administration.
Changes to the rating procedureOne change that has had an effect on all of the BEC levels except BEC Higher(where the rating procedure has not changed) is the fact that writingperformance is now rated using a different scale.
Items Main Skill Focus Focus Format Marks
1 Report writing –describing, comparing,inferring
Short written output (120–140words – input from simplegraphic)
Free writing 10
2 Report or proposal orcorrespondence writing
Written output (200–250 words– based on limited written inputand some scaffolding)
Free writing(choice from 3)
20
Changes in the Writing papers
147
For all tasks and levels, two mark schemes are used:
1. General Mark Scheme: this included six criteria, each with detaileddescriptors at five levels or bands: These werei) contentii) vocabulary and structure rangeiii) accuracyiv) organisation of information and textv) appropriacy of register and formatvi) effect on target reader.
2. Task Specific Mark Scheme: this gave guidance to the rater on the features ofan appropriate response at the different levels.
With the revised BEC papers, the situation differs depending on the test level.While all tasks at all levels are rated using two separate scales (General and TaskSpecific), at BEC Preliminary level the first of the two tasks is scored using aversion of the General Mark Scheme in which task achievement only isaddressed.
One of the advantages to using a simplified scale such as this is that thedescriptors are easily kept in mind as they are so short. The fact that this scale isused in conjunction with a task specific scale (i.e. the specific ‘content points’referred to above are outlined in detail) makes the rating of this task very reliable– as it is relatively easy for raters to make consistent estimates of performancelevel.
Figure 4.7 BEC3 Part 2 Writing
Source: Cambridge ESOL BEC Handbook (2000:77)
4 Changes in the BEC papers
148
Figure 4.8 BEC Higher Part 2 Writing
Source: Cambridge ESOL BEC Higher, Examination Report and Past Examination Papers (2002:35)
Changes in the Writing papers
149
For the second task at BEC Preliminary and all tasks at BEC Vantage andHigher, a different type of General Impression Mark Scheme is used (seeFigure 4.9, which shows the BEC Preliminary version of the new GIMS).
This scheme is far more detailed and includes reference to the criteria used inthe original BEC suite, e.g. content, range and accuracy of vocabulary and
Band 5 Full realisation of the task set:• all four content points achieved• confident use of language; errors are minor, due to ambition and non-
impeding• good range of structure and vocabulary• effectively organised, with appropriate use of simple linking devices• register and format consistently appropriate.Very positive effect on the reader
Band 4 Good realisation of the task set:• three or four content points achieved• ambitious use of language; some non-impeding errors• more than adequate range of structure and vocabulary• generally well-organised, with attention paid to cohesion• register and format on the whole appropriate.Positive effect on the reader
Band 3 Reasonable achievement of the task set:• three or four content points achieved• a number of errors may be present, but are mostly non-impeding• adequate range of structure and vocabulary• organisation and cohesion is satisfactory, on the whole• register and format reasonable, although not entirely successful.Satisfactory effect on the reader
Band 2 Inadequate attempt at the task set:• two or three content points achieved• numerous errors, which sometimes impede communication• limited range of structure and vocabulary• content is not clearly organised or linked, causing some confusion• inappropriate register and format. Negative effect on the reader
Band 1 Poor attempt at the task set:• one or two content points achieved• serious lack of control; frequent basic errors• little evidence of structure and vocabulary required by task• lack of organisation, causing breakdown in communication• little attempt at appropriate register and forma Very negative effect on the reader
Band 0 Achieves nothing. Either fewer than 25% of the required number of words or totallyillegible or totally irrelevant.
Figure 4.9 BEC Preliminary revised General Impression Mark Scheme
4 Changes in the BEC papers
150
grammar, organisation, register, and effect on the reader. The raters award asingle impression score based on the descriptors.
The fact that a separate task specific scheme is used for each task and thatexaminers are familiar with the interpretation levels for the BEC suite, meansthat the system can result in reliable and consistent rating.
The GMS is interpreted at the following levels:
BEC Preliminary Cambridge/ALTE Level 2BEC Vantage Cambridge/ALTE Level 3BEC Higher Cambridge/ALTE Level 4
The great value of this method is that it reinforces the link to the Cambridge/ALTE levels (and therefore to the Common European Framework). While wehave seen that the BEC suite examinations have been developed with theseexternal performance criteria in mind, the fact that the rating of an individual’stest performance is based directly on the criteria reinforces the link to thosecriteria and as such offers evidence of test validity.
As can be seen from this section, there have been a number of quitesignificant changes to the Writing papers, particularly with the choice nowoffered at BEC Higher for the second writing task. The other changes include anincrease in the length of the required output for the initial writing task at alllevels, and for Task 2 at BEC Preliminary and BEC Vantage, the separation ofthe Reading and Writing papers at BEC Vantage and the use of a commonGeneral Impression Mark Scheme, but interpreted at different performancelevels and tied to the Cambridge/ALTE levels. These changes combine to makethe revised Writing papers more reliable and valid – in that they represent aclearer business orientation – in terms of context, output text type and length.
Changes in the Listening papersFrom the following description of the old and revised BEC Listening papers, wecan see that there have been few changes made. This is because there was ageneral satisfaction with the Listening papers on the part of the developers andthose people who were asked to comment on the test during the review stage.
BEC1 and BEC Preliminary
We can see from Tables 4.13 and 4.14 that there have been few substantivechanges to the Listening paper. While the sections remain essentially the same,there has been an attempt made to spread out the items more evenly over the foursub-tests. The revised paper continues to test a variety of sub-skills using a rangeof test formats, again in keeping with the Cambridge ESOL approach outlined inChapter 2.
Changes in the Listening papers
151
It is at the lowest level that the only substantial change has occurred. In BEC1Part 2 (see Figure 4.10) the listener is required to identify a series of fournumbers from a short listening text and then use these to complete a simple form.
Table 4.13 BEC 1 Listening paper outline
Total time allowed 40 minutes (30 items)
Table 4.14 BEC Preliminary Listening paper outline
Total time allowed 40 minutes (30 items)
BEC1 Part 3 (Figure 4.11), then asks the candidates to listen to a conversationfor specific ‘words or a number’. In a second conversation, the listenercompletes a form while listening to non-number based details. Between the twoparts there are a total of 14 items, though there appears to be an overlap in focusbetween Parts 2 and 3. This overlap is both confusing (what are the items tryingto test?) and at best potentially redundant (if the items are testing the samething).
Items Main Skill Focus Focus Format Marks
1–8 Listening for detail Information transfer (shortconversations/monologues)
MCQ (3 option) 8
9–12 Listening for detail Information transfer (shortconversations/monologues)
Gap filling (numbers) 4
13–22 Listening andwriting
Form completion Gap filling (words andnumbers)
10
23–30 Listening forspecific information
General comprehension anddetailed listening
MCQ (3 option) 8
Items Main Skill Focus Focus Format Marks
1–8 Listening for detail Information transfer (shortconversations/monologues)
MCQ (3 option) 8
9–15 Listening for detail Information transfer (shortconversations/monologues)
Gap filling (words,numbers, letters)
7
16–22 Listening andwriting
Form/note completion Gap filling (1 or 2words)
7
23–30 Listening forspecific information
General comprehension anddetailed listening
MCQ (3 option) 8
4 Changes in the BEC papers
152
Figure 4.10 BEC1 Listening Part 2
Source: Cambridge ESOL BEC Handbook (2000:30)
Figure 4.11 BEC1 Listening Part 3
Changes in the Listening papers
153
Source: Cambridge ESOL 2000: BEC Handbook (2000:31)
4 Changes in the BEC papers
154
The revised paper deals with this problem by expanding Part 2 to include sevenitems involving listening for specific detail in the form of a ‘word, numbers orletters’ (see Figure 4.12). As with the original version, there was some supportoffered to the listener as some of the details in the form were included.
Figure 4.12 BEC Preliminary Listening Part 2
Source: Cambridge ESOL BEC Preliminary, Examination Report and Past Examination Papers(2002:41)
BEC Preliminary Part 3 (Figure 4.13) then focuses on completing a set of noteswith seven items which focus on using ‘one or two words’. This task is thereforesomewhat different from Part 2, in that the focus is now clearly on words only.The result of these changes is to maintain the same number of items, whilemaking the two parts more clearly distinct.
Figure 4.13 BEC Preliminary Listening Part 3
Source: Cambridge ESOL BEC Preliminary, Examination Report and Past Examination Papers(2002:42)
BEC2 and BEC Vantage
At the next levels (BEC2/BEC Vantage; BEC3/BEC Higher), we can see thatthere have been no changes made to the Listening papers (Tables 4.15 to 4.18).
Changes in the Listening papers
155
Table 4.15 BEC2 Listening paper outline
Total time allowed 40 minutes (30 items)
Table 4.16 BEC Vantage Listening paper outline
Total time allowed 40 minutes (30 items)
BEC3 and BEC Higher
Table 4.17 BEC3 Listening paper outline
Total time allowed 40 minutes (30 items)
4 Changes in the BEC papers
156
Items Main Skill Focus Focus Format Marks
1–12 Listening for detail Information transfer (shortconversations/monologues)
Form/note completion 12
13–22 Listening to identifytopic, context,function etc.
Listening for specificinformation from 2 shortmonologues/dialogues
Matching extract tostatement (5 items to 8options)
10
23–30 Listening forspecific information
General comprehension anddetailed listening
MCQ (3 option) 8
Items Main Skill Focus Focus Format Marks
1–12 Listening for detail Information transfer (shortconversations/monologues)
Note completion 12
13–22 Listening to identifytopic, context,function, opinionetc.
Information transfer (shortconversations/monologues)
Matching extract tostatement (reasonsand reactions)
10
23–30 Listening forspecific information
General comprehension anddetailed listening
MCQ (3 option) 8
Items Main Skill Focus Focus Format Marks
1–12 Listening for detail Information transfer (shortconversations/monologues)
Form/note completion 12
13–22 Listening to identifytopic, context,function etc.
Listening for specificinformation from 2 shortmonologues/dialogues
Matching extract tostatement (5 items to 8options)
10
23–30 Listening forspecific information
General comprehension anddetailed listening
MCQ (3 option) 8
Table 4.18 BEC Higher Listening paper outline
Total time allowed 40 minutes (30 items)
The Listening papers reviewed here represent the least altered papers of the BECsuite. The changes that were made were based on feedback from the consul-tation exercise. The Listening papers of the suite have not been seen, either bythe developers or by the stakeholders, as being problematic over the years. Theyrepresent a practically effective set of papers that offer a view of listening forspecific purposes where the tasks and the language are both set in a businesscontext.
Changes in the Speaking papersIt is in the Speaking papers that the most obvious changes have been made.Criticism of the BEC1 Speaking paper tended to focus on the lack of specificityof the task topics – with half of the test devoted to a personal informationexchange task and the other to an information transfer task, which, although itwas set in a business context, did not really reflect the type of speaking tasktypical of the domain (see Table 4.19 and Figure 4.14). In the revised version,the number of tasks has been increased to three (Table 4.20), with the intro-ductory task greatly reduced in scope – the task still operates as a sort of ‘lowimpact’ introduction to the test event, in terms of cognitive demand andcandidate anxiety.
In terms of the tasks included in the revised version of the test, the second taskmarks the singular most important change. The introduction of the individuallong turn with follow-up questions/comments by another candidate adds animportant dimension to the test event, namely that of broadening the potentialfor the test as a whole to elicit a greater range of language functions. Thispotential has been demonstrated by O’Sullivan, Weir and Saville (2002) in theirreport on the development of a set of ‘Observation Checklists’, used by taskwriters to predict the linguistic outcomes of Speaking test tasks in terms of infor-mational, interactional and discourse management functions, and again byvalidation researchers to establish empirically that the predictions could be
Changes in the Speaking papers
157
Items Main Skill Focus Focus Format Marks
1–12 Listening for detail Information transfer (shortmonologues)
Gap fill, note completion(up to three words or anumber)
12
13–22 Listening to identifytopic, context,function etc.
Information transfer (shortconversations/monologues)
Matching extract tostatement (reasons andreactions)
10
23–30 Listening forspecific information
General comprehensionand detailed listening
MCQ (3 option) 8
supported. Essentially, the checklists allow the researcher/validator to generatea profile of a test task in use. This profile, based on the elicitation of languagefunctions, can be used to make working descriptions of the tasks through whichmeaningful comparisons can then be made. Figure 4.14 represents a mapping ofthe probable function pattern (or profiles) elicited by the three different tasksused in the revised BEC suite.
The profiles, based on data reported by O’Sullivan, Weir and Saville (2002)and modified to predict the outcome of the tasks in the BEC suite, show how the
4 Changes in the BEC papers
158
Task 1 Task 2 Task 3
Informational Functions
Provide personal information
Present
Past
Future
Expressing opinions
Elaborating
Justifying opinions
Comparing
Speculating
Staging
Describing a scene
Expressing preferences
Interactional Functions
Agreeing
Disagreeing
Modifying
Asking for opinions
Negotiating of meaning understanding
respond to req. clarification
Managing Interaction
Initiating
Changing
Reciprocating
Figure 4.14 Profile of language elicited by tasks used in the revised BECsuite Speaking paper
Key: Task 1 – one-to-one interviewTask 2 – Individual Long Turn (with follow-up comments etc.)Task 3 – Two-way interaction (candidate-to-candidate)
three different interaction types tend to generate radically different profiles.They also offer some evidence in favour of including as wide a variety of taskstructures as possible in this type of test of speaking as it clearly results in a widerrange of language functions and offers the candidates an opportunity to displaytheir linguistic range to a greater degree.
The final major changes to the Speaking papers are the introduction of aninterlocutor frame and a change in the way in which the scores were awarded (adifferent rating scale was used) and reported. This change will be discussed afterthe papers at the different levels are reviewed.
BEC1 and BEC Preliminary
The first part of the BEC1 Speaking paper (Table 4.19) involved a brief (approx.two minute) informal one-to-one interview between the examiner and each ofthe candidates in turn. This task did not feature input material but was unscriptedand based on personal information exchange. As such, it was problematic fromthe perspective of equivalence (each test was essentially a unique event), lack ofspecificity (there was no obvious ‘business’ context) and an associated absenceof authenticity. In the revised paper, this first part has been shortened to approx-imately one minute per candidate and is seen as an opener, designed to settle thecandidates.
Table 4.19 BEC1 Speaking paper outline
Total time allowed 10 minutes (two tasks/parts)
The second task in the BEC1 Speaking paper involved information exchangebetween the two candidates. Figure 4.15 shows an example of one of the two setsof task cards used by the candidates.
This set of cards shows clearly where the criticism of the BEC1 Speakingpaper originated. The main focus of the criticism was the lack of real interactionin performing the task. Basically, the candidates were simply asked to create aseries of three questions based on the prompts contained on the ‘YourQuestions’ card. From the example shown we can see that it would be quite easyto complete the task by converting the prompts into simple questions and forone’s interlocutor (the other candidate) to respond to these questions with
Changes in the Speaking papers
159
Structure Task Format Input Output Time Marks
1 interlocutor1 assessor2 candidates
(possible3 at end ofsession)
1 One-to-oneinterview
Oralquestions
Personal information
Agreeing, disagreeing,preferences, opinions
4–5minutes
1 – Higher
2 – Minimumsatisfactory
No Grade – lessthan satisfactory
2 Two-waycollaborativetask
Writtenpromptand spokenrubric
Interactional
Eliciting and givinginformation
4minutes
language taken primarily (or even only) from the text of the card, for example:
Question: [What] [is the] title of [the] magazine?Response: [It is called] Commercial Life.
Another worry about this type of item is the unlikelihood of any extendeddiscourse resulting from the questions asked, certainly if the candidates areexpected to stick to the information provided in their prompt cards. This intro-duces the possibility that the task can only be performed well if the individualcandidate is able to create both language and context from the prompt. In other
4 Changes in the BEC papers
160
Figure 4.15 BEC1 Speaking Part 2
Source: Cambridge ESOL BEC Handbook (2000:36–37)
words, successful performance is, to a large extent, dependent on non-languageability such as imagination/creativity or background knowledge.
Table 4.20 BEC Preliminary Speaking paper outline
Total time allowed 12 minutes (three tasks/parts)
In the revised version of the paper (outlined in Table 4.20), we can see that thesecond task is based around an individual long turn (see Figure 4.16). The profileof this task in Figure 4.14 implies that it is quite similar to the first task, though itshould be remembered that the profile only tells part of the story – thesedarkened areas simply show the expected functions in the candidate response;they are not meant to quantify the number of functions. The task involves thecandidate in a single long turn, in which they are first given one minute toprepare and then expected to produce at least one minute of continuous output.Finally, there is an opportunity for the candidate who is not speaking to ask aquestion or make a point related to what has been said and for the speaker to thenrespond. This will obviously involve the use of a broader variety of linguisticand strategic language use. The final advantage to this type of task is that it ismore clearly related to the business context than the information exchange task.Figure 4.16 also indicates that a choice of topic is available to the candidate.
Task 3 in the revised BEC Preliminary (see Figure 4.17) is a two-way (orthree-way where there are three candidates tested during one session) inter-action task, in which the candidates are introduced to the task by the interlocutor(see the interlocutor frame in Figure 4.20) and given an additional bullet-pointedprompt card (see Figure 4.17). In this task, the candidates are asked to speakfor approximately two minutes, with the interlocutor supporting the presen-tation where he or she deems it appropriate. Finally, the interlocutor may askadditional questions (again scripted) that are related to the theme of the presen-tation.
While this task type tends to lead to a broadening of the range of language
Changes in the Speaking papers
161
Structure Task Format Input Output Time Marks
1 examiner1 observer2 candidates
(possible3 at end ofsession)
1 One-to-oneinterview
Oralquestions
Personal information
Agreeing, disagreeing,preferences, opinions
2 minutes 1 mark awarded by interlocutorusing holistic scale
4 marks awarded byobserver usinganalytic scale(grammar andvocabulary;discoursemanagement;pronunciation;interactivecommunication)
2 Individuallong turn
Writtenpromptwithbulletedsuggestions
Mini presentation 5 minutes(includes1 minutepreparationtime)
3 Two-waycollaborativetask
Writtenprompt andspokenrubric
Interactional
Eliciting and givinginformation
5 minutes
functions elicited (see the discussion above), there is always a danger that theintervention of the interlocutor will reduce the interactive or conversationalnature of the event to that of an interview (with the interlocutor engaging inwhat is essentially a series of individual question-and-answer based interac-tions with each candidate in turn). This presents the developer with somethingof a conundrum; if the interlocutor is instructed not to intervene there may be acomplete breakdown in the interaction, particularly at this level. On the otherhand, this very intervention can alter the nature of the communication! As withalmost any other such decision, there is no perfect answer, and the decision hereto allow for interventions is based on the only really pragmatic solution – if theinteraction breaks down totally there is no language to base a judgement on.
4 Changes in the BEC papers
162
Figure 4.16 BEC Preliminary Speaking Part 2
Source: Cambridge ESOL BEC Preliminary, Examination Report and Past Examination Papers (2002)
BEC2 and BEC Vantage
Tables 4.21 and 4.22 show that the format has been changed at this level in thesame way.
Table 4.21 BEC 2 Speaking paper outline
Total time allowed 12 minutes (two tasks/parts)
Changes in the Speaking papers
163
Figure 4.17 BEC Preliminary Speaking Part 3
Source: Cambridge ESOL, BEC Preliminary, Examination Report and Past Examination Papers (2002)
Structure Task Format Input Language Time Marks
1 examiner1 observer2 candidates
(possible3 at end ofsession)
1 One-to-oneinterview
Oralquestions
Personal Informationexchange
Agreeing, disagreeing,preferences, opinions
3–4minutes
1 – Higher
2 – Minimumsatisfactory
No Grade – lessthan satisfactory
2 Paired task Writtenprompt andspokenrubric
Non-personalinformation transfer
Eliciting and givinginformation
7–8minutes
The task shown in Figure 4.18 (Task 2) shows that the candidate is offered achoice from a set of three semi-scaffolded task variations. These are semi-scaffolded in that there are just two bulleted suggestions included in the prompt,with an indication that other points can be added. The prompts are all designedto elicit a single long turn on one of a range of business-related topics. Onepotential problem with offering a choice, such as has been done here, is thatthere may be some options that are more difficult for candidates to achieve highscores on. While this can be addressed to a large extent in the design of the task,and in the writing of the different versions of the task through a checklist typeframework such as that suggested by O’Sullivan and Weir (2000), it is alsonecessary to empirically test for bias in the test data.
4 Changes in the BEC papers
164
Structure Task Format Input Output Time Marks
1 examiner1 observer2 candidates
(possible3 at end ofsession)
1 One-to-oneinterview
Oralquestions
Personalinformation
Giving opinions,speculating etc.
3 minutes 1 mark awardedby interlocutorusing holisticscale
4 marks awardedby observerusing analyticscale (grammarand vocabulary;discoursemanagement;pronunciation;interactivecommunication)
2 Individuallong turn
Writtenprompt withbulletedsuggestions
Mini presentation
Giving informationand justifyingopinions
5 minutes(includes1 minutepreparationtime)
3 Two-waycollaborativetask + follow-updiscussion
Writtenprompt andspoken rubric
Oral promptfor follow-updiscussion
Interactional
Eliciting andgiving information,justifying opinions,makingcomparisons,agreeing anddisagreeing etc.
5 minutes
Table 4.22 BEC Vantage Speaking paper outline
Total time allowed 14 minutes (three tasks/parts)
Figure 4.18 BEC Vantage Speaking Part 2
Source: Cambridge ESOL BEC Vantage, Examination Report and Past Examination Papers (2002:42)
Changes in the Speaking papers
165
A: WHAT IS IMPORTANT WHEN...?
Entertaining clients
• Types of activities
• Cost
•
•
B: WHAT IS IMPORTANT WHEN...?
Choosing retail premises to rent
• Location
• Length of contract
•
•
C: WHAT IS IMPORTANT WHEN...?
Deciding on packaging for products
• Image
• Production process
•
•
Task Card 17
BEC 3 and BEC Higher
Table 4.23 BEC 3 Speaking paper outline
Total time allowed 14 minutes (three tasks/parts)
Table 4.24 BEC Higher Speaking paper outline
Total time allowed 14 minutes (three tasks/parts)
Figure 4.19 shows the task cards for Part 3 of the BEC Higher. In this task, whichis designed to elicit a sample of interaction-based language, candidates areallowed thirty seconds to read the task card and are then expected to speak for
4 Changes in the BEC papers
166
Structure Task Format Input Output Time Marks
1 examiner1 observer2 candidates
(possible3 at end ofsession)
1 One-to-oneinterview
Oralquestions
PersonalInformationexchangeExpressing opinions
3 minutes 1 mark awardedby interlocutorusing holisticscale
4 marks awardedby observerusing analyticscale (grammarand vocabulary;discoursemanagement;pronunciation;interactivecommunication)
2 Individuallong turn
Writtenprompt withbulletedsuggestions
Mini presentation
Giving informationand justifyingopinions
6 minutes(includes1 minutepreparationtime)
3 Two-waycollaborativetask + follow-updiscussion
Writtenprompt andspoken rubric
Oral promptfor follow-updiscussion
Interactional
Eliciting and givinginformation,justifying opinions,makingcomparisons,agreeing anddisagreeing etc.
7 minutes
Structure Task Format Input Language Time Marks
1 examiner1 observer2 candidates
(possible3 at end ofsession)
1 One-to-oneinterview
Oralquestions
Personal Informationexchange
Expressing opinions
3–4minutes
1 – Higher
2 – Minimumsatisfactory
No Grade – lessthan satisfactory2 Paired
taskWrittenprompt andspoken rubric
Non-personalinformation exchange
Explaining, persuading,justifying, etc.
4 minutes
3 Individuallong turn
Writtenprompt andspoken rubric
Monologue (based onwritten input)
Describing, explaininggiving and justifyingopinions, etc.
6 minutes
approximately three minutes. The topics are clearly business-focused, and canbe realistically expected to elicit the sort of profile outlined in Figure 4.14 – witha range of language functions across the three types.
In order to deal with the situation where there are three candidates present, thetask has been added to slightly – with an additional element in the expectedoutcome, see Task 26 in Figure 4.19. There is a potential danger here that thelanguage elicited under the two conditions may be different, as the two condi-tions involve both different numbers of candidates and different expectedoutcomes. However, there is no evidence that candidates involved in paired orthree-way interactions are biased either towards or against – the format has beensuccessfully used for almost a decade in the Cambridge ESOL Main Suiteexaminations and has been adopted in other tests around the world. It is certainlyan area in which further research is required in order to ensure that there is nounintentional bias present in the Speaking papers of the revised BEC.
Figure 4.19 BEC Higher Speaking Part 3
Source: Cambridge ESOL BEC Higher, Examination Report and Past Examination Papers (2002:47–48)
Changes in the Speaking papers
167
There are obvious advantages to the inclusion of the different task types in theBEC suite. These can be summarised as adding to the test in terms of:
• authenticity – since presentations and peer discussion and decision-making are seen as being of particular relevance to thearea of business
• specificity – the inclusion of these tasks has the effect of making thepaper more clearly specific to the language use domain
• generalisability – as the task introduces the potential for a wider varietyof language function use (see O’Sullivan, Weir andSaville 2002).
Other changes to the Speaking paperThe other major changes to the Speaking papers are the use of an ‘interlocutorframe’ and the way in which the performances are scored.
The introduction of an interlocutor frame
In the earlier versions of the BEC examinations, the interlocutor frame as wenow know it was not used. However, work carried out in the early 1990s, partic-
4 Changes in the BEC papers
168
Figure 4.20 BEC Higher Speaking Part 3 – interlocutor frame
ularly that of Lazaraton (1992, 1996), suggested that the lack of control over thelanguage input was having a measurable impact on the performance of candi-dates in oral interview type tests.
The introduction of the scripted interlocutor frame allows the test developerto more fully control what is happening in the test event. The frame is a scriptedtext which guides the examiner through the event, limits the examiner in termsof input (ensuring that all candidates receive the same directions) and timing(guaranteeing that all candidates will have an opportunity to perform all of thetasks provided in the test) – see Figure 4.20 for a copy of the frame that goes withthe two candidate version of the task described in Figure 4.19.
We can see from this example that there are times in which clarification maybe offered at the discretion of the examiner. Clearly, it would be unwise not toallow for some flexibility as all test events will be in some way different, andcandidates of different ability will require more or less help from the examiner.The advantage of allowing the examiner a choice in the follow up questions
Other changes to the Speaking paper
169
Figure 4.20 BEC Higher Speaking Part 3 – interlocutor frame (continued)
Source: Cambridge ESOL BEC Vantage, Examination Report and Past Examination Papers(2002:53–54)
means that the questions can be chosen to engage with the candidates’ output,while still allowing for the control over the event vital to reliability.
Of interest here are the findings of a research study undertaken by O’Sullivanand Lu (2003), who investigated the impact on candidates’ linguisticperformance of deviations from the interlocutor frame by examiners in theIELTS Speaking test. By making comparisons between transcribed segments ofthe output of learners taken before and after the deviation, they found that therewas no significant impact on a number of measures (discourse features,linguistic accuracy and complexity or fluency). The indication is that as long asthe interlocutor maintains the integrity of the test through a systematic, thoughnot dogmatic, use of the interlocutor frame, there will be little perceptible impacton the output of the candidates, thus supporting the decision not to make theinterlocutor frame so tight as to eliminate any individual expression on the partof the examiner.
While the Cambridge ESOL move to the paired format has been criticised(Foot 1999) the anecdotal nature of the criticism when coupled with the failureto make a realistic critique of the limitations of the one-to-one interview formatlimit the value of this criticism. The other major limitation of the criticism wasthe lack of awareness of the model of language competence which lay behind themove, a problem possibly caused or certainly exacerbated by the lack ofpublished information on the construct at that time. In their response to thecriticism, Saville and Hargreaves (1999) provided a well argued rationale for theformat, demonstrating the essential weakness in any test design that did notencourage interactive communication – though it should also be pointed out thatthis strength may also be a weakness, as it is now accepted that the nature of anylanguage of the interaction is co-constructed.
The introduction of a new rating and reporting scale
This original rating procedure has been replaced with a scale which is moretypical of the Cambridge ESOL examinations. While the focus is still on thesame criteria as were used in the pre-revision tests (grammar and vocabulary,discourse management, pronunciation and interactive communication), thedescriptors have been revised and rewritten to reflect the descriptors used in theequivalent Main Suite tests. Using the rating scales in the same way as they areused in the Main Suite, also minimises any negative effect that using a verydifferent type of scale might have on the examiners. Where examiners use afamiliar scale, and are making judgements at a level with which they arefamiliar, there is a far greater likelihood that they will be consistent than if theyare asked to use very different scales for each examination they are asked to rate.On the negative side of this is the argument that the rating scales lack anyspecific business domain orientation. This means that the aspects of language
4 Changes in the BEC papers
170
which distinguish the business domain (for example appropriacy of lexis,register, format, and rhetorical structure) are not taken into account, thusquestioning the potential of this aspect of the test to tell us about the candidates’ability to perform linguistically in the business domain.
Since the arrangement of the Speaking test now more closely resembles thatof the Main Suite examinations in terms of task type and assessment type, moreaccurate and meaningful comparisons between the Main Suite examinationsand external criteria such as the Common European Framework and the threeBEC levels can be made (see Figure 4.21 – which represents a schematicdiagram of the format of the Main Suite and BEC Speaking tests).
Figure 4.21 Structure of the revised BEC Speaking paper
SummaryIn this chapter I have tried to outline briefly the changes made to the individualpapers in the BEC suite. As you can see from the above, the most significantchanges have come as a response to criticisms of the Speaking paper, while theother papers that were considered to be working well were left relativelyuntouched. This is in line with the ‘continuity and change’ dimensions referredto in Weir and Milanovic (2003).
Before concluding the chapter, it might be useful to review the BEC exami-nations in terms of the criteria used to review the other tests in Chapter 1. Thiswill allow the reader to make comparative judgements on the different tests andwill, I hope, demonstrate how the value or usefulness of the BEC suite has beenincreased with this revision.
As the first two of the criteria (a brief introduction to and description of thetest) have been dealt with in this chapter, I will focus on the remaining criteria inthe brief review contained in Table 4.25. As can be seen from this overview, thechanges to BEC have resulted in some areas of significant improvement, and inother areas of similar performance. Even these areas of relatively little or no
Summary
171
Performance
Performance
INTERLOCUTOR
OBSERVER
CANDIDATE B
CANDIDATE A
Holistic Score
Analytical Scores
Grammar & VocabularyDiscourse ManagementPronunciationInteractive Communication
Table 4.25 Brief overview of the old and revised BEC suite
change are relevant however, as the lack of change is not due to any inertia, butis based on a thorough review of the entire test system. In fact, this suggests afurther criterion for test evaluation, that of systematic self-monitoring. In thecase of the BEC suite (as in other Cambridge ESOL examinations) this constantmonitoring and revising of tests is a feature which seems to ensure that the tests
4 Changes in the BEC papers
172
Original BEC Revised BEC
3. An outline ofthe constructupon whichthe testfocuses
Not clearly defined, though appearsto have been based on acommunicative, four skills’definition of the construct.
Now more explicitly designed to reflect themulti-componential approach to competence astypified in Bachman’s (1990) model.
4. The testmethod
A variety of task and item formatsare used throughout the differentpapers.
A variety of task and item formats are usedthroughout the different papers.
5. Skills’coverage
Listening, Speaking, Reading andWriting
Listening, Speaking, Reading and Writing
6. Measurementqualities
Not available in the public domain. Reported here in Chapter 2.
7. Degree ofspecificity/Authenticity
Relatively specific, though in thecase of the Speaking paper,criticised for being too general.
In general meets any ‘authenticity’criticism, with the exception of theSpeaking paper.
Retains former degree of specificity, thoughrevised Speaking paper is much more specific.
Tasks across the different papers attempt toengage concepts of both situational andinteractional authenticity.
8. Impact ofnon-languagefactors
Seem to be unproblematic, thoughthe use of a single interlocutor in theSpeaking paper may heighten theimpact of any ‘interlocutor’ effect.
Likelihood that backgroundknowledge of the business languagedomain might impact to somedegree on performance in somepapers.
Seems to be unproblematic – potential problemin Speaking paper rectified through theintroduction of interaction (with peer andexaminer) and monologic discourse.
The additional specificity in the Speaking papersuggests that background knowledge mayimpact on performance.
9. Reportingof testperformance
Candidates received one grade forReading, Writing and Listening (A,B or C at levels BEC2 and 3, and A,B, C or D at BEC1). As theSpeaking paper was considered aseparate entity a separate grade wasawarded (1 or 2).
All candidates receive an overall estimate oftheir ability based on their performance on eachof the four papers (each is worth 25% of the totalavailable marks).
For BEC Preliminary, results are reported as aPass with Merit or a Pass or as one of two failinggrades – Narrow Fail or Fail. At the other levelsthere are three passing grades (A, B or C) andtwo failing grades (D or E).
The certificates for all three levels also include agraphical profile (see Figure 3.1). This profile isof particular diagnostic value to the candidate –indicating areas of strength and/or weakness.
are continuously being brought up-to-date to reflect changing views of whatlanguage ability really consists of, and of how it might best be assessed. It alsoallows for a test to be constantly monitored for appropriacy as the candidate basechanges over time or where the uses of the test evolve.
Summary
173
174
Conclusions and the wayforward
SummaryIn the first chapter of this book, I reviewed the brief history of the testing oflanguage for business purposes (TLBP). This review demonstrated the relative‘newness’ of the area and highlighted the tendency for these tests to be‘industry-driven’ with a more pragmatic than theoretical foundation. The onlytheoretical perspective that has gained recognition is that of Douglas(2000:281), who sees language for specific purpose (LSP) tests as beingpremised on the fact that language performance varies between specific contextsand that the language of these specific contexts is precise, that it is distin-guishable from other language use contexts or domains. Criticism of thisdefinition (Elder 2001) focused on the fact that there were three areas in whichLSP tests were problematic. These were:
• specificity • authenticity • impact of non-language factors.
Before going on to look at current practice in the area of business languagetesting, I offered a somewhat different perspective on LSP tests and the abovecriticism. In this perspective, I suggested that there were four key points thatshould be taken into account when theorising on LSP testing in general. Thesewere
1. As all tests are in some way ‘specific’, it is best to think of language tests asbeing placed somewhere on a continuum of specificity, from the broadgeneral purpose test (such as CPE) to the highly specific test.
2. Very highly specific tests tend to be very poor in terms of generalisability,while the opposite can be said of non-specific (or general proficiency) tests,though this is not a binary choice if we accept that tests can be developedalong a specificity continuum.
3. Where a test is situated closer and closer to the more highly specified end ofthe continuum, the focus on authenticity also changes.
4. The more highly specific a language test is the more it entails a focus on theevent rather than on the language of the event. The degree to which non-
5
Summary
175
language factors impact on a candidate’s test performance will reflect thedegree of specificity of that test. Therefore, in a highly specific language testit may not be possible to separate the language from the specific event.
Following this, a review of currently available small and large scale tests (interms of test-taking population) was undertaken. The tests reviewed werelooked at from a 9-point perspective based on an overview of the theoretical andpractical issues, these were:
1. A brief introduction to the test.2. A brief description of the test.3. An outline of the construct upon which the test is based.4. The test method.5. Skills’ coverage.6. Measurement qualities.7. Degree of specificity/authenticity.8. Impact of non-language factors.9. Reporting of test performance.
From these reviews it became clear that the practical operationalisation of theTLBP concept appears to be quite uneven in some regards, certainly in terms ofthe availability of research and/or support material.
Some interesting points can be taken from these reviews
• Large-scale tests tend to have originally been produced at the behest ofgovernment agencies (though the trend is that international tests are beingproduced more and more to meet either perceived or established marketneeds – in other words the TLBPs are more and more market driven).
• As the markets (and the test-taking population) change there is little signthat the tests have been revised to meet the change, and where change hascome, there has been no information on that change made available in thepublic domain.
• There has not really been a tendency for changes in proficiency languagetesting practice to be reflected in TLBP practice with regards to context-based, theory-based and scoring validity.
• Few TLBPs include papers related to the four skills of speaking, writing,reading and listening.
• There appears to be a relatively low level of support material available,though the UK-based tests tend to offer practice or past papers at no chargeto test takers – these can usually be downloaded from the web (see theReferences section at the end of this book).
• There is a clear tendency against very highly specific tests, for example atest of language for chartered accountants. Instead, the tests on the marketappear to be more general in nature, context-oriented rather than context-focused.
5 Conclusions and the way forward
176
The overview then showed how the review of the Cambridge ESOL BusinessEnglish Certificate (BEC) suite offered an interesting insight into how sucha large scale test system might be revised. The main points highlighted bythese chapters are the complexity of the process, for example in demonstratinghow different stakeholders’ views were taken into account (in different ways),the need for tests to take into account changes in language testing theoriesand the reinforcement of an all-skills approach.
The implications of this work are twofold, theoretical and practical. Theformer, is of interest to LSP testing in general, while the latter will focusprimarily on the BEC suite of tests though will identify areas of LSP in generalthat might benefit.
Theoretical implicationsDouglas essentially sees authenticity as the central issue in his definition of LSPtests:
. . . a specific purpose language test is one in which test content and methodsare derived from an analysis of a specific purpose language use situation, sothat test tasks allow for an interaction between the test taker’s languageability, on the one hand, and the test tasks on the other (2000:90).
In the first chapter of this book I suggested that it may be better to see thisas just one aspect of LSP tests, and instead argue that other central issues werethe potential for distinguishing language use in a specific situation and, fromthe operational perspective, the assessment or evaluation of the performance.Of the points made in Chapter 1, and reiterated above, I would now like to revisitthe notion of degree of specificity, because it appears to me to be at the heartof the issue.
In Figures 1.5 and 1.31, I suggested that all tests lie on a ‘degree of specificity’continuum. Reflecting now on that suggestion, having reviewed both the liter-ature and current practice in business language testing, I see that it seems tooversimplify the situation. In actual fact, there are a number of elements which combine to help us draw inferences as to the degree of specificity of anLSP test. These elements are related to such concepts as authenticity as well asgeneralisability and distinguishability.
In order to explain what I mean by this we need to go back to the criticismmade by Davies (2001) of the lack of a theoretical basis for LSP testing. In hispaper, Davies argues that it is not possible to fully distinguish specific languageuse domains. The point to be made here is that, by its very nature, language is noteasily defined, and the language of a specific use domain is no different. Withinany such situation there will be a specific ‘core’ language, which may refer to aspecific use of language or a specific lexicon – see for example the work of Ball(2002) in using a corpus linguistics approach to producing a series of updated
Theoretical implications
177
wordlists for the BEC suite. As Davies and Elder argue, there is no distinctboundary between this core and what I have labelled here the general languageuse domain. Instead, there is an area of transition, in which language use isshared with other domains. Figure 5.1 attempts to graphically represent thisnotion, though it is limited to two dimensions, while the actual situation shouldbe visualised as being multi-dimensional.
Figure 5.1 The notion of core and general language use domains
Taking this idea a step further, Figure 5.2, we can now see that the notion ofdegree of specificity brings with it the related notions of generalisability andsituational authenticity. Where a test is seen to be positioned towards thespecific end of the continuum, the potential for generalisation from testperformance beyond the specific situation is reduced – it is difficult to imagine atest that could be placed at the extreme end of the continuum as this would befocused only on a very limited ‘core’ language. In the same way, that test wouldbe seen as being more situationally authentic were it manipulated to move it evercloser to the specific end of the continuum. A completely specific language testwould therefore be focused only on language unique to a specific use domainand would be tested in use within that domain. Performance on the test couldthen be related only to that domain. This is clearly neither practical nordesirable.
Figure 5.2 is again limited by my ability to represent the notion of general andcore in anything but a two dimensional diagram. In reality, once we movebeyond the distinguishable core we are in the domain of general language use –the figure implies that only a part of this domain can be represented in the testsample. What is successfully represented in the figure is the idea that when a testis more ‘specific’ in its focus, the greater will be the importance of the core andwhen as the test is more ‘general’ in focus, a less important role will be played bylanguage from a specific core. The question we must again ask is how do weknow that a test is either specific or general in focus?
General Language Use Domain
CORE
5 Conclusions and the way forward
178
Figure 5.2 Extension of the ‘specificity continuum’ concept
Locating specificity
The notion of specificity, if it is to be of practical use to the test developer, mustbe tied to an understanding of test validity. One such perception of testvalidation is suggested by the socio-cognitive spoken language frameworkdiscussed by O’Sullivan and Weir (2002) and developed as a series of frame-works for all four skills by Weir (2004). In these frameworks, validity is seenfrom a socio-cognitive perspective – a perspective which appears similar to thatsuggested by Chalhoub-Deville (2003) and Chapelle (1998). In the followingexample, I will refer to the framework developed for validating tests ofspeaking, though any of the other three frameworks would obviously workequally well.
Figure 5.3 gives an idea of what the entire framework looks like. In thisoutline, we can see that there are a number of elements, each of which should beattended to by the test developer. Evidence is required at each level, in order tomake validity claims for a test. I have added to the framework by highlightingthe fact that the test taker can be described in terms of a number of characteristics
MORE
MORE
LESS
LESS
Specific Core
SPECIFIC
SITUATIONALLY AUTHENTIC
GENERALISABLE
. .
..
General LanguageUse Domain
Theoretical implications
179
(physical/physiological; psychological and experiential) and by the internalprocessing (unique to the individual) which takes place during test performance.The test can be described in terms of its context validity and in terms of thepotential for successful test tasks to result in appropriate processing. It is thisnotion of what Weir (2004) calls theory-based validity that forms the linkbetween the test and the test taker.
Figure 5.3 Format of validation frames
Source: based on Weir (2004)
Since I am hoping to provide a theoretical basis for LSP tests in general andbusiness language tests in particular, I will now demonstrate how I feel theabove framework can be shown to be related to the two aspects of authenticity.
Context validity (see Figure 5.4), is concerned with aspects of the demands ofthe task and text, as well as detailing the test setting. In terms of the view of LSPtests offered here, it should become clear that when we are talking about testspecificity, we are actually referring to test context, and this is expressed in theframework as being comprised of task and text demands.
When we consider the difficulty in defining language proficiency and use (forexample the ‘boundary’ issue raised by Davies 2001 and Elder 2001) we can seethat context validation is always going to be problematic. The operations andconditions suggested in the framework presented here are based on Weir (1993)and have been used with some success in test development projects for a decade,though they remain tentative in that there is no empirical evidence that these are
The Test Taker
The Test
Criterion-Related Validity
ConsequentialValidity
ContextValidity
Responses
Scoring Validity
Theory-BasedValidity
Test TakerCharacteristics
Score/Grade
5 Conclusions and the way forward
180
the only operations and conditions applicable to a test of speaking (see Weir2004 for a more detailed and updated version of the frameworks).
Figure 5.4 Aspects of context validity for speaking
Source: Weir (2004)
Test specificity might therefore be expressed as the degree to which the opera-tionalisation of each of these demands can be considered to be uniquely relatedto a specific language use domain. In practice, this entails making value judge-ments of the degree of specificity along a continuum for each aspect of both taskdemands and text demands (see Figure 5.5). This may be seen as being toosubjective a task to be of practical use. However, the real value of the exercise isin the breadth of the exercise. Specificity is now seen as a multi-dimensionalperspective of a test, and judgements are at least being made on a systematicbasis; a criticism of my early reviews of the various tests is that the judgementswere essentially intuitive and, as no systematic approach was taken, thisintuition may not always have been based on similar criteria (the same criticismcan be made of almost all multiple-test reviews).
In order to demonstrate this, I undertook a small experiment in which a groupof language specialists was asked to take two test papers (of Reading) and tomake judgements on the papers based on a simple Likert scales’-basedinstrument. The instructions to the specialists asked that they should try todecide where on the scales (one scale for each of the aspects of context validityshown in Figure 5.6) each of the two papers might hypothetically lie, with
Theoretical implications
181
1 meaning ‘very specific’ and 7 ‘general’ – where an aspect was considered‘neutral’ it was decided that a rating of 4 should be awarded. The papers weretaken from an LSP test, BEC Vantage, and a general proficiency test, the FCE, asthese two tests are designed to allow for inferences to be made at the same CEFlevel – Level B2. Figure 5.6 shows that there were clear differences seen by thespecialists in terms of the task demands. This clearly different profile can betaken as empirical evidence of the distinguishability of LSP and general tests.
Figure 5.6 Differences in task demands between LSP and generalproficiency test papers
Figure 5.5 A multi-componential view of specificity
100%Gener
100%Gener
100%General
100%General
100%Specific
Task and Text Demands
Points on eachcontiniuum
Purpose
Response FormatResponse Format
Weighting Weighting
Known Criteria Known CriteriaOrder of Items Order of Items
Time Constraints Time Constraints
Intended Operations Intended Operations
Purpose
Task Demands LSP Task Demands General
5 Conclusions and the way forward
182
When the participants were asked to repeat the exercise for the same papers butthis time with a focus on text demands, the differences are even more obvious(Figure 5.7).
Figure 5.7 Differences in text demands between LSP and generalproficiency test papers
The evidence from this admittedly very small study, suggests that judgementson the degree of specificity of an LSP test can be made in a systematic way. Italso suggests that the notion of test specificity is closely linked to that of situa-tional authenticity. Of course it could be argued that even a supposedly‘specific’ test such as BEC, or even a ‘highly specific’ test, such as the one for airtraffic controllers described by Teasdale (1994), can never reach a positionwhere the shaded area in the figures is minimised – indicating that the test hasachieved a high degree of specificity from all perspectives. The evidence heresupports the view that tests can never hope to do more than simulate authen-ticity, and intuition suggests that this same evidence will be found where othertests are analysed using the methodology suggested here – however highlyspecific the test developer claims it to be.
The second aspect of validity I will look at is that of theory-based validity (seeWeir 2004:Chapter 1), which is concerned with the cognitive processing duringtest task performance. Test validity, from this perspective, is thereforeconcerned with the degree to which the processing in the test situation reflectsthat of the language use domain. While this perspective on test validity isrelatively new (though see both Chapelle 1998 and Douglas 2000 for argumentsthat lend support to this view of validity) there is encouraging evidence fromongoing research in China and Malaysia that evidence can be elicited to supportthe making of comparative judgements between different test task types.
The symbiotic nature of the relationship between content and theory-basedvalidity can be illustrated by showing how decisions taken with regard toelements of context validity have significant effects on the cognitive processingof test takers who must perform the tasks in the test situation. An example of thiscan be found in Porter and O’Sullivan (1999) who demonstrated that by
LinguisticChannel
Discourse Mode
Text Length
Writer/Reader Relationship
Nature of Information
Topic Familiarity
Lexical Range
Structural Range
Functional Range
LinguisticChannel
Discourse Mode
Text Length
Writer/Reader RelationshipNature of Information
Topic Familiarity
Lexical Range
Structural Range
Functional Range
Task Demands GeneralTask Demands LSP
Theoretical implications
183
changing the description of the addressee of letters written by Japanese EFLlearners, significant changes were observed in terms of both orthography andlanguage use, suggesting that there had been a significant impact on the goal-setting part of the executive processing dimension of written production.
If we show that the cognitive processes involved in an LSP test taskperformance reflect those of the specific language use domain they are designedto reflect, then we can claim with some confidence that our test task demon-strates interactional authenticity. It is quite possible that such processing maywell differ in important respects from general purpose task performance, forexample in the recourse to different areas of executive resources. By demon-strating differences in internal processing between LSP and general purposetasks we are offering an additional argument in favour of the distinguishabilityof language use domains.
This is obviously pertinent to all areas of language testing, supporting, forexample Bachman and Palmer (1996:23) who see authenticity as ‘the degree ofcorrespondence of a given language test to the features of the TLU (targetlanguage use) task’. The argument here is that we need to go beyond the notionof content validity and situational authenticity to include a working perspectiveon the interactional aspect of authenticity. This way of looking at validity offersjust that perspective.
While it is relatively straightforward to establish the situational authenticityof a test task, it is only by an a posteriori empirical exploration of testperformance that evidence of the interactional authenticity of any test can beestablished.
In a way, this brings us back full circle to the definition of LSP tests offered byDouglas, and quoted above. When I quoted Douglas’s definition, I suggestedthat there were limitations to it; however, when these limitations are seen in the
Figure 5.8 Aspects of theory-based validity for speaking
THEORY-BASED VALIDITY
INTERNAL PROCESSES
• Conceptualiser
• Pre verbal message
• Linguistic formulator
• Phonetic plan
• Articulator
• Overt speech
• Audition
• Speech comprehension
MONITORIN
G
EXECUTIVE RESOURCES
• Content knowledge
• Internal
• External
• Language knowledge
• Grammatical
• Discoursal
• Functional
• Sociolinguistic
Source: Weir (2004)
5 Conclusions and the way forward
184
light of the twin perspectives of authenticity, we can see that my argumentsactually support the definition – though, I hope, adding to it so that the criticismsvoiced by Davies, Elder and Douglas himself can be, to some considerableextent, rejected.
Practical implicationsBefore concluding, I would like to first suggest a series of practical implicationsfor the BEC tests (and other tests of language for specific purposes). There aresome implications that are relevant to all papers and others that are specific toeach of the four.
Reliability
The issue of reliability is of great importance to all test developers. As we saw inChapter 2 of this book, there are problems associated with the way in whichinternal consistency is estimated for tests with truncated populations (existingprocedures can result in low estimates as there tends to be restricted variationwithin the population), and also with the way in which reliability is estimated fortests based on performance such as writing and speaking. Weir (2003b:475)suggests a number of alternatives to the current practice in estimating reliabilitywhere a truncated population is involved, as does Luoma (2004:183) whoargues that the Standard Error of Measurement (SEM) should be routinelyreported ‘as a useful quality check’. SEM may serve as a practical device whenit comes to dichotomously scored test items, and the internal consistencyestimates and SEM reported earlier for the BEC suite are quite satisfactory.However, the fact that SEM is premised on being able to accurately estimateinternal consistency means that there is an even greater difficulty with applyingthe formula to a speaking or writing test. One reason for this is that the mostcommon reported procedure for estimating inter-rater reliability, correlation, isproblematic for a number of reasons:
• There is some concern that the intervals represented in rating scales are notequal. This suggests that the researcher/tester should use the SpearmanRank Order Correlation statistic – a less powerful non-parametric estimateof association than Pearson’s Rho.
• Correlation statistics can only tell us about the association between thepattern of scoring of raters, not between their level of agreement. So thecorrelation between the scores awarded by a very harsh rater and a verylenient rater will still be very high if for example the scores they award placethe test takers in the same order.
• The correlation between two variables may be due to the impact of other,unobserved variables.
Practical implications
185
• Even if there are two raters who agree totally, it is possible that they areboth either very harsh or lenient – or both inconsistent, but in a similarway.
While I agree that alternatives should be sought, I feel, as does Weir, that thoseoffered to date are simply a short term solution. In the longer term we really needto look more closely at the whole area of reliability. Until recently, it wasconsidered acceptable to discuss test validity in somewhat simplistic terms andit was generally considered acceptable that evidence need only be gathered inrelation to a single aspect of validity. Indeed, in the psychology literature it isstill common to find a single numerical estimate of test validity. The same viewof reliability is common today – with the reporting of a single internal consis-tency coefficient considered adequate evidence of the reliability of a test.
The format of the validation frames (Figure 5.3), offers, I believe, the basisfor a more viable alternative to existing practice. Here, we can see that the notionof reliability has been replaced with the more helpful concept of ‘scoringvalidity’.
Bachman (1990:163–166) argues that test scores are affected by a numberof factors: the communicative language ability of the candidate; test methodfacets (systematic aspects of test delivery for example); personal attributesof the candidate (both individual and group characteristics); and finallyrandom factors (unpredictable and unsystematic factors that impact on testperformance). Consideration of the reliability of a test would be greatlyimproved by conceptualising reliability in terms of these different factors. Weir(2004) suggests that, for tests of reading comprehension, for example, theframework should consist of those elements contained in Figure 5.9.
Figure 5.9 Aspects of scoring validity in tests of reading
Source: Weir (2004)
While this figure is very similar to the suggestions made by Bachman (1990),when we look at the possible aspects of scoring validity for a test of speaking(Figure 5.10) the situation is now quite a lot more complex with the inclusion ofa number of new elements in the equation, in particular the rating scale, therating process and the rater. When we consider the discussion of rating scale
5 Conclusions and the way forward
186
development and use in Chapter 1, we can see that this is an area of someinterest. In addition, the whole rating process is still relatively unexplored; wecontinue to know far too little about the effects of rater training (though seeWeigle 1994, and Rethinasamy in progress), or of the value of standardisation(see O’Sullivan and Rignall 2001, 2002) or of what happens in the minds of theraters when they are awarding scores (see Lumley 2000).
As for the other areas referred to in Figure 5.10, there is little or no empiricalevidence of how they impact on rating performance. O’Sullivan (2002b) arguedthat the test taker should be described in terms of a series of characteristics(physical, psychological, experiential) and that research should be carried outinto the effect on performance of the interaction between these variables and thetest event. In the same way that characteristics of the test taker will influence testtask performance, it is clear that similar characteristics of the examiner or raterwill affect their performance in awarding scores for those performances. Thereis evidence of how performances can achieve very different scores dependingon the examiner (see for example Congdon and McQueen 2000, Engelhard1994, Fisher 1994, Lamprianou and Pillas 2003, Longford 1994, Lumley,Lynch and McNamara 1994, Lunz and Stahl 1990, Lunz, Wright and Linacre1990, Myford and Wolfe 2002) but relatively little that I could find on howcharacteristics of the rater might have some impact on rating performance(though see Lumley 2000, Lumley and McNamara 1995, McNamara andLumley 1997, O’Sullivan 1999, 2000a, 2002).
Figure 5.10 Aspects of reliability or measurement validity for speaking
Source: Weir (2004)
Computers
Another area in which there has been a great amount of interest over the pastdecade in particular is in the delivery of tests using computers. There are a
Practical implications
187
number of issues here that remain somewhat unexplored. In a debate on the useof computers in language testing at the IATEFL conference in Brighton (O’Sul-livan 2003) I talked about the fact that computers are used as part of the devel-opmental process, as a delivery mechanism, and as an analysis tool. Thepotential of computers to positively impact on test development has beenreferred to elsewhere (for example Chalhoub-Deville 1999, 2001, Brown 1997),while the growing importance of complex statistical analysis tools (such as G-Theory, Multi-faceted Rasch and other IRT-based procedures) would be incon-ceivable without computers and the programs that make their application apractical consideration.
It is in the area of test delivery that the impact of computers has been mostdisappointing. While tests such as BULATS, described in Chapter 1, are at thecutting edge of computer-delivery of language tests, even here there is atendency to limit the test to items that are reminiscent of the old psychometric–structuralist era, that is the test items tend to focus on discretedecontextualised aspects of language use. In fact, there are very few, if any,examples in the literature of really new or innovative item types. One exampleof how computer delivery of tests of reading comprehension could add to ourunderstanding of the reading construct is the relative ease of designing deliveryplatforms that allow for an element of timing of tasks, thus allowing thedeveloper to introduce the concept of expeditious reading (see the followingsection).
In my conclusion to the IATEFL debate I suggested that as language testerswe should not be overly dazzled by the technology and look beneath the deliverymechanisms to the underlying tests. I also argued, perhaps a little unkindly, thatthe tests delivered using computers represented a step backwards in terms of theapproach to testing they typically represent. While many tests clearly fall intothis category, there are a small number in which efforts have been made to cometo terms with the new technology, though the great leap forward first promisedby the introduction of computers has not happened.
In the case of testing language for business purposes, for example, we haveseen how technology has had a profound impact on the revision of the test –through the impact of the ALTE ‘Can Do’ statements’ projects (in helping todefine the levels of the tests) and in the project undertaken to update thebusiness-related wordlists which help define the language use domain which istested in the BEC suite (Ball 2002).
However, when we consider the notion of theory-based validity it is obviousthat there is a great deal of research needed into the degree to which different testdelivery mechanisms (e.g. pencil and paper and computer) impact on thecognitive and meta-cognitive processing of test candidates. In other words, doesthe platform affect the interactional authenticity of the task? This work has onlyjust begun, with O’Sullivan and Weir (2003) investigating this area in terms ofdelivering a writing test on computer as compared with the more traditionalpencil and paper.
5 Conclusions and the way forward
188
Suggestions for future research in specific skills’ testing
In the following section I will outline some of the possible areas for research inthe testing of the different skills in the area of business language. In this sectionI will be referring to the BEC suite. However, I believe that the suggestions willbe equally applicable to other test systems.
ReadingWhile there were relatively few changes made to the Reading papers in thisrevision, this is not to say that there is no additional work needed in the area.Weir (2004) argues for research into the impact of (and need for) expeditiousreading (skimming for gist, search reading or scanning for specifics) oncandidate performance and though he was looking at the situation from theperspective of general proficiency testing in the revision of the CambridgeESOL Certificate of Proficiency in English (CPE), the same argument can bemade for LSP tests such as BEC. Before this can begin, it is first necessary tolook again at the kind of reading undertaken in the domain of business languagethrough continued investigation using needs’ analysis techniques such as obser-vation and interviews. The indications from this book are that we shouldadditionally attempt to investigate the cognitive processing associated withthese different activities in order to help us gain evidence of the interactionalauthenticity of future test versions.
WritingAs we saw in Chapter 4, the section related to the changes in the Writing papersindicated that there have been a number of quite significant changes. Thesechanges include the decision to separate the reading and writing skills in BECVantage and Higher, and the choice now offered at BEC Higher for the secondwriting task. There is also an increase in the length of the required output for theinitial writing task at all levels and the use of a common General ImpressionMark Scheme (GIMS). This GIMS is interpreted at different performance levelsand tied to the Cambridge/ALTE levels.
Obviously all of these changes require monitoring and evaluation. Forexample, in offering a choice of task to the candidates at BEC Higher it isnecessary to establish a systematic framework for ensuring that the choices arelikely to be equivalent, to routinely trial these different choices, and to establishmonitoring systems to ensure that candidates are not negatively affected by theirchoice – it is possible, for example, that some choices may be inherently moreproblematic than others. There is also a clear need to ensure that raters see eachtask as being equal (through training, monitoring of performance and research).There is evidence from speaking test research (McNamara and Lumley 1997)that raters compensate candidates for what they see as poor performance byexaminers where they feel that the candidate has been negatively affected; it is
Practical implications
189
also very much worthwhile exploring the impact on scores of raters’ perceptionof the different writing tasks.
The use of the GIMS at the different test levels is also an interesting devel-opment. The difficulty with devising and applying rating scales within a testsuite which tests at different levels was briefly discussed above. The solutionadapted by Cambridge ESOL for the BEC suite marks an effort to standardisethe way in which raters come to decisions at the different levels (a vital elementof internal consistency – intra-rater reliability) and moves the focus away fromthe individual rater to the training and standardisation procedures. There areworries about this approach, however. The argument that a specific purpose testwill, by its very definition, involve elements of both language ability anddomain ability, implies that any rating scale devised for use in this type of testmust include some reference to performance in the domain. The ‘strong’ view ofthis argument calls for the sort of indigenous scale developed by Abdul-Raof(2002), and while the BEC tests reviewed here would not appear to justify sucha scale (as the test is context-oriented rather than context-focused) it may well bethat the existing scale does not capture important elements of the performance incontext (i.e. of the business language domain). It is a matter for future researchto investigate this aspect of scoring validity.
Another effect of using this type of scale (the GIMS) is that the importance ofrater training and standardisation becomes even greater than in the past – as it isvital in this system that raters fully understand the process of applying the samescale at different levels and that they are aware of what constitutes acceptableperformance at each level. Since there is evidence that systematic feedbackduring a rating exercise can have a negative effect on rating performance(O’Sullivan and Rignall 2001, 2002) it might be useful to investigate methods ofself-retraining (see Kenyon 1997), using the web. For example the kit developedby the Center for Applied Linguistics (CAL) (CAL 2001) offers an opportunityfor raters, working in their own time and place, to gain a detailed understandingof a test and its scoring procedures. Raters can also gain certification from CALon completion of the training programme through an accreditation procedurewhere their ratings of a set of test performances are compared to those of a set of‘expert raters’. Since this whole area of rater training is relatively unexplored(though see Weigle 1994, 1998) a systematic agenda for research is clearlyneeded. This agenda should take into account some or all of the following:
• the rater – Are there identifiable characteristics which typify the(un)successful rater? (physical, psychological, experi-ential, individual)Are there identifiable behaviours or strategies whichtypify the (un)successful rater?
• the scale – Are there substantive differences in how differentscales are applied? (indigenous vs. linguistic forexample)
5 Conclusions and the way forward
190
• the tasks – Are there particular task types that are susceptible tobias on the part of raters? If so, how is this manifested?
• rater training – In what ways do raters benefit from training?What kind of training is most beneficial? (this refers tomode, format and content)
• rating conditions – In what ways can manipulation of rating conditionsaffect rating performance?
• standardisation – Does standardisation of raters work?How long should we claim its effects last?Is there a best method – or are different methodssuitable for different types of test?
These are just a few of the many possible questions that could be asked of thewhole rating process – I have not included any questions related to the inter-action of some or all of these variables, though obviously this is relevant to anyresearch design.
ListeningThough the Listening papers in the BEC suite were basically unaltered in therevision process, there are some issues that might be investigated.
Weir (2003b:477) refers to an internal report from Cambridge ESOL (Field2000) in which an argument is made for the use of more explicitly authentic textsin the Cambridge Proficiency in English (CPE). Authenticity appears to be seenby Field as being related to both content and delivery. While authenticity ofcontent is dealt with in the BEC suite in terms of the task types and topics that aredesigned to reflect the business language use domain, the area of delivery hasnot really been explored in any depth. Aspects of authenticity such as accent,speed of delivery, ‘reality’ (the degree to which differences between recorded‘real world’ texts and purposefully recorded texts affect the listening process)are all in need of exploration. Listening comprehension tests typically involveboth aural and read input. Where alternative visual input is included (drawings,charts, still photographs, moving images) listening will be affected. The issue ishow and to what extent. While Coniam (2001) suggests that there is evidencethat visuals may detract listeners, in general, there has been very little researchinto the effect on performance of different types of input or of the effects ofinvolving the listener in dealing with a number of different types of input.
Recent advances in neural science have included the development of brainimaging, a procedure that allow us to see what is actually happening in the brain(in terms of neural activity) when people are engaged in high level cognitivetasks. Just et al (2001) used functional magnetic resonance images of brainactivity to investigate how trying to perform two non-related tasks affectedperformance on the two tasks (sentence comprehension, and the mental rotationof three-dimensional objects). Just et al found that when participants attempted
Practical implications
191
to perform the two tasks simultaneously, they did neither task well. In a similarvein, Rubenstein, Meyer and Evans (2001) found that as the difficulty of tasksincreased so did the related time costs (where time is lost in switching from onetask to another).
The implication here is that the inclusion of additional input to a listening taskessentially adds an element to the task which results in test candidates failing toperform that task to the best of their ability (i.e. where a secondary element is notadded). Where this additional input is visual (i.e. it is quite different in nature tothe original aural input) the difference may well be exaggerated. This wouldappear to add support to Skehan’s (1998) ‘code complexity’ idea – wheremanipulation of input will impact on the difficulty of a task – and also thefindings of Coniam (2001).
Buck (2001) suggests that other areas of interest to the researcher mightinclude collaborative listening and the identification of the sub-skills oflistening. The former presents particular difficulties for the tester, with theproblems of identification of the contribution of individual candidates (alwaysproblematic with collaborative tasks), though in the area of business language,there may be an argument for including such a task because this type of listeningis quite typical of the area. It may well be that such a task could be positionedwithin the context of the Speaking paper. There appears to be a danger that testswhich are made up of a single type of listening task or item may be focusing ontoo limited an aspect of an individual’s listening ability and as such may notallow us to draw broad inferences on candidates’ listening ability. Looking backover the BEC suite Listening papers (Chapter 4) we can see that an attempt hasbeen made to identify different focuses for the different elements of the papers.However, the central focus still appears to be on listening for specific details orfor information in a text.
SpeakingThis was the most changed section of the BEC suite of examinations and stillrepresents a great challenge to language testers, despite some quite majoradvances during the past decade. As far back as 1972, a major needs analysisundertaken by the LCCIEB identified speaking as an area of particular interest inbusiness English. This is reflected in the general profusion of Speaking papers orseparate tests among the examinations reviewed in Chapter 1 – with theexception of the TOEIC tests. However, many of the Speaking papers reviewedsuffer from the same non-business orientation as the original BEC Speakingpapers.
We saw in Chapter 4 (Figure 4.15) that different types of task tend to result inquite unique profiles of language functions (O’Sullivan, Weir and Saville 2002).This lends support to the inclusion in a test of speaking of tasks that require thecandidate to perform under different conditions, in the case of the revised BECpapers these are one-to-one with the examiner, alone in an individual long turn
5 Conclusions and the way forward
192
and in interaction with a second (or third) candidate. A score is awarded to eachcandidate immediately following the test by both the examiner/interlocutor andthe examiner/observer. The fact that a single score is awarded by each examinermeans that all tasks are treated as one. This may act to reduce the reliability ofthe scores and at the same time limit the amount of information available to thetest developer. The evidence that there are differences between the tasks couldbe supported by evidence from examiner scores, at present we cannot distin-guish between performance on the different tasks in any post examinationreview. This is certainly an area in which research is required – into the practi-cality of awarding scores for individual tasks in operational conditions, into thedegree to which each task adds to the overall test performance, and into theperception of task importance by the examiners.
It has been argued (McNamara, 1997) that where candidates interact linguis-tically in order to perform a test task, the resulting language is a reflection of theability of all concerned in the interaction, that is, the language is co-constructedby the interacting candidates. The difficulty then is in separating the individualfrom the group or even in identifying unique contributions to the group, sinceeven these apparently unique contributions will have been influenced by theother group members. There is evidence (O’Sullivan 1999, 2000a, 2000b,2002a) that the affective reaction of an individual candidate to characteristicsassociated with their interlocutor can have a systematic and significant impacton subsequent performance. However, the number of potential characteristics,and the interactions between these characteristics, means that the whole area istoo complex to be dealt with without a programme of extensive research inwhich the major characteristics are identified and interactions between thesecharacteristics observed. O’Sullivan (2000a) represents a beginning of thisprocess.
In terms of practicality, we seem to be caught between including tasks whichrequire interaction (and both the candidate-to-candidate discussion and the one-to-one interview are such tasks) or limiting speaking tests to individualmonologues – though even here I would argue that there is still an audience(perceived in an audio or video recorded format, actual in a ‘live’ event) so thepotential for impact on performance is still a factor. Clearly then, it is necessaryto investigate the impact on performance of factors such as interlocutorvariables (e.g. sex or age) and candidate perceptions of the interlocutor/audience (e.g. relative language level, age, personality, status etc.). There is alsoevidence that the effects on candidate performance may be group or culturespecific, for example Porter (1991a) reports that Arabic learners and Europeanlearners demonstrate very different behaviours depending on particular charac-teristics associated with their interviewer. A similar phenomenon was notedalso by O’Sullivan (2000a, 2000b, 2002a). This implies that the culture ofbusiness language may need to be investigated as a separate entity. Ignoring theimpact of audience is simply not an option.
Conclusion
193
Other immediate concerns include monitoring the new interlocutor frames(or scripts) at the different levels and the output language for the different tasks.In terms of the first of these areas, O’Sullivan and Lu (2003) have analyseddeviations from the interlocutor frame in the IELTS and shown that there is littleimpact on the candidate when pre- and post-deviation language is analysed. Thistype of research is clearly relevant to the BEC Speaking papers as is the work ofO’Sullivan, Weir and Saville (2002) in identifying the language functionselicited by speaking test tasks. Brooks (2003) is demonstrating how the check-lists can be adapted for use with a specific test, a process that could quite easilybe employed in monitoring the BEC Speaking papers.
Apart from these concerns, there are still many other questions to beanswered about tests of speaking. O’Sullivan and Weir (2002) have outlined abroad research agenda based on a socio-cultural perspective on language testingbased very much on the type of validation framework outlined in Figure 5.3.
ConclusionI have tried to demonstrate in this book that tests of language for businesspurposes are different, in their theoretical basis, their content and their intendedaudience. I do not believe that I have suggested anything that is completelynew about the subject, but hope that I have offered a perspective on businesslanguage testing (and LSP testing in general) that is supportable from bothpractical and theoretical perspectives. The book has added support to adefinition of LSP tests presented in terms of authenticity (Douglas 2000),though it has suggested that the way we look at authenticity should be with agreater degree of complexity than hitherto conceived.
All tests can be seen as lying on a specificity continuum, between the highlyspecific and the general purpose. This continuum is multi-componential andincludes the twin aspects of authenticity – situational and interactional. Aspecific purpose test will be distinguishable from other tests (both specific andgeneral purpose) in terms of the domain represented by the demands of its tasksand texts, and in terms of the cognitive processing it elicits.
The book has presented a broad outline of current theory and practice in thetesting of language for business purposes. The description of the test devel-opment and revision framework which drives the work of Cambridge ESOL andhow this led to particular changes in the BEC suite of examinations demon-strates the necessity for a set of clear and unambiguous developmental proce-dures. Any LSP, or in this case business language, test development or revisionproject is dependent on having an understanding of the language use domain tobe tested, an awareness of the degree to which authenticity decisions impact onthe specificity of the test and the ability to deliver instruments of a high quality.
Appendices
196
APPENDIX 1.1
JOCT Evaluation Criteria
A+ Superior Japanese communication skills for a wide variety of business situations:In addition to Level A abilities, Level A+ speakers can skilfully summarize their ideas,speak convincingly, pick out essential facts, recognize nuances and generallycommunicate on superior Levels from both technical and cultural perspectives
A Thorough Japanese communication skills for normal business situations:Level A speakers correctly use the special terms and expressions of business. Speech isfluid and any mistake in pronunciation or grammar does not create problems. Their skillsfor handling unfamiliar situations are sufficient, although not complete. Overall, however,they have thorough skills for normal communication using business Japanese.
B+ Require improvement in selected areas to reach Level A:Level B+ speakers can use the special vocabulary and expressions of business, but not aswell as Level A speakers and sometimes with inadequacies. Speech is fluid, but occasionalmistakes with pronunciation, grammar, etc. can cause problems in communication. Theysometimes cannot suitably handle unfamiliar situations.
B Require improvement in many areas to reach Level A:Level B speakers can generally use the special vocabulary and expressions of business, butinadequacies are quite noticeable compared to Level A speakers. Speech is not alwaysfluid and repeated mistakes with pronunciation and grammar cause communicationproblems. They often cannot suitably handle unfamiliar situations.
C Limited communication skills:Level C speakers understand the gist of discussions, but limited knowledge of businessvocabulary and expressions, as well as Japanese business itself, prevents them fromhandling matters suitably. Daily conversations are possible, but their communication lackssmoothness. Mistakes in pronunciation and grammar are frequent.
D Insufficient Japanese communication skills for business:Level D speakers do not have sufficient skills of comprehension and expression, whichprevents them from communicating in Japanese to conduct normal business.
197
APPENDIX 1.2
CEFLS Pilot Test
Appendix 1.2
198
Appendix 1.2
199
Appendix 1.2
200
Appendix 1.2
201
Appendix 1.2
202
Appendix 1.2
203
Appendix 1.2
204
205
APPENDIX 1.3
CEIBT – Test of Reading and Writing –June and November 1992
Appendix 1.3
206
Appendix 1.3
207
Appendix 1.3
208
Appendix 1.3
209
Appendix 1.3
210
Appendix 1.3
211
Appendix 1.3
212
Appendix 1.3
213
Appendix 1.3
214
Appendix 1.3
215
Appendix 1.3
216
Appendix 1.3
217
Appendix
218
Appendix 1.3
Appendix 1.3
219
Appendix
220
Appendix 1.3
221
Appendix 1.3
Appendix
222
Appendix 1.3
223
Appendix 1.3
Appendix
224
Appendix 1.3
225
A Member of the Association of Language Testers in Europe (ALTE)
English Version: EN00
Standard Test
Sample Question Paper
APPENDIX 1.4
BULATS – Standard Test English
Appendix
226
Appendix 1.4
3
Can
dida
te In
form
atio
n
Can
dida
te n
ame:
Fam
ily n
ame:
……
……
……
……
……
……
Firs
t nam
e:…
……
……
……
……
……
…
Can
dida
te N
umbe
r:…
……
……
……
……
……
…
Exam
inat
ion
Cen
tre:
……
……
……
……
……
……
Dat
e:…
……
……
……
……
……
…
Test
Abo
ut 1
10 m
inut
es
Lis
teni
ngA
bout
50
min
utes
.A
s you
list
en, w
rite
your
ans
wer
s in
this
Que
stio
n Pa
per.
Whe
n th
e lis
teni
ng t
est
finis
hes,
you
have
5 m
inut
es t
o co
pyyo
uran
swer
s ont
o yo
ur A
nsw
er S
heet
.
Rea
ding
&Sp
end
60 m
inut
es.
Lan
guag
e K
now
ledg
eW
rite
your
ans
wer
s on
your
Ans
wer
She
et.
2
BLA
NK
PAG
E
227
Appendix 1.4
Appendix
228
Appendix 1.4
7
Wha
t doe
s th
e an
noun
cer s
ay a
bout
the
train
to P
orts
mou
th?
ATh
e tra
in w
ill le
ave
at 1
0.37
.
BTh
e de
partu
re p
latfo
rm h
as b
een
chan
ged.
8
CPa
ssen
gers
will
be u
nabl
eto
get
food
on
the
train
.
Wha
t doe
s th
e w
oman
wan
t her
col
leag
ue to
do?
Atra
in n
ew e
mpl
oyee
s
Bde
mon
stra
te a
mac
hine
9
Cgi
ve a
talk
Who
is th
e m
an o
n th
e ph
one
talk
ing
to?
Ahi
s bo
ss
Ba
cust
omer
10
Chi
s as
sist
ant
6
5 W
hen
will
the
new
pro
duct
be
read
y fo
r tes
ting?
12
34
56
79
1112
13
1415
1617
1819
20
2122
2324
2526
27
2829
3031
810
13
45
6
79
1112
13
1415
1617
1819
20
2122
2324
2526
27
2829
3031
810
12
35
6
79
1112
13
1415
1617
1819
20
2122
2324
2526
27
2829
3031
8
22
10
4
A B
C
Who
is th
e sa
les
assi
stan
t in
the
shop
talk
ing
to?
Ahe
r bos
s
Ban
othe
r ass
ista
nt
6
Ca
cust
omer
Wha
t doe
s M
ike
do a
t the
trai
ning
cen
tre?
AH
e's
a st
uden
t.
BH
e's
a re
cept
ioni
st.
7
CH
e's
a te
ache
r.
229
Appendix 1.4
9
Con
vers
atio
n Th
ree
Que
stio
ns 1
9 –
22
�Lo
ok a
t the
not
es b
elow
.�
You
will
hear
a w
oman
cal
ling
abou
t an
orde
r.
Com
pany
:(19
) ……
……
……
……
……
……
……
……
……
……
.…..
In:
Leon
,(20)
……
……
……
……
……
……
……
……
……
……
…. S
pain
.
Theyy
wan
t:30
0 o
f Mod
el X
42 b
y(2
1) …
……
……
……
……
……
……
……
……
……
……
…. a
t th
e la
test
.
Desp
atch
by(
22) …
……
……
……
……
(the
y wi
ll pa
y).
8
PAR
T TW
OQ
uest
ions
11
– 22
�Yo
uw
ill he
ar th
ree
conv
ersa
tions
.�
Fill
in th
e nu
mbe
red
spac
eson
the
form
s, u
sing
the
info
rmat
ion
you
hear
.�
You
will
hear
eac
h co
nver
satio
non
ce o
nly.
Con
vers
atio
n O
neQ
uest
ions
11
– 14
�Lo
ok a
t the
form
belo
w.
�Yo
uw
ill he
ar a
man
cal
ling
topl
ace
an o
rder
.
OR
DE
R F
OR
M
CU
STO
MER
DET
AIL
S
Nam
e: K
en (1
1)...
......
......
......
......
......
......
......
......
......
......
......
...
Com
pany
: G
reen
light
Com
mu
nic
atio
ns
201
Hal
l Roa
d, M
anch
este
rTe
l: 0
161
3139
88Fa
x:(1
2)...
......
......
......
......
......
......
......
.
OR
DER
DET
AIL
S
Item
:(13
).....
......
......
......
......
......
......
......
......
......
......
M
odel
: X
T519
Qua
ntity
/ Am
ount
: (1
4)...
......
......
......
......
......
......
......
......
......
...
Con
vers
atio
n Tw
oQ
uest
ions
15
– 18
�Lo
ok a
t the
form
belo
w.
�Yo
uw
ill he
ar a
wom
an m
akin
g a
com
plai
nt.
CO
MPL
AIN
T FO
RM
Nam
e: Mrs Hector
Addr
ess:31,
(15)
......
......
......
......
......
......
......
......
......
......
......
......
, Rossington.
Tel:
01923 951975
Dat
e:5 April
Dat
e of
Com
plai
nt (i
f diff
eren
t): (1
6)...
......
......
......
...
Bra
nch:
(17)
......
......
......
......
......
......
......
......
....
Rea
son
forC
ompl
aint
: Goods damaged due to bad
(18)
......
......
......
......
.....
Actio
n: Issue credit note.
Appendix
230
Appendix 1.4
11
PAR
T FO
UR
Sect
ion
One
Que
stio
ns 3
3 –
38
�Yo
u w
ill he
ar a
con
vers
atio
n be
twee
na
univ
ersi
ty s
tude
nt,
Sally
, an
d a
com
pany
repr
esen
tativ
e,D
an,
at a
rec
ruitm
ent
sem
inar
. S
ally
is in
tere
sted
in w
orki
ng f
or D
an’s
com
pany
, Man
son’
s pl
c.�
For q
uest
ion
33
– 38
, circ
le o
ne le
tterA
, B o
r C fo
r the
cor
rect
answ
er.
�Yo
uw
ill he
ar th
e co
nver
satio
ntw
ice.
33Sa
lly fi
nish
es h
er s
tudi
es
Ain
two
wee
ks.
Bin
a m
onth
.C
in s
ixw
eeks
.
34Sa
llyw
ould
like
to w
ork
in
Am
arke
ting.
Bre
tailin
g.C
finan
ce.
35M
anso
n's
have
div
isio
ns in
AEu
rope
onl
y.B
Euro
pe a
nd H
ong
Kong
.C
Hon
g Ko
ng o
nly.
36M
anso
n's
wan
t em
ploy
ees
who
are
Aac
adem
ical
ly c
leve
r.B
dyna
mic
per
sona
litie
s.C
keen
to le
arn.
37In
an
empl
oyee
's fi
rst y
ear,
Man
son'
s of
fer t
rain
ing
in
Am
anag
emen
t.B
sale
s te
chni
ques
.C
mar
ket d
evel
opm
ent.
38Tr
aine
es a
re a
sses
sed
ever
y
A3
mon
ths.
B6
mon
ths.
C12
mon
ths.
10
PAR
T TH
REE
Sect
ion
One
Que
stio
ns 2
3 –
27
�Yo
u w
ill he
ar fi
ve p
eopl
e an
swer
the
ques
tion
‘Wha
t do
you
like
abou
t you
r wor
k?'
�As
you
list
en to
eac
h on
e, d
ecid
e w
hat t
he p
erso
n lik
es m
ost.
�C
hoos
e yo
uran
swer
from
the
list A
– I
and
writ
e th
e co
rrect
lette
r in
the
spac
e pr
ovid
ed.
�Yo
uw
ill he
ar th
e fiv
epi
eces
once
onl
y.
Exam
ple:
……
…. I
……
….
Am
eetin
g lo
ts o
f peo
ple
Bgo
od s
alar
y
23Pe
rson
1 .
......
......
......
......
......
...C
wor
king
on
my
own
24Pe
rson
2 .
......
......
......
......
......
...D
varie
ty
25Pe
rson
3 .
......
......
......
......
......
...E
com
pany
has
goo
d re
puta
tion
26Pe
rson
4 .
......
......
......
......
......
...F
good
offi
ce c
ante
en
27Pe
rson
5 .
......
......
......
......
......
...G
deve
lopi
ng u
sefu
l ski
lls
Hne
ar h
ome
Ifo
reig
n tra
vel
Sect
ion
Two
Que
stio
ns 2
8 –
32
�Yo
uw
ill he
ar fi
vepe
ople
talk
ing.
�As
you
list
en, d
ecid
e w
hat e
ach
of th
em is
talk
ing
abou
t.�
Cho
ose
your
answ
er fr
om th
e lis
t A–
I, an
d w
rite
the
corre
ct le
tter i
n th
e sp
ace
prov
ided
.�
You
will
hear
the
five
piec
eson
ce o
nly.
Exam
ple:
……
…. I
……
….
Aa
plan
fora
new
offi
ce
Ba
prob
lem
at w
ork
28Pe
rson
1 .
......
......
......
......
......
...C
a bu
sine
ss m
eetin
g
29Pe
rson
2 .
......
......
......
......
......
...D
a st
aff m
eetin
g
30Pe
rson
3 .
......
......
......
......
......
...E
a co
nfer
ence
31Pe
rson
4 .
......
......
......
......
......
...F
a jo
b in
terv
iew
32Pe
rson
5 .
......
......
......
......
......
...G
a ne
w c
olle
ague
Hsa
fety
pre
caut
ions
Ia
pay
rise
231
Appendix 1.4
13
Sect
ion
Thre
eQ
uest
ions
45
– 50
�Yo
uw
ill he
ar a
Pers
onne
l Man
ager
inte
rvie
win
g an
app
lican
t for
a jo
b.�
For q
uest
ions
45
– 50
, circ
le o
ne le
tterA
, B o
r C fo
r the
cor
rect
ans
wer
.�
You
will
hear
the
conv
ersa
tion
twic
e.
45In
his
cur
rent
job,
Dav
id h
as to
Ase
e if
certa
in w
ork
has
been
finis
hed.
Bas
sem
ble
parts
of a
mac
hine
.C
help
peo
ple
prog
ress
in th
eir c
aree
rs.
46M
ost o
fthe
tim
e, D
avid
wor
ks in
Ath
e Sa
les
Dep
artm
ent.
Bth
e m
ain
offic
e bl
ock.
Cth
e pr
oduc
tion
area
.
47W
hat i
mpr
ovem
ent d
oes
Dav
id s
ay c
ompu
ters
hav
e m
ade?
APr
oble
ms
are
deal
t with
imm
edia
tely
.B
Prod
uctio
n st
affh
ave
less
to d
o.C
Mor
e de
taile
d in
form
atio
n is
ava
ilabl
e.
48In
Dav
id's
opi
nion
, the
mos
t com
mon
pro
blem
is
Ahu
man
erro
r.B
mac
hine
bre
akdo
wn.
Cm
issi
ng p
arts
.
49D
avid
feel
s he
is s
uita
ble
for t
he n
ew jo
b be
caus
e it
requ
ires
Aw
orki
ngw
ith s
imila
r pro
duct
s.B
prob
lem
-sol
ving
ski
lls.
Ca
know
ledg
e of
com
pute
rs.
50D
avid
rega
rds
him
self
as
Aa
natu
ral l
eade
r.B
a go
od te
am m
embe
r.C
a se
nsiti
ve p
erso
n .
That
is th
e en
d of
the
List
enin
g Se
ctio
n. Y
ou n
ow h
ave
5 m
inut
es to
cop
yyo
ur a
nsw
ers
onto
you
r Ans
wer
She
et.
12
Sect
ion
Two
Que
stio
ns 3
9 –
44
�Yo
uw
ill he
ar a
con
vers
atio
nbe
twee
n tw
oem
ploy
ees
of a
24-
hour
sup
erm
arke
t dis
cuss
ing
som
e te
nder
s th
ey h
ave
rece
ived
for a
cle
anin
g co
ntra
ct. H
elen
isa
Purc
hasi
ng o
ffice
r, an
dTo
ny is
Hea
d of
Mai
nten
ance
.�
For q
uest
ions
39
– 44
, circ
le o
ne le
tter A
, B o
rC fo
r the
cor
rect
ans
wer
.�
You
will
hear
the
conv
ersa
tion
twic
e.
39W
hat i
s th
e pr
oble
m w
ith th
eir p
rese
nt c
ontra
ctor
s?
ATh
ey’re
not
hon
est.
BTh
ey’re
not
relia
ble.
CTh
ey’re
not
sui
tabl
y sk
illed.
40H
elen
thin
ks th
at B
ento
n’s
and
Qui
ckco
Aof
fer v
ery
diffe
rent
dea
ls.
Bdo
n’t d
iffer
ver
y m
uch.
Cha
ve tw
o m
ain
diffe
renc
es.
41W
hen
do th
ey w
ant t
he n
ew c
lean
ers’
con
tract
to s
tart?
Ain
Aug
ust
Bin
Sep
tem
ber
Cin
Dec
embe
r
42H
elen
thin
ks a
key
fact
or in
dec
idin
g w
ho g
ets
the
cont
ract
is
Ath
e sp
eed
of th
e cl
eane
rs.
Bth
e nu
mbe
r ofc
lean
ers.
Cth
e co
st o
f the
cle
aner
s.
43To
ny is
kee
n fo
r Qui
ckco
to g
et th
e co
ntra
ct b
ecau
se th
ey
Aha
ve a
goo
d re
puta
tion.
Bpr
esen
ted
thei
r ten
der w
ell.
Cof
fere
d a
trial
per
iod.
44H
ow d
o th
eyfe
elab
out t
heir
final
dec
isio
n?
ATh
ey a
re c
onfid
ent a
bout
it.
BTh
ey d
ecid
e th
ey n
eed
som
e re
fere
nces
.C
They
wan
t to
disc
uss
som
e is
sues
furth
er.
Appendix
232
Appendix 1.4
15
53
The F
&B G
rou p
Prof
it An
alysis
prop
erty
leis
ure
cent
res
5% 40%
mai
l ord
er c
atal
ogue
hote
ls
10% 45
%
AF&
B m
akes
mos
t of i
ts p
rofit
out
of l
eisu
re c
entre
s an
d pr
oper
ty.
BF&
B re
lies
on le
isur
e ce
ntre
s an
d ho
tels
for t
he g
reat
er p
art o
f its
pro
fit.
CF&
B's
bigg
est p
rofit
-mak
ers
are
its m
ail o
rder
cat
alog
ue a
nd le
isur
e ce
ntre
s.
54In
ter-
Offi
ce M
emo
From
:Ph
ilip J
ones
To:
Mik
e W
illiam
sSu
bjec
t:Fi
nanc
e m
eetin
g, F
riday
Jun
e 8
I am
free
at 1
0.00
on
Frid
ay, b
ut o
nly
fora
bout
an
hour
. Cou
ld w
e ha
ve th
efin
ance
mee
ting
in th
e af
tern
oon,
if p
ossi
ble?
The
re's
a lo
t to
disc
uss.
I'm
free
from
2 o'
cloc
k. If
we
can
begi
n th
en it
wou
ld g
ive
us th
e w
hole
afte
rnoo
n.
Philip
Jon
es w
ants
to
Aca
ncel
the
finan
ce m
eetin
g on
Frid
ay.
Bre
sche
dule
the
finan
ce m
eetin
g fo
r2 o
'clo
ck.
Cm
ake
the
mee
ting
on F
riday
mor
ning
sho
rter.
55
AIn
tern
atio
nal s
ales
rem
aine
dfa
irly
cons
tant
thro
ugho
ut th
e ye
ar.
BSa
les
with
in E
urop
e in
crea
sed
in th
e fo
urth
qua
rter.
CD
omes
tic s
ales
col
laps
ed in
the
seco
nd q
uarte
roft
he y
ear.
020406080100
1st Q
tr2n
d Q
tr3r
d Q
tr4t
h Q
tr
Dom
estic
Euro
pe
Inte
rnat
iona
l
Sale
s fig
ures
for 2
001
14
REA
DIN
G a
nd L
ANG
UAG
E K
NO
WLE
DG
EPA
RT
ON
E
Sect
ion
One
Que
stio
ns 5
1 –
57
�Lo
ok a
t the
follo
win
g m
essa
ges
and
notic
es.
�Fo
r que
stio
ns 5
1 –
57, m
ark
one
lette
rA, B
or C
on
your
Answ
er S
heet
.
Exam
ple: ST
OR
E C
ON
TE
NT
S IN
A C
OO
L P
LA
CE
OU
T O
F D
IRE
CT
SU
NL
IGH
T
ATh
e co
nten
ts s
houl
d be
kep
t at a
con
stan
t tem
pera
ture
.B
The
cont
ents
are
sen
sitiv
e to
hea
t and
ligh
t.C
The
cont
ents
mus
t be
kept
froze
n.
AB
C0
51
See
encl
osed
bro
chur
e fo
r det
ails
and
leve
lsof
com
pens
atio
n.
AYo
u sh
ould
writ
e to
us
for d
etai
ls a
bout
com
pens
atio
n.B
Det
ails
abo
ut c
ompe
nsat
ion
are
give
n in
a s
epar
ate
docu
men
t.C
You
will
find
mor
e in
form
atio
n on
com
pens
atio
n on
the
next
pag
e.
52
He
re a
re d
et
ails
of
th
em
ark
et
ing
se
min
ar
I t
old
yo
u a
bo
ut
.I
sh
an
'tb
e a
ble
to
go
mys
elf
th
at
da
y.
Le
t m
e k
no
w i
fy
ou
th
ink
it's
wo
rth
wh
iles
en
din
g
so
me
on
e e
lse
fro
m t
his
div
isio
n.
Pa
ul.
APa
ul d
oesn
't w
ant t
o se
nd a
nyon
e to
the
sem
inar
.B
Paul
wan
ts y
ou to
repr
esen
t you
r div
isio
n at
the
sem
inar
.C
Paul
wan
ts y
our o
pini
on a
bout
whe
ther
som
eone
sho
uld
go to
the
sem
inar
.
233
Appendix 1.4
17
PAR
T O
NE
Sect
ion
Two
Que
stio
ns 5
8 –
63
�C
hoos
e th
e w
ord
or p
hras
e w
hich
bes
t com
plet
es e
ach
sent
ence
.�
For q
uest
ions
58
– 63
, mar
kon
e le
tterA
, B, C
orD
on
your
Ans
wer
She
et.
My
job
frequ
ently
invo
lves
hav
ing
to w
ork
……
……
……
.. in
tens
e pr
essu
re.
Abe
low
Bun
der
Cbe
neat
h
58
Dun
dern
eath
This
type
of d
ecis
ion
has
to b
e m
ade
at b
oard
……
……
……
.. .
Ala
yer
Bra
nkC
grad
e
59
Dle
vel
The
Man
agin
g D
irect
or is
now
urg
ently
……
……
……
.. to
app
oint
a H
ead
of O
pera
tions
.
Ase
ekin
gB
purs
uing
Cse
arch
ing
60
Dhu
ntin
g
The
com
pany
has
dec
ided
to …
……
……
…..
a sh
are
optio
n sc
hem
e, s
tarti
ng n
ext y
ear.
Ain
trodu
ceB
inno
vate
Cco
nfer
61
Dem
bark
Lack
of o
rder
s ha
s m
eant
that
a n
umbe
r of e
mpl
oyee
s ha
ve b
een
laid
……
……
……
.. .
Ado
wn
Bon
Cup
62
Dof
f
The
com
pany
has
goo
d in
dust
rial …
……
……
…..
and
disp
utes
are
rare
.
Aw
orki
ngs
Baf
fairs
Cte
rms
63
Dre
latio
ns
16
56
Stor
age
Uni
ts -
from
£10
0 ea
chSA
PELE
or T
EAK
mel
amin
e w
ood
effe
ct fi
nish
.Al
so b
lack
ash
, lig
ht o
ak o
r whi
te -
add
10%
to p
rice.
For g
rey
add
17%
to p
rice.
AYo
u pa
y m
ore
for t
eak
units
than
for w
hite
one
s.B
You
pay
the
sam
e fo
r gre
y or
bla
ck a
sh u
nits
.C
You
pay
less
for w
hite
uni
ts th
an y
ou d
o fo
r gre
y.
57Sy
mte
x C
ompa
ny R
esul
ts
Sale
s Ta
rget
Sale
s Th
is Y
ear
Sale
s La
st Y
ear
7,00
07,
200
6,70
0
ATh
is y
ear's
sal
esfig
ures
wer
e no
t as
good
as
last
yea
r's.
BLa
st y
ear's
sal
esfig
ures
did
not
reac
h th
e sa
les
targ
et.
CTh
is y
ear's
sal
esfa
iled
to re
ach
the
sale
s ta
rget
.
Appendix
234
Appendix 1.4
19
64O
ne-th
ird o
f new
Briti
sh b
usin
esse
s
Aco
nsis
t of c
hain
com
pani
es.
Bar
e ow
ned
or m
anag
ed b
y w
omen
.C
oper
ate
with
in th
e fo
od s
ecto
r.
65W
hat d
oes
the
seco
nd p
arag
raph
tell
us a
bout
the
com
pany
cal
led
Surp
rise?
ATh
e co
mpa
ny d
eliv
ers
its p
rodu
cts
by p
ost.
BTh
e co
mpa
ny h
as m
ade
a pr
ofit
ever
yye
ar.
CTh
e co
mpa
ny o
pera
tes
from
a g
arag
e.
66W
hat d
id N
ina
Tayl
or th
ink
help
ed h
er w
hen
she
star
ted
herb
usin
ess?
Aha
ving
a g
ood
know
ledg
e of
the
law
Bbe
ing
the
owne
r ofa
war
ehou
seC
know
ing
noth
ing
abou
t the
boo
k tra
de
67W
hat s
how
s th
at D
enlo
ws
Rec
ruitm
ent A
genc
y ha
s be
en a
suc
cess
?
AIt
has
been
in b
usin
ess
for m
ore
than
two
year
s.B
It ha
s re
crui
ted
25 n
ew e
mpl
oyee
s.C
It ha
s op
ened
sev
eral
new
offi
ces.
68C
offe
e C
hoic
e C
ompa
ny w
as e
stab
lishe
d
Ato
fill a
gap
in th
em
arke
t.B
desp
ite s
trong
com
petit
ion.
Cw
ith fi
nanc
e fro
m th
e U
S.
69Ac
cord
ing
to th
efin
alpa
ragr
aph
wom
en w
ant t
o ru
n th
eir o
wn
busi
ness
esbe
caus
e th
ey
Aen
joy
lear
ning
new
ski
lls.
Blik
e m
akin
g pl
ans
for t
hefu
ture
.C
pref
er w
orki
ng fo
r sm
all c
ompa
nies
.
18
PAR
T O
NE
Sect
ion
Thre
eQ
uest
ions
64
– 69
�R
ead
the
artic
le b
elow
abo
ut w
omen
who
run
thei
row
n bu
sine
ss a
nd a
nsw
erqu
estio
ns64
– 6
9on
the
oppo
site
pag
e.�
For q
uest
ions
64 –
69,
mar
kon
e le
tterA
, B o
rC o
n yo
ur A
nsw
erSh
eet.
WO
MEN
RU
NN
ING
TH
EIR
OW
N B
USI
NES
S
Acco
rdin
g to
rec
ent
rese
arch
, a
third
of
new
com
pani
es s
et u
p in
Brita
in a
re r
unby
wom
en. T
ypic
al e
xam
ples
are
Lis
a Si
mon
s, w
ho s
tarte
d up
her
own
high
lysu
cces
sful
clot
hing
sho
ps in
Lond
on w
ith a
loan
of £
15,0
00 fr
om h
er b
ank,
and
Kat
eR
oger
s, w
hose
t up
Cel
larC
afés
five
yea
rs a
go a
nd s
old
outt
o a
maj
orre
stau
rant
cha
in th
is m
onth
,at
a pr
ofit
of£3
milli
on.
Ther
e ar
e m
any
othe
r ex
ampl
es. S
urpr
ise,
a m
ail o
rder
com
pany
sel
ling
unus
ual g
ifts,
had
a tu
rnov
er o
f $4
milli
on la
st y
ear.
The
com
pany
was
sta
rted
by C
laire
Ful
ler
five
year
s ag
o in
a g
arag
e, b
ut l
ater
mov
ed i
nto
prem
ises
in
the
cent
reof
Cov
entry
.Si
mila
rly, N
ina
Tayl
or s
tarte
d he
rcom
pany
, NC
Boo
ks, i
n an
old
war
ehou
se in
Bris
tol.
Ala
wye
r by
trai
ning
, she
had
no
prev
ious
exp
erie
nce
of th
e re
tail
trade
but
bel
ieve
s th
atth
is w
orke
d to
her
adv
anta
ge.
‘Hav
ing
no k
now
ledg
e of
the
boo
k tra
de,’
she
says
,‘a
llow
ed m
e to
brin
g fre
sh id
eas
into
the
busi
ness
.’
Youn
g w
omen
are
als
o ru
nnin
g su
cces
sful
busi
ness
es, l
ike
Mar
iaFe
llow
san
d C
hris
tine
Cra
ig,
both
still
in t
heir
early
tw
entie
s. T
oget
her
they
set
up
Den
low
s R
ecru
itmen
tAg
ency
in a
n of
fice
in L
iver
pool
less
than
two
year
s ag
o; th
e fir
mno
w h
as s
ixof
fices
and
a st
affo
f 38.
Ano
ther
you
ng w
oman
, Am
eric
an-b
orn
Amy
Baile
y,co
uld
not f
ind
any
good
cof
fee
shop
s in
Engl
and
so s
he s
et u
pC
offe
e C
hoic
e C
ompa
ny i
n 19
94.
Two
year
s ag
o a
larg
e Am
eric
an g
roup
bou
ght t
he c
ompa
ny, a
lthou
gh A
my
cont
inue
s to
pla
ya
lead
ing
part
in th
e bu
sine
ss.
Ther
e ar
em
any
reas
ons
why
wom
en w
ant t
o ru
n th
eir o
wn
busi
ness
es. O
ne is
that
they
are
incr
easi
ngly
conf
iden
t tha
t the
y ar
e ca
pabl
e of
doi
ng th
is. A
noth
er is
that
man
y no
long
erw
ant
to w
ork
for
com
pani
es w
hich
fai
l to
offe
r sa
tisfy
ing
care
ers.
‘W
omen
’sbu
sine
sses
are
ofte
n m
ore
succ
essf
ul th
at th
ose
run
by m
en,’
says
Den
ise
John
son,
of
the
Nat
iona
lBus
ines
s Ba
nk,
‘bec
ause
wom
en a
re g
ood
at lo
okin
g ah
ead
and
seei
ngw
hat w
ill be
bes
t for
a b
usin
ess
– th
ey fi
nd it
ver
y sa
tisfy
ing.
’
235
Appendix 1.4
21
REA
DIN
G a
nd L
ANG
UAG
E K
NO
WLE
DG
E
Part
Tw
o
20
PAR
T O
NE
Sect
ion
Four
Que
stio
ns 7
0 –
74
�Fo
r que
stio
ns 7
0 –
74, r
ead
the
text
bel
owan
d th
ink
of th
e w
ord
whi
ch b
est f
its e
ach
spac
e.�
Writ
e on
ly o
new
ord
in e
ach
spac
e on
you
rAns
wer
She
et.
Exam
ple:
He
is v
ery
inte
rest
ed …
……
.……
…. c
ompu
ters
.
Answ
er:
Mar
ia's
Bea
uty
Bou
tique
s
Mar
ia W
ilson
is th
e no
w f
amou
s M
anag
ing
Dire
ctor
ofB
eaut
y Bo
utiq
ue I
nter
natio
nal p
lc,
one
of…
……
(70)
…...
.. w
orld
’s
beau
ty
prod
uct
chai
ns.
The
daug
hter
of
Sp
anis
h
imm
igra
nts,
she
...…
(71)
.....
bor
n an
d ed
ucat
ed in
Brita
in.
She
open
ed h
erfir
st B
eaut
y Bo
utiq
ue in
1976
and
it w
as .
....
(72)
....
. po
pula
r th
at s
he
open
ed f
ive
mor
esh
ops
in 3
yea
rs.
......
. (7
3) .
....
then
bus
ines
sha
s de
velo
ped
at a
n
amaz
ing
rate
, and
the
chai
n of
Bea
uty
Bout
ique
s ...
.. (7
4) ..
... n
ow e
xpan
ding
all
over
the
wor
ld.
0in
Appendix
236
Appendix 1.4
23
A
Fina
ncia
l Ana
lyst
£20
-£25
,000
+ b
enef
itsAs
a
youn
g an
d en
thus
iast
icfin
alis
t/new
ly
qual
ified
ac
coun
tant
, yo
u sh
ould
dem
onst
rate
the
pote
ntia
l to
prog
ress
qui
ckly
with
in th
e co
mpa
ny a
nd to
dev
elop
an
exce
llent
und
erst
andi
ng o
f the
com
mer
cial
nee
ds o
fthe
bus
ines
s. W
ithan
anal
ytic
alap
proa
ch a
nd s
trong
com
mun
icat
ion
skills
, you
will
prov
ide
supp
ort o
n pu
rcha
se p
rice
varia
nces
, pre
pare
fina
ncia
l rep
orts
for
the
Euro
pean
Hea
d O
ffice
and
unde
rtake
ad
hoc
proj
ects
.
B
Exec
utiv
e D
irect
or
The
Wom
en’s
Env
ironm
ent N
etw
ork
(WEN
) is
look
ing
for a
hig
hly
mot
ivat
ed, i
nspi
ring
wom
an to
lead
one
of B
ritai
n’s
mos
t suc
cess
ful
envi
ronm
enta
l cam
paig
ning
and
info
rmat
ion
orga
nisa
tions
.
Expe
rienc
e re
quire
d in
clud
esm
anag
emen
t and
orga
nisa
tiona
l ski
lls,
med
ia, w
ritin
g an
d ed
iting
expe
rienc
e an
d a
know
ledg
e of
the
envi
ronm
ent a
nd/o
r wom
en’s
mov
emen
t. Yo
u w
ill ne
ed to
run
cam
paig
ns a
nd m
otiv
ate
a te
am o
fabo
ut 3
0 vo
lunt
eers
and
pai
d st
aff.
C
Wan
ted:
EXPE
RIE
NC
ED IN
TER
NAT
ION
ALSA
LESP
ERSO
Nre
pres
entin
g le
adin
g tra
vel c
ompa
ny to
com
pani
es a
nd m
ultin
atio
nals
in th
e U
K.Ap
plic
ants
are
exp
ecte
d to
be
able
to d
ispl
ay w
ord-
proc
essi
ng s
kills
and
fam
iliarit
yw
ith d
atab
ase
softw
are.
Bas
ic s
alar
y, e
xcel
lent
com
mis
sion
and
com
pany
car
.W
eha
ve a
full-
scal
e Lo
ndon
offi
ce.
D
Off
ice
Eq
uip
me
nt
Sa
les
Sta
ff�
Ne
w B
us
ine
ss
�S
ale
s p
eo
ple
Req
uire
d by
su
cces
sful
of
fice
equi
pmen
t su
pplie
r to
deve
lop
busi
ness
in
Euro
pe.
Mus
t ha
vepr
oven
abi
lity
tose
ll at
seni
or l
evel
s in
top
com
pani
es.
The
succ
essf
ulap
plic
ant w
ill b
e flu
ent i
n at
leas
t tw
o fo
reig
n la
ngua
ges.
22
PAR
T TW
O
Sect
ion
One
Que
stio
ns 7
5 –
81
�R
ead
thes
e se
nten
ces
and
the
job
adve
rtise
men
ts o
n th
e op
posi
te p
age.
�W
hich
job
does
eac
h se
nten
ce 7
5 –
81 re
fer t
o?�
For e
ach
sent
ence
, mar
kon
e le
tterA
, B,C
or D
on
your
Ans
wer
Shee
t.
Exam
ple:
0Yo
u ne
ed to
be
able
to s
peak
two
orm
ore
fore
ign
lang
uage
s.
Answ
er:
AB
CD
0
75Yo
u ne
ed to
hav
e ex
perie
nce
of w
orki
ngw
ith n
ewsp
aper
s, T
V or
radi
o.
76Yo
u ha
ve to
be
able
to u
se a
com
pute
r effi
cien
tly.
77Th
e jo
b w
ill in
volv
e pr
ovid
ing
seni
or s
taff
with
dat
a.
78Yo
u sh
ould
hav
e su
cces
sful
sal
es e
xper
ienc
e w
ith le
adin
g co
mpa
nies
.
79Yo
u ne
ed to
be
able
to g
et th
e be
st o
ut o
fpeo
ple
who
wor
k fo
r you
.
80Yo
ur e
arni
ngs
will
depe
nd o
n ho
w m
uch
you
sell.
81Yo
u ne
ed to
sho
w y
ou a
re s
uita
ble
fore
arly
pro
mot
ion.
237
Appendix 1.4
25
PAR
T TW
O
Sect
ion
Thre
eQ
uest
ions
87
– 91
�R
ead
the
artic
le b
elow
abo
ut a
bus
ines
sman
who
mad
e a
fortu
ne fr
om h
is c
ar m
agaz
ines
.�
For e
ach
ques
tion
87 –
91,
writ
e on
ew
ord
in th
e sp
ace
on y
our A
nsw
er S
heet
.
Exam
ple:
He
is v
ery
inte
rest
ed …
……
.……
…. c
ompu
ters
.
Answ
er:
Mill
iona
irew
ho ju
stlo
ves
cars
John
Paj
acko
wsk
i is
a ta
ll, th
in,f
it 54
yea
r old
who
is w
orth
an
estim
ated
£14
5m
illion
than
ks to
his
Car
Tra
der m
agaz
ines
.
Back
in …
……
(87)
……
… 1
960s
, Joh
n w
as w
orki
ng in
Am
eric
a, s
ellin
g Br
itish
spor
ts c
ars.
It w
as h
ard
wor
k bu
t ……
…. (8
8)…
……
.. he
was
ther
e, h
e sa
w a
mag
azin
e w
ith p
ictu
res
of c
ars
for s
ale.
The
mag
azin
e w
as re
gion
al a
ndan
yone
……
…. (8
9)…
……
…. a
dver
tise
thei
r car
in it
for a
rela
tivel
ysm
all f
ee.
Wha
t im
pres
sed
John
mos
t was
that
……
……
(90)
……
…. s
ingl
e ad
verti
sem
ent
had
a pi
ctur
e of
the
car,
unlik
e no
rmal
new
spap
er a
dver
ts, w
hich
just
pro
vide
da
writ
ten
desc
riptio
n.
Ret
urni
ng to
Brit
ain,
Joh
n pu
t all
of h
is s
avin
gs in
to p
rodu
cing
a m
agaz
ine
like
the
one
he …
……
.. (91
)……
……
see
n in
the
Stat
es.
He
star
ted
in th
e Lo
ndon
regi
on b
ut w
as s
oon
prod
ucin
g si
mila
r mag
azin
es fo
r tw
elve
mor
e re
gion
s an
d,by
199
0,fo
r thr
ee o
ther
cou
ntrie
s as
wel
l.
0in
24
PAR
T TW
O
Sect
ion
Two
Que
stio
ns 8
2 –
86
�R
ead
this
lette
r abo
ut b
uyin
g a
com
pute
r.�
Cho
ose
the
best
wor
d to
fill
each
spa
ce fr
om th
e w
ords
bel
ow.
�Fo
r eac
h qu
estio
n 82
– 8
6, m
ark
one
lette
rA, B
, C o
r D o
n yo
ur A
nsw
er S
heet
.
1 A
pril
Dea
r M
rW
hite
head
Re: Su
pply o
f XR6
Wor
ksta
tion
sI
have
ple
asur
e in
con
firm
ing
our
abili
ty t
o m
eet
your
requ
irem
ents
for
the
Sili
con
Grap
hics
wor
ksta
tion
. In
vie
wof
you
rsp
ecia
l nee
ds, I
sug
gest
tha
t yo
u pl
ace
your
orde
r fo
r th
e ag
reed
equ
ipm
ent
as s
oon
as p
ossi
ble.
The
.....(8
2)...
.. ti
me
for
hard
ware
for
exa
mpl
e is
6 w
eeks
fro
m r
ecei
pt o
f or
der
to ..
...(8
3)...
.. .
Thus
, an
orde
r pl
aced
wit
h us
tom
orro
w wi
ll ...
..(84
).....
del
iver
y to
you
r si
te b
y th
e we
ekco
mm
enci
ng M
onda
y, 1
5th
May
. A
ll or
ders
mus
t be
acc
ompa
nied
by
a ...
..(85
).....
of
20%
of
the
tota
l am
ount
sho
wn o
n th
e at
tach
ed ..
...(8
6)...
.. .
Your
s si
ncer
ely
Vinc
ent
Law
Sale
s M
anag
er
82A
delivery
Barriv
alC
trans
port
Dpo
stag
e
83A
institu
tion
Binitiation
Cintro
duction
Dinstallatio
n
84A
compe
lB
ensu
reC
prom
ise
Dmaintain
85A
portion
Bco
stC
depo
sit
Ddisc
ount
86A
cheq
ueB
quotation
Cch
arge
Dde
man
d
Appendix
238
Appendix 1.4
27
Onc
e fu
rther
inve
stm
ent h
as b
een
……
……
……
.. , t
he p
lans
for g
row
th c
an b
eca
rried
out.
Ase
cure
dB
oblig
edC
boun
d
95
Dfo
rced
This
app
roac
h w
ill en
able
us to
get
the
max
imum
ben
efit
from
……
……
……
.. of
sca
le.
Asa
ving
sB
com
pens
atio
nsC
reso
urce
s
96
Dec
onom
ies
My
boss
pro
mis
ed to
……
……
……
.. on
boar
d th
e su
gges
tions
I m
ade
at th
e m
eetin
g.
Ata
keB
let
Cho
ld
97
Dse
t
26
PAR
T TW
O
Sect
ion
Four
Que
stio
ns 9
2 –
97
�C
hoos
e th
e w
ord
or p
hras
e w
hich
bes
t com
plet
es e
ach
sent
ence
.�
For q
uest
ions
92
– 97
, mar
kon
e le
tterA
, B, C
, orD
on
your
ans
wer
she
et.
The
succ
essf
ulap
plic
ant w
ill ha
ve a
pro
ven
track
……
……
……
.. in
pro
ject
man
agem
ent.
Ahi
stor
yB
reco
rdC
curri
culu
m
92
Dpe
rform
ance
Whe
n re
plyi
ng, p
leas
e …
……
……
…..
the
abov
e re
fere
nce
num
ber.
Are
fer
Bre
prod
uce
Cqu
ote
93
Dal
lude
Like
man
y co
mpa
nies
, we
wer
e af
fect
ed b
y th
e …
……
……
…..
in th
e w
orld
eco
nom
y.
Ado
wns
ide
Bdo
wnt
urn
Cdo
wng
rade
94
Ddo
wnf
all
239
Appendix 1.4
29
98W
hat d
oes
the
writ
er s
ay a
bout
car
eers
adv
ice
in th
e fir
st p
arag
raph
?
ATh
ose
rece
ivin
g it
have
som
etim
es d
isag
reed
with
it.
BIt
has
favo
ured
cer
tain
type
s of
peo
ple
over
oth
ers.
CSo
me
of th
e st
anda
rd a
dvic
e gi
ven
has
been
app
ropr
iate
.D
It ha
s ac
quire
d a
mor
e ap
prop
riate
title
than
it u
sed
to h
ave.
99W
hat d
oes
the
writ
er s
ay a
bout
org
anis
atio
ns in
the
seco
nd p
arag
raph
?
ATh
ey c
an tu
rn s
uita
ble
empl
oyee
s in
to u
nsui
tabl
e on
es.
BM
any
ofth
emha
ve a
low
opi
nion
of c
aree
rs a
dvis
ors.
CTh
ey a
re in
sens
itive
to th
e ef
fect
thei
rmet
hods
hav
e on
em
ploy
ees.
DM
any
are
uncl
ear a
bout
wha
t to
expe
ct fr
om e
mpl
oyee
s.
100
The
writ
er s
ays
that
whe
n in
divi
dual
s m
ake
chan
ges
to th
eir j
obs,
Ath
ey m
ay n
ot b
e aw
are
that
they
are
doi
ng th
is.
Bth
ey g
ener
ally
hav
e th
e be
st o
f int
entio
ns.
Cth
ey a
re o
ften
afra
id o
fthe
con
sequ
ence
s.D
they
freq
uent
ly p
rete
nd th
ey a
re n
ot d
oing
this
.
101
The
writ
er’s
mai
n po
int i
n th
e fo
urth
par
agra
ph is
that
Ape
ople
hav
e be
com
e le
ss s
ure
abou
t whi
ch c
aree
r wou
ld s
uitt
hem
.B
peop
le n
o lo
nger
wan
t to
stay
in th
e sa
me
job
fora
long
tim
e.C
the
spee
d of
cha
nge
in th
e w
orld
of w
ork
has
caus
ed c
onfu
sion
.D
care
ers
advi
ce c
an n
o lo
nger
focu
s on
the
natu
re o
f spe
cific
jobs
.
102
The
writ
er c
ontra
sts
cons
cien
tious
peo
ple
with
neu
rotic
peo
ple
with
rega
rd to
Ath
e ki
nd o
f adv
ice
they
can
be
give
n.B
thei
rcha
nces
of f
indi
ngem
ploy
men
t.C
the
impr
essi
on th
ey g
ive
to c
olle
ague
s.D
thei
r willi
ngne
ss to
take
advi
cefro
m o
ther
s.
103
Wha
t is
the
writ
er’s
gen
eral
vie
w o
n ca
reer
s ad
vice
inth
efin
alpa
ragr
aph?
AIt
has
had
to a
llow
for t
he fa
ct th
at p
eopl
e no
w h
ave
high
er a
spira
tions
.B
Whe
ther
it is
usef
ul o
r not
has
bec
ome
hard
er to
ass
ess.
CPr
edic
ting
futu
re d
evel
opm
ents
has
bec
ome
its k
ey in
gred
ient
.D
It ha
s be
com
e m
uch
less
rele
vant
in to
day’
s w
orld
of w
ork.
28
PAR
T TW
O
Sect
ion
Five
Que
stio
ns 9
8 –
103
�R
ead
the
artic
le b
elow
abo
ut c
aree
rs a
dvic
e an
d an
swer
ques
tions
98
– 10
3on
the
oppo
site
pag
e.�
For
ques
tions
98
– 10
3, m
ark
one
lette
rA
, B
, C
or
D f
or t
he c
orre
ct a
nsw
er o
n yo
urAn
swer
She
et.
Key
s to
unl
ock
path
of c
aree
r ful
film
ent
It us
ed to
be
calle
d vo
catio
nal g
uida
nce.
The
n it
beca
me
care
ers
advi
ce a
nd c
ouns
ellin
g. S
ince
the
late
1920
s th
ere
have
bee
n va
rious
test
s to
hel
pgu
ide
bew
ilder
ed a
nd g
uile
less
sch
ool l
eave
rs in
tojo
b ca
tego
ries
suite
d to
thei
r int
eres
ts a
nd a
bilit
ies.
Hen
ce e
xtro
verts
have
bee
n ad
vise
d,w
isel
y,to
go i
nto
sale
s an
d m
arke
ting,
whe
re t
hey
thriv
e on
the
var
iabi
lity,
peop
le-c
onta
ctan
dai
rof
optim
ism
.Int
rove
rts,
on t
he o
ther
han
d, f
ind
the
quie
tw
ork
ofac
coun
ts,
stor
es a
nden
gine
erin
g,w
here
they
can
wor
k al
one
in a
less
peo
ple-
orie
ntat
ed, f
rene
tic a
tmos
pher
e, m
ore
toth
eir t
aste
.
Butg
ood
care
ers
advi
sors
need
to lo
ok a
t oth
erfa
ctor
s w
hen
givi
ng a
dvic
e. F
or e
xam
ple,
car
eer
coun
sellin
gis
flaw
ed if
it fa
ils to
allo
w fo
r the
pos
sibi
lity
of p
eopl
e ad
aptin
gto
, and
cha
ngin
g,th
eir
jobs
onc
e th
eyar
e in
them
. M
ost
orga
nisa
tions
atte
mpt
thr
ough
var
ious
exp
licit
(indu
ctio
n,m
ento
ring,
train
ing,
app
rais
al) a
nd im
plic
it (re
lianc
e on
obs
erva
tion)
tech
niqu
es to
mou
ld b
ehav
iour
into
an a
ccep
tabl
e pa
ttern
.Th
is m
eans
that
attit
udes
and
eve
n ap
titud
es o
f em
ploy
ees
may
be
chan
ged
over
the
first
yea
r of
em
ploy
men
t, so
met
imes
, but
not
alw
ays,
in th
e di
rect
ion
desi
red
byth
e or
gani
satio
n. T
hus
wha
t was
a ‘f
it’ m
ay e
asily
and
qui
ckly
dev
elop
into
a m
isfit
and
vic
e ve
rsa .
Indi
vidu
als
also
cha
nge
thei
r jo
bs w
ithou
t lea
ving
them
.The
yre
arra
nge
furn
iture
, use
spa
cean
dte
chno
logy
diffe
rent
ly a
nd p
erso
nalis
e di
ffere
nt a
spec
ts o
fth
e jo
b.Th
ey c
an n
egot
iate
with
colle
ague
s, e
arn
spec
ial p
rivile
ges
and
use
othe
r m
eans
to im
prov
e th
eir
role
and
out
put.
In th
isse
nse,
ver
y so
on th
ey a
re d
oing
the
job
diffe
rent
lyfro
mth
eirp
rede
cess
ors,
and
poss
ibly
from
the
way
reco
mm
ende
d by
the
com
pany
. How
ever
, bot
h ad
apta
tion
of p
erso
nal w
ork-
styl
e an
d at
tem
pts
to c
hang
e th
e w
ay o
f do
ing
the
job
are
mor
e lik
ely
to le
ad t
oa
high
erle
vel o
f‘fi
t’,be
caus
e th
ech
ange
s ar
e us
ually
all
atte
mpt
s to
incr
ease
‘fit’
.
And
jobs
the
mse
lves
cha
nge.
Org
anis
atio
nal r
estru
ctur
ing,
the
dev
elop
men
t of
new
tech
nolo
gy,
chan
ges
in th
e m
arke
t, an
d so
on,
all
mea
n th
atjo
bs e
volv
efa
irly
fast
. Bec
ause
oft
he s
peed
ofte
chno
logy
, al
l jo
bs a
re i
n a
stat
e of
flux.
Job
ana
lysi
s is
the
refo
re b
ecom
ing
less
rel
evan
t to
care
ers
advi
ce.C
ouns
ello
rs’ a
naly
sis
has
tofo
cus
inst
ead
on p
erso
nal p
oten
tial r
athe
r tha
n cu
rrent
know
ledg
e or
ski
ll. A
nd th
efe
atur
es o
fsuc
h an
alys
is a
re p
oten
tial s
peed
and
thor
ough
ness
in th
eac
quis
ition
of n
ewkn
owle
dge
and
skills
.
The
area
s th
at a
re p
roba
bly
mos
t pre
dict
ive
ofth
ese
are
inte
lligen
ce a
nd p
erso
nalit
y. In
tellig
ence
ispr
obab
ly t
he b
est
pred
icto
r of
spe
ed o
f le
arni
ng.
Ofte
n, i
ntel
ligen
t pe
ople
are
cur
ious
and
self-
conf
iden
t an
d he
nce
happ
y to
tac
kle
new
tas
ks.
Inte
lligen
t pe
ople
are
bet
ter
and
quic
ker
atan
alys
ing
both
lo
gica
l an
d de
duct
ive
and
crea
tive
prob
lem
s.
Prob
ably
the
mos
tim
porta
ntpe
rson
ality
dim
ensi
ons
are
cons
cien
tious
ness
and
neu
rotic
ism
.Con
scie
ntio
us p
eopl
e so
onge
ta
good
repu
tatio
n w
hich
ser
ves
them
in g
ood
stea
d. C
oupl
ed w
ith a
bilit
y, th
is tr
aiti
s a
sure
-fire
car
eer
win
ner.
On
the
othe
r han
d ne
urot
icis
m is
a d
eep
and
abid
ing
hand
icap
.It i
s no
t eas
yto
‘cur
e’ a
ndca
n ha
ve a
last
ing
effe
ct o
n a
care
er.N
euro
tics
are
unha
ppy
and
tend
tobe
dis
satis
fied,
stre
ssed
and
com
plai
ning
in a
ll jo
bs th
ey h
ave.
The
cons
iste
ncy
ofth
eir b
ehav
iour
lead
s th
em to
deve
lop
apo
orre
puta
tion
in th
e w
orkp
lace
whi
ch o
f cou
rse
can
be s
elf-f
ulfil
ling.
Succ
ess
in th
e jo
bin
202
0 ca
n be
ass
esse
d no
w. W
eha
ve li
ttle
idea
wha
tthe
wor
ld o
f wor
k w
ill be
like
but
advi
sors
can
do
a re
ason
able
job
in
asse
ssin
gth
e po
tent
ial
ofth
ein
divi
dual
with
in i
tbe
caus
e th
eykn
ow th
e pr
edic
tors
of s
ucce
ss. A
nd it
is th
ese
pred
icto
rs th
at c
ouns
ello
rsne
edto
atte
nd to
ifth
ey a
re to
give
the
best
adv
ice.
Appendix
240
Appendix 1.4
31
BLA
NK
PAG
E
30
PAR
T TW
O
Sect
ion
Six
Que
stio
ns 1
04–
110
�Yo
ur s
ecre
tary
has
giv
en y
ou th
is le
tter t
o ch
eck.
�In
som
e lin
es th
ere
is o
ne w
rong
wor
d.�
If th
ere
is a
wro
ng w
ord,
writ
e th
e co
rrect
wor
don
you
r ans
wer
she
et.
�If
ther
e is
no
mis
take
, put
a ti
ck (�
) on
your
Ans
wer
She
et.
Exam
ple:
One
oft
he it
ems
you
orde
red
from
our
cata
logu
e0
�
is te
mpo
rary
out
of s
tock
.00
tem
pora
rily
Dea
r Mr R
ose
,
It w
as a
ple
asur
e to
mee
t you
the
othe
rday
. I w
as v
ery
grat
eful
104
that
you
wer
e ab
le to
find
som
e tim
es in
you
r bus
y sc
hedu
le to
vis
it us
.
105
I tho
ught
it s
houl
d be
hel
pful
ifI p
ut o
n pa
per s
ome
of th
e po
ints
we
106
agre
ed o
n at
our
mee
ting
and
indi
cate
d so
me
with
the
actio
n po
ints
.
107
We
agre
ed w
hich
my
com
pany
will
act a
s yo
ur a
gent
in n
orth
ern
108
Euro
pe. A
s yo
ur a
gent
, we
will
oper
ate
on a
com
mis
sion
bas
is. W
e w
ill
109
char
ge y
ou o
nth
e ra
teof
20%
for s
ales
up
to 2
milli
on E
uros
. On
sale
s ab
ove
110
such
figu
rew
ew
ill ch
arge
com
mis
sion
on
asl
idin
g sc
ale
up to
a m
axim
um
rate
of 2
5%.
Your
s si
ncer
ely
John
Sm
ith
English Version: EN40
Speaking Test
Sample Question Paper
A Member of the Association of Language Testers in Europe (ALTE)
APPENDIX 1.5
BULATS – Speaking Test
241
Appendix 1.5
242
In P
art
2 o
f th
e t
est
yo
u w
ill b
e a
ske
d t
o g
ive
a s
ho
rt p
rese
nta
tio
n o
n o
ne
of
thre
eto
pic
s.
Acco
rdin
g t
o y
ou
r w
ork
exp
erie
nce
, th
e t
op
ics m
ay b
e o
n g
en
era
l w
ork
,te
ch
nic
al w
ork
, o
r stu
dy.
Th
is p
art
will
la
st
ap
pro
xim
ate
lyfo
ur
min
ute
s.
PA
RT
2
P
res
en
tati
on
INS
TR
UC
TIO
NS
Ple
ase
re
ad
all
TH
RE
E t
op
ics b
elo
w c
are
fully
.
Ch
oo
se
ON
E w
hic
h y
ou
fe
el yo
u w
ill b
e a
ble
to
ta
lk a
bo
ut
for
on
e m
inu
te.
Yo
u h
ave
on
e m
inu
te t
o r
ea
d a
nd
pre
pa
re y
ou
r ta
lk.
Yo
u m
ay t
ake
no
tes.
To
pic
A
De
scrib
e a
n im
po
rta
nt
bu
sin
ess m
ee
tin
g y
ou
att
en
de
d.
Yo
u s
ho
uld
sa
y:
wh
ere
it
wa
s;
wh
at
it w
as a
bo
ut;
wh
y it
wa
s im
po
rta
nt.
Wh
at
we
re t
he
mo
st
inte
restin
g m
om
en
ts?
To
pic
B
De
scrib
e s
om
eo
ne
yo
u p
art
icu
larly e
njo
y w
ork
ing w
ith
.Y
ou
sh
ou
ld s
ay:
wh
at
this
pe
rso
n d
oe
s;
wh
at
so
rt o
f w
ork
yo
u d
o w
ith
th
is p
ers
on
;w
hy y
ou
lik
e w
ork
ing w
ith
th
is p
ers
on
.
Wo
uld
yo
u c
ha
nge
an
yth
ing a
bo
ut
this
pe
rso
n?
G
ive
re
aso
ns f
or
yo
ur
an
sw
er.
To
pic
C
De
scrib
e t
he
be
st
wo
rkp
lace
yo
u h
ave
eve
r h
ad
.Y
ou
sh
ou
ld s
ay:
wh
ere
th
e w
ork
pla
ce
wa
s;
wh
at
yo
u w
ere
do
ing t
he
re;
wh
y y
ou
lik
ed
to
wo
rk t
he
re.
Wo
uld
yo
u c
ha
nge
an
yth
ing a
bo
ut
it?
G
ive
re
aso
ns f
or
yo
ur
an
sw
er.
BU
LA
TS
Sp
eakin
g T
est
Sam
ple
Mate
rials
Th
ere
are
th
ree
pa
rts t
o t
he
sp
ea
kin
g t
est;
in
Pa
rt 1
, yo
u w
ill h
ave
an
Inte
rvie
w.
Th
is w
ill in
clu
de
qu
estio
ns f
rom
th
e s
ectio
ns b
elo
w:
�a
ll ca
nd
ida
tes w
ill b
e a
ske
d q
ue
stio
ns f
rom
Se
cti
on
1;
�p
eo
ple
in
wo
rk w
ill b
e a
ske
d q
ue
stio
ns f
rom
Se
cti
on
2;
�stu
de
nts
will
be
aske
d q
ue
stio
ns f
rom
Se
cti
on
3;
�a
ll ca
nd
ida
tes w
ill t
he
n b
e a
ske
d q
ue
stio
ns f
rom
on
e o
f th
e S
ec
tio
ns
4 -
7.
Th
is p
art
will
la
st
ap
pro
xim
ate
lyfo
ur
min
ute
s.
PA
RT
1In
terv
iew
Secti
on
1
Intr
od
ucti
on
In t
his
pa
rt y
ou
will
be
aske
d q
ue
stio
ns a
bo
ut
yo
ur
se
lf,
wh
ere
yo
u liv
e a
nd
wh
ere
yo
u s
tud
y o
r w
ork
.
Secti
on
2C
urr
en
t w
ork
In t
his
pa
rt y
ou
will
be
aske
d m
ore
de
tails
ab
ou
t w
ha
t yo
u d
o in
yo
ur
job
.
Secti
on
3C
urr
en
t stu
die
sIn
th
is p
art
yo
u w
ill b
e a
ske
d m
ore
de
tails
ab
ou
t yo
ur
stu
die
s.
Secti
on
4T
ravel
In t
his
pa
rt y
ou
will
be
aske
d a
bo
ut
pla
ce
s y
ou
ha
ve
vis
ite
d.
Secti
on
5L
an
gu
ag
e learn
ing
In t
his
pa
rt y
ou
will
be
aske
d a
bo
ut
yo
ur
exp
eri
en
ce
s o
f stu
dyin
g E
nglis
h.
Secti
on
6F
utu
re c
are
er
pro
sp
ects
In t
his
pa
rt y
ou
will
be
aske
d a
bo
ut
yo
ur
futu
re c
are
er
de
ve
lop
me
nt.
Secti
on
7In
tere
sts
In t
his
pa
rt y
ou
will
be
aske
d a
bo
ut
yo
ur
ho
bb
ies a
nd
in
tere
sts
.
Appendix 1.5
243
In P
art
3 o
f th
e t
est
yo
u w
ill b
e a
ske
d t
o t
ake
pa
rt in
a c
om
mu
nic
ative
activity w
ith
the
exa
min
er
wh
ich
is m
ad
e u
p o
f a
n in
form
atio
n e
xch
an
ge
an
d a
dis
cu
ssio
n.
Acco
rdin
g t
o y
ou
r w
ork
exp
erie
nce
, th
is a
ctivity m
ay b
e o
n a
ge
ne
ral w
ork
,te
ch
nic
al w
ork
, o
r stu
dy t
op
ic.
Th
is p
art
will
la
st
ap
pro
xim
ate
lyfo
ur
min
ute
s.
PA
RT
3C
om
mu
nic
ati
ve A
cti
vit
y
Co
nfe
ren
ce a
rran
gem
en
ts
Yo
u h
ave
on
e m
inu
te t
o r
ea
d t
hro
ugh
th
is t
ask.
Info
rmati
on
Exch
an
ge
Yo
u a
re m
akin
g t
he
arr
an
ge
me
nts
fo
r a
on
e-d
ay c
on
fere
nce
at
a lo
ca
l h
ote
l.
Th
eE
xa
min
er
is t
he
Co
nfe
ren
ce
Org
an
ise
r fo
r th
e h
ote
l a
nd
is v
isitin
g y
ou
to
dis
cu
ss
the
co
nfe
ren
ce
.
Fin
d o
ut
this
in
form
atio
ni)
the
siz
e o
f th
e la
rge
st
co
nfe
ren
ce
ro
om
ii)th
e c
ost
for
tha
t ro
om
iii)
equ
ipm
en
t a
va
ilab
le
Do
yo
u t
hin
k t
he
ho
tel is
off
erin
g y
ou
a g
oo
d s
erv
ice
fo
r th
e p
rice
it
is c
ha
rgin
g?
Dis
cu
ssio
n
No
w d
iscu
ss t
his
to
pic
with
th
e E
xa
min
er:
Wha
t mak
es a
suc
cess
ful c
onfe
renc
e?
244
Deutsch Version: DE00
Standard Test
Modelltest
Mitglied der Vereinigung von Sprachpr�fungsanbietern in Europa (ALTE)
APPENDIX 1.6
BULATS – Standard Test German
245
Appendix 1.6
3
Kan
dida
teninformation
Nam
e de
s Kan
dida
ten:
Fam
ilien
nam
e:…
……
……
……
……
……
…
Vor
nam
e:…
……
……
……
……
……
…
Kan
dida
tenn
umm
er:
……
……
……
……
……
……
Pr�fu
ngsz
entru
m:
……
……
……
……
……
……
Dat
um:
……
……
……
……
……
……
Test
Zeit:
ca.
110
Min
uten
H�rv
erst
ehen
Ca.
50
Min
uten
.Sc
hrei
ben
Sie
w�hr
end
des
H�re
ns Ih
re L�su
ngen
in d
as A
ufga
benb
latt.
Wen
n de
r Tes
t zum
Hör
vers
tehe
n be
ende
t ist
, hab
en S
ie 5
Min
uten
Zei
t,um
Ihre
Lösu
ngen
in d
en A
ntw
ortb
ogen
zu
schr
eibe
n.
Lese
vers
tehe
n&
60 M
inut
en.
Spra
chke
nntn
is
Schr
eibe
n Si
e Ih
re L�su
ngen
in d
en A
ntw
ortb
ogen
.
2
LEER
E SE
ITE
Appendix 1.6
246
5
3W
ann
will
Frau
Wol
f wie
der i
mBü
ro s
ein?
4W
ie s
ieht
Her
r Fel
dman
nau
s?
C
BA
Mon
tag
12.
09M
ittw
och
14.0
9
Don
ners
tag
15.0
9
C
BA
4
HÖ
RVE
RST
EHEN
TEIL
EIN
SAu
fgab
en 1
– 1
0
•Si
e h�
ren
zehn
kur
ze G
espr�
che.
•Kr
euze
n Si
e im
Auf
gabe
nbla
tt be
i den
Auf
gabe
n 1
– 10
die
richt
ige
Lösu
ngA
, B o
der
Can
. Es
gibt
nur
eine
rich
tige
Lösu
ng.
•Si
e h�
ren
jede
s G
espr�
chzw
eim
al.
1W
elch
en S
chre
ibtis
ch m
öcht
e Fr
au D
r. Sc
hmitt
für i
hr B
üro?
2W
elch
eW
ähru
ng b
rauc
ht d
er M
ann?
C
BA
C
BA
£$
€
Appendix 1.6
247
7
Sie
höre
n ei
n G
espr
äch.
Wel
chen
Flu
g so
ll da
s R
eise
büro
buc
hen?
AH
err H
ilper
t will
in N
ewYo
rk u
nd S
an F
ranc
isco
Zw
isch
enst
ops
mac
hen.
BH
err H
ilper
t will
jetz
t am
19.
Apr
il fli
egen
.
8
CH
err H
ilper
t will
non
stop
nac
h Lo
s An
gele
s fli
egen
.
Sie
höre
n ei
n G
espr
äch.
Wie
sol
ldie
neu
e H
omep
age
auss
ehen
?
AD
ie H
omep
age
soll
eine
zus
ätzl
iche
Sei
te b
ekom
men
.
BD
ie S
eite
zw
ei s
oll n
eu g
esta
ltet w
erde
n.
9
CD
er In
halt
soll
gekü
rzt w
erde
n.
Sie
höre
n ei
n G
espr
äch.
Was
sag
en d
ie b
eide
n Ko
llegi
nnen
?
ASi
e fin
den
den
Che
f zu
kriti
sch.
BSi
e gl
aube
n, d
ass
ein
Ges
präc
h Er
folg
hat
.
10
CSi
e w
olle
n m
it de
m C
hef s
prec
hen.
6
5W
as b
ekom
mt d
ie F
rau
des
Che
fs?
Sie
höre
n ei
n Te
lefo
nges
präc
h.W
ann
brau
cht H
err H
offm
ann
eine
Sek
retä
rin?
AH
eute
Nac
hmitt
ag
BM
orge
n N
achm
ittag
6
CSo
fort
Sie
höre
n im
Rad
io e
ine
Nac
hric
ht v
on e
iner
Mes
se.
Was
ist i
n di
esem
Jah
rand
ers?
AD
as W
ette
r
BD
ie A
nzah
l der
Auss
telle
r
7
CD
ie A
nzah
l der
Besu
cher
C
BA
Appendix 1.6
248
9
Ges
pr�ch
Dre
iAu
fgab
en 1
9 –
22
•Se
hen
Sie
sich
das
For
mul
ar u
nten
an.
•Si
e h�
ren
eine
Fra
u, d
ie e
ine
Rek
lam
atio
n m
acht
.
Tele
foni
sche
Rek
lam
atio
n
Nam
e:Fa
. Hau
ser
& Co
.Li
efer
ung
von:
(19)
……
……
……
..
Gru
nd d
erR
ekla
mat
ion:
(20)
……
……
……
..
Lief
erda
tum
: (21
) ……
……
……
……
……
……
……
……
……
Kund
en-N
r.: 10
28
Kund
enw
unsc
h:(2
2) …
……
……
…..
8
TEIL
ZW
EIAu
fgab
en 1
1 –
22
•Si
e h�
ren
drei
kur
ze T
elef
onge
sprä
che.
•Fü
llen
Sie
die
num
mer
ierte
n Lü
cken
in d
en F
orm
ular
en m
it de
n In
form
atio
nen
aus,
die
Sie
höre
n w
erde
n. S
chre
iben
Sie
Ihre
Lös
unge
n in
die
Form
ular
e, w
ähre
nd S
ie d
ieG
espr
äche
hör
en.
•Si
e hö
ren
jede
s G
espr
äch
nure
inmal
.
Ges
pr�ch
Ein
sAu
fgab
en 1
1 –
14
•Se
hen
Sie
sich
das
For
mul
ar u
nten
an.
•Si
e h�
ren
eine
Fra
u, d
ie e
inen
Auf
trag
gibt
.
Auftr
agsf
orm
ular
Nam
e:Li
a S
chrö
der
Gru
nd d
es A
nruf
s:(1
1) …
……
……
…..
Erre
ichb
ar z
wis
chen
:(12
) ……
……
……
..
Adre
sse:
(13)
……
……
……
……
……
……
……
……
……
……
……
.. Te
l.: 0
89/2
4560
7
Zahl
ungs
wun
sch:
(14)
.....
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
.
Ges
pr�ch
Zw
eiAu
fgab
en 1
5 –
18
•Se
hen
Sie
sich
das
For
mul
ar u
nten
an.
•Si
e h�
ren
eine
n M
ann,
der
ein
e Be
stel
lung
mac
ht.
Bes
tellu
ng
Nam
e:M
anfr
ed H
ausm
ann
Kund
ennu
mm
er:(
15) …
……
……
…..
Artik
el:
(16)
……
……
……
……
……
……
Nr.
L264
Kata
log:
(17)
……
……
……
…
Lief
erze
it:(1
8) …
……
……
…..
Appendix 1.6
249
11
TEIL
VIE
R
Absc
hnitt
Ein
sAu
fgab
en 3
3 –
38
•Si
e h�
ren
eine
n ku
rzen
Vor
trag
über
die
Eig
ensc
hafte
n ei
nes
gute
n Ve
rkäu
fers
.•
Kreu
zen
Sie
beid
en A
ufga
ben
33 –
38im
Auf
gabe
nbla
tt di
e ric
htig
e Lö
sung
A, B
ode
rCan
. Es
gib
t nur
ein
e ric
htig
e Lö
sung
.•
Sie
höre
n da
s G
espr
äch
zwei
mal
.
33Es
han
delt
sich
um
eine
Sch
ulun
gfü
r
AM
arke
ting-
Man
ager
.B
Mita
rbei
ter i
m C
all-C
ente
r.C
Mita
rbei
ter i
m V
ertri
eb.
34D
ie M
itarb
eite
rinne
n un
d M
itarb
eite
r hab
en in
Kris
enze
iten
eine
n si
cher
en J
ob, w
eil
Ada
s U
nter
nehm
en a
uf s
tark
en A
bsat
z an
gew
iese
n is
t.B
die
Firm
a vi
el G
eld
in ih
reSc
hulu
ngen
inve
stie
rtha
t.C
ihre
Geh
älte
r zie
mlic
h ni
edrig
sin
d.
35D
ie M
itarb
eite
rinne
n un
d M
itarb
eite
r sol
len
dara
uf a
chte
n,
Aob
der
Kun
de d
as G
espr
äch
abbr
eche
n m
öcht
e.B
ob d
er K
unde
noc
h un
sich
er is
t.C
wan
n de
r Kun
de b
erei
t ist
, etw
as z
u ka
ufen
.
36D
er R
edne
r sag
t, da
ss
Adi
e pe
rsön
liche
Wirk
ung
wic
htig
er is
t als
das
Pro
dukt
.B
der V
erka
ufse
rfolg
nur
vom
Pro
dukt
abh
ängt
.C
der V
erka
ufse
rfolg
sta
rk v
om A
uftre
ten
abhä
ngt.
37D
er R
edne
r sag
t auc
h, d
ass
AKe
nntn
isse
übe
r den
Ges
präc
hspa
rtner
seh
r nüt
zlic
h si
nd.
Bm
an d
en G
espr
ächs
partn
er n
ach
priv
aten
Din
gen
frage
n so
llte.
Cm
an G
espr
äche
übe
r Hob
bies
ver
mei
den
sollt
e.
38D
er R
edne
r sag
t auß
erde
m, d
ass
man
Abe
stim
mte
Fra
gete
chni
ken
gebr
auch
en s
ollte
.B
die
Frag
en d
em G
espr
ächs
partn
er ü
berla
ssen
sol
lte.
Cve
rsuc
hen
sollt
e, m
öglic
hst k
urze
Fra
gen
zu s
telle
n.
10
TEIL
DR
EI
Absc
hnitt
Ein
sAu
fgab
en 2
3 –
27
•Si
e h�
ren
fünf
kur
ze T
exte
. Hör
en S
ie z
u un
d en
tsch
eide
n Si
e, w
as d
ie L
eute
übe
r den
Geb
rauc
h de
s „d
u“ o
derd
es „S
ie“ s
agen
.•
W�
hlen
Sie
bei
jede
m T
ext I
hre
Antw
ort a
us d
er L
iste
A–I a
us u
nd s
chre
iben
Sie
den
richt
igen
Buc
hsta
ben
in d
ie le
ere
Stel
lehi
nter
der
Num
mer
.•
Sie
h�re
n je
den
Text
nur
einm
al.
Bei
spie
l: …
……
. I …
……
.
AJe
de F
irma
soll
eige
ne R
egel
nm
ache
n.
BD
as„S
ie“ f
ür a
lle s
chaf
ft kl
are
Verh
ältn
isse
.
23Te
xt 1
……
……
……
…..
CD
as „D
u“ is
t bes
serf
ür d
ie Z
usam
men
arbe
it.
24Te
xt 2
……
……
……
…..
DC
hefs
sol
lten
die
Mita
rbei
ter n
icht
duz
en.
25Te
xt 3
……
……
……
…..
ETr
aditi
onel
leBr
anch
en g
ebra
uche
n da
s „S
ie“.
26Te
xt 4
……
……
……
…..
FW
enn
man
sic
h du
zt, g
ibt e
s ke
inen
Res
pekt
.
27Te
xt 5
……
……
……
…..
GEs
spi
elt g
ar k
eine
Rol
lefü
r das
Arb
eits
klim
a.
HM
an k
ann
den
Che
fnic
ht d
uzen
.
IM
an s
oll n
ur s
eine
Fre
unde
duz
en.
Absc
hnitt
Zw
eiAu
fgab
en 2
8 –
32
•Si
e h�
ren
fünf
kur
ze T
exte
. Hör
en S
ie z
u un
d en
tsch
eide
n Si
e, w
as d
ie L
eute
übe
r ein
enAu
slan
dsau
fent
halt
sage
n.•
W�
hlen
Sie
bei
jede
m T
ext I
hre
Antw
ort a
us d
er L
iste
A–I a
us u
nd s
chre
iben
Sie
den
richt
igen
Buc
hsta
ben
in d
ie le
ere
Stel
lehi
nter
der
Num
mer
.•
Sie
h�re
n je
den
Text
nur
einm
al.
•B
eisp
iel:
……
…. I
……
….
AFi
rma
mus
s 1x
pro
Jah
r Flu
g na
ch H
ause
zah
len.
BBe
rufs
tätig
keit
der E
hefra
u sp
ielt
groß
e R
olle
.
28Te
xt 1
……
……
……
…..
CM
an m
uss
über
den
Ort
gena
u Be
sche
id w
isse
n.
29Te
xt 2
……
……
……
…..
DM
an m
uss
das
Klim
a gu
t ver
trage
n kö
nnen
.
30Te
xt 3
……
……
……
…..
EM
an m
uss
im A
usla
nd m
ehr G
eld
verd
iene
n.
31Te
xt 4
……
……
……
…..
FG
ute
Schu
lmög
lichk
eite
n fü
rdie
Kin
der.
32Te
xt 5
……
……
……
…..
GD
er A
ufen
thal
t sol
l max
imal
3 J
ahre
dau
ern.
HM
an m
uss
die
Land
essp
rach
e gu
t spr
eche
n.
ID
er A
ufen
thal
tmus
s gu
t für
die
Kar
riere
sei
n.
Appendix 1.6
250
13
Absc
hnitt
Dre
iAu
fgab
en 4
5 –
50
•Si
e h�
ren
ein
Ges
präc
h m
it Fr
au M
eure
r, di
e in
der
Per
sona
labt
eilu
ng e
iner
gro
ßen
Firm
aar
beite
t. Fr
au M
eure
r spr
icht
übe
r die
Kar
riere
chan
cen
in te
chni
sche
n Be
rufe
n.•
Kreu
zen
Sie
beid
en A
ufga
ben
45 –
50
im A
ufga
benb
latt
die
richt
ige
Lösu
ng A
, B o
derC
an.
Es g
ibt n
ur e
ine
richt
ige
Lösu
ng.
•Si
e hö
ren
das
Ges
präc
hzw
eim
al.
45Es
gib
t wen
ig B
ewer
berin
nen
fürt
echn
isch
e Be
rufe
, wei
l
Adi
e Fi
rmen
kei
ne S
telle
n fü
r Fra
uen
auss
chre
iben
.B
Frau
en d
ie A
rbei
t in
tech
nisc
hen
Beru
fen
zu s
chw
er fi
nden
.C
Frau
engl
aube
n, s
ie s
eien
für t
echn
isch
e Be
rufe
ung
eeig
net.
46Fr
au M
eure
r gla
ubt,
dass
Am
ännl
iche
Mita
rbei
ter d
ie b
esse
ren
Cha
ncen
hab
en.
Bte
chni
sche
Ber
ufe
imm
erei
ne D
omän
e de
r Män
nerb
leib
en.
Cte
chni
sche
Ber
ufe
beso
nder
s zu
kunf
tssi
cher
sin
d.
47In
den
tech
nisc
hen
Beru
fen
hat d
ie F
irma
von
Frau
Meu
rer
Afa
st k
eine
Bew
erbu
ngen
von
Fra
uen.
Bm
ehr F
raue
n al
s M
änne
rein
gest
ellt.
Cre
lativ
vie
le F
raue
n ei
nges
tellt
.
48D
ie F
irma
von
Frau
Meu
rer
Abi
etet
Fra
uen
inte
chni
sche
n Be
rufe
n hö
here
Geh
älte
r.B
hilft
Fra
uen
bei d
er F
inan
zier
ung
eine
r tec
hnis
chen
Aus
bild
ung.
Clä
sst F
raue
n in
tech
nisc
hen
Beru
fen
schn
elle
r auf
stei
gen.
49D
ie M
itarb
eite
rinne
n in
der F
irma
von
Frau
Meu
rer
Akö
nnen
Beru
fund
Fam
ilie s
chw
er m
itein
ande
r ver
einb
aren
.B
könn
en s
ich
ihre
Arb
eits
zeite
n in
divi
duel
l ges
talte
n.C
wün
sche
n si
ch m
ehrM
öglic
hkei
ten
der T
eilz
eita
rbei
t.
50Fr
au M
eure
r ist
der
Mei
nung
, das
s di
e
AFi
rmen
noc
h vi
el m
ehr f
ür ju
nge
Frau
en tu
n kö
nnte
n.B
jung
en F
raue
n ih
re C
hanc
en b
esse
r wah
rneh
men
müs
sten
.C
Lebe
nspa
rtner
die
jung
en F
raue
n ni
cht g
enüg
end
unte
rstü
tzen
.
Das
ist d
as E
nde
des T
ests
zum
Hör
vers
tehe
n. Ü
bert
rage
n Si
e nu
n Ih
re L
ösun
gen
auf
den
Ant
wor
tbog
en. S
ie h
aben
daz
u 5
Min
uten
Zei
t.
12
Absc
hnitt
Zw
eiAu
fgab
en 3
9 –
44
•Si
e h�
ren
ein
Ges
präc
h zw
isch
en z
wei
Mita
rbei
tern
übe
r ein
Pro
blem
in ih
rer F
irma.
•Kr
euze
n Si
e be
iden
Auf
gabe
n39
–44
im A
ufga
benb
latt
die
richt
ige
Lösu
ng A
, B o
derC
an.
Es g
ibt n
ur e
ine
richt
ige
Lösu
ng.
•Si
e hö
ren
das
Ges
präc
hzw
eim
al.
39Fr
au S
cher
er u
nd H
err H
uber
Abe
klag
en s
ich
über
ihre
vie
len
Term
ine.
Bsp
rech
en ü
ber d
ie E
inha
ltung
von
Lie
ferte
rmin
en.
Cüb
erle
gen,
wie
sie
Kun
den
gew
inne
n kö
nnen
.
40Es
gab
Pro
blem
e be
i der
Verp
acku
ng d
erW
aren
, wei
l
Adi
e G
läse
r nic
ht in
die
Kar
tons
pas
sten
.B
dort
viel
e Au
shilf
en a
rbei
tete
n.C
eine
Mas
chin
e ni
cht f
unkt
ioni
erte
.
41Im
Aus
liefe
rung
slag
er g
ab e
s Sc
hwie
rigke
iten,
wei
l
Ade
r Lag
erm
eist
erer
kran
kte.
Bdo
rt Pe
rson
alm
ange
l her
rsch
t.C
fals
che
Lief
ersc
hein
e au
sges
tellt
wur
den.
42H
err H
uber
krit
isie
rt an
dem
Tra
nspo
rtunt
erne
hmen
, das
s
Ade
r neu
e C
hef k
eine
Erfa
hrun
g ha
t.B
es G
ewin
ne m
ache
n w
ill.C
es z
u vi
ele
Auftr
äge
anni
mm
t.
43H
err H
uber
sch
lägt
vor
,
Afir
men
eige
ne L
KWs
anzu
scha
ffen.
Bm
it de
m n
euen
Firm
ench
ef d
as P
robl
em z
u be
spre
chen
.C
mit
eine
m a
nder
enTr
ansp
ortu
nter
nehm
en z
usam
men
zuar
beite
n.
44Fr
au S
cher
er b
ittet
Her
rn H
uber
,
Aei
ne K
alku
latio
n de
rTra
nspo
rtkos
ten
zu e
rste
llen.
Bm
it m
ehre
ren
Tran
spor
tfirm
enKo
ntak
tauf
zune
hmen
.C
sich
für e
in n
eues
Tra
nspo
rtunt
erne
hmen
zu
ents
chei
den.
Appendix 1.6
251
15
53
Lieb
e K
olle
ginn
en u
nd K
olle
gen,
Uns
ere
für l
etzt
en M
ittw
och
gepl
ante
Abt
eilu
ngsb
espr
echu
ng, d
ie w
egen
Kra
nkhe
itau
sfal
len
mus
ste,
kan
n nu
n au
ch d
iese
Woc
he n
icht
stat
tfind
en, d
a ic
h am
Mitt
woc
hei
nen
wic
htig
en a
nder
en T
erm
in h
abe.
Best
e G
rüße
H. K
unstm
ann
Die
Abt
eilu
ngsb
espr
echu
ngfin
det d
iese
Woc
he n
icht
sta
tt, w
eil H
err K
unst
man
n
Aei
nen
ande
ren
Term
in v
orsc
hläg
t.B
kein
e Ze
it ha
t.C
kran
k is
t.
54
Lieb
e M
itarb
eite
r und
Mita
rbei
terin
nen,
Die
Per
sona
labt
eilu
ng w
eist
noc
h ei
nmal
dar
auf
hin,
das
s al
le R
estu
rlaub
e bi
s En
deA
pril
des
Folg
ejah
res
geno
mm
en s
ein
müs
sen.
Nur
in
Aus
nahm
efäl
len,
wen
n au
sbe
trieb
liche
n G
ründ
en d
ringe
nd e
rfor
derli
ch,
kann
ein
Urla
ubsa
nspr
uch
auf
eine
nsp
äter
en Z
eitp
unkt
, sp
ätes
tens
jed
och
bis
Ende
Aug
ust
des
Folg
ejah
res,
über
trage
nw
erde
n.
Die
Mita
rbei
terin
nen
und
Mita
rbei
ter
Akö
nnen
ein
enTe
il de
s U
rlaub
s im
näc
hste
n Fr
ühja
hr n
ehm
en.
Bkö
nnen
ihre
n ga
nzen
Urla
ub a
uf d
as n
ächs
te J
ahr v
ersc
hieb
en.
Cm
üsse
n ih
ren
Urla
ub im
lauf
ende
n Ka
lend
erja
hrne
hmen
.
55
Lieb
e K
olle
ginn
en,
Frau
Mar
tin h
at si
ch w
iede
rhol
t bes
chw
ert ü
ber s
chm
utzi
ges G
esch
irr, d
as in
der
Küc
he h
erum
steh
t. Ic
h de
nke,
sie
hat R
echt
und
wir
sollt
en d
iese
s Pro
blem
irge
ndw
iere
geln
. Ich
schl
age
vor,
wir
setz
en u
ns m
orge
n 11
Uhr
kur
z zu
sam
men
und
übe
rlege
nge
mei
nsam
, wie
wir
dies
es u
nlie
bsam
e Th
ema
aus d
er W
elt s
chaf
fen
könn
en.
Wilm
a Fr
itz
Frau
Frit
z sc
hläg
t vor
,
Ada
s G
esch
irr g
emei
nsam
zu
spül
en.
Bsi
ch ü
ber F
rau
Mar
tin z
u be
schw
eren
.C
sich
zu
treffe
n un
d da
s Pr
oble
m z
u di
skut
iere
n.
14
LESE
VER
STEH
EN u
nd S
PRAC
HK
ENN
TNIS
TEIL
EIN
S
Absc
hnitt
Ein
sAu
fgab
en 5
1 –
57
•Le
sen
Sie
die
folg
ende
n N
otiz
en u
nd N
achr
icht
en.
•M
arki
eren
Sie
bei
den
Auf
gabe
n 51
– 5
7im
Ant
wor
tbog
en d
ie ri
chtig
e Lö
sung
A, B
oder
C. E
s gi
bt n
urei
ne ri
chtig
e Lö
sung
.
Bei
spie
l: DE
N IN
HA
LT
K�
HL
LA
GE
RN
UN
D V
OR
SO
NN
EN
LIC
HT
SC
H�
TZE
N
AD
er In
halt
mus
s be
i gle
ichb
leib
ende
r Tem
pera
tur l
ager
n.B
Der
Inha
lt is
t em
pfin
dlic
h ge
gen
Hitz
e un
d Li
cht.
CD
er In
halt
mus
s ge
frore
n la
gern
.
AB
C0
51O
ffizi
elle
Brie
fesc
hrei
ben
leic
ht g
emac
htD
er
Rat
gebe
rm
it Be
ispi
elen
, An
regu
ngen
un
d M
uste
rtext
en
für
Anfra
gen,
Rek
lam
atio
nen,
R
echn
unge
n.
Best
elle
n Si
e no
ch
heut
eun
sere
ko
sten
lose
Bros
chür
e.
Die
sen
Rat
gebe
r bek
omm
en S
ie
Abe
i Rek
lam
atio
nen.
Bge
gen
Rec
hnun
g.C
umso
nst.
52 Die
Deu
tsch
en k
aufe
n im
Inte
rnet
Am
ehr K
leid
er a
ls C
Ds.
Bvo
r alle
m R
eise
n.C
wen
iger
Büc
her a
ls C
ompu
ter.
Eink
äufe
imIn
tern
et
Reise
n34
%
Kleid
ung
15%
Com
pute
r15
%
CDs
21%
Büch
er15
%
Reise
n
Kleid
ung
Comp
uter
CDs
Büch
er
Lösu
ng:
Appendix 1.6
252
17
TEIL
EIN
S
Absc
hnitt
Zw
eiAu
fgab
en 5
8 –
63
•W�
hlen
Sie
das
Wor
t, da
s am
bes
ten
in d
ie L�
cke
im S
atz
pass
t.•
Mar
kier
en S
ie b
ei d
en A
ufga
ben
58 –
63
im A
ntw
ortb
ogen
die
ric
htig
eLö
sung
A, B
, Cod
er D
. Es
gib
t nur
ein
e ric
htig
e Lö
sung
.
Dur
ch…
……
……
…..
Prei
se g
ewin
nen
wir
wah
rsch
einl
ich
neue
Kun
den.
Ani
edrig
enB
nied
riger
Cni
edrig
ere
58
Dni
edrig
eren
Das
Sem
inar
war
lang
wei
lig. I
ch h
abe
……
……
……
.. N
eues
erfa
hren
.
Afa
stB
kein
Cni
cht
59
Dni
chts
Wei
l der
Fah
rer e
inen
Unf
allh
atte
, kam
die
War
e zu
……
……
……
.. .
Aal
tB
kapu
ttC
lang
sam
60
Dsp
ät
Näc
hste
Woc
he b
ekom
men
wir
eine
……
……
……
.. Ko
llegi
n.
Ane
uB
neue
Cne
uere
61
Dne
uere
s
Der
Brie
f mus
s er
st m
orge
n ra
us. S
ie k
önne
n si
ch a
lso
ruhi
g et
was
……
……
…..
lass
en.
AM
omen
tB
Paus
eC
Ruh
e
62
DZe
it
Ich
halte
die
se F
irma
……
……
……
.. se
hr g
ut. S
ie k
önne
n ih
r den
Auf
trag
gebe
n.
Aal
sB
best
imm
tC
f�r
63
Dw
ie
16
56
051015202530
1. Q
rtl.
2. Q
rtl.
3. Q
rtl.
4. Q
rtl.
Um
satz
u. G
ewin
n
Um
satz
Gew
inn
Der
Um
satz
Ais
t im
2. Q
uarta
l ges
tiege
n.B
ist i
m 3
. Qua
rtal s
o ho
ch w
ie im
4. Q
uarta
l.C
ist i
m 4
. Qua
rtal h
öher
als
im1.
Qua
rtal.
57
Sehr
gee
hrte
Fra
u Sc
hmid
t,
es tu
t uns
sehr
Lei
d, d
ass u
nser
e Li
efer
ung
nich
t zu
Ihre
r Zuf
riede
nhei
t aus
gefa
llen
ist.
Selb
stve
rstä
ndlic
h ne
hmen
wir
die
Send
ung
zurü
ck u
nd sc
hrei
ben
Ihne
n de
nR
echn
ungs
betra
ggu
t. Fa
lls S
ie m
it ei
nem
and
eren
Arti
kel e
inve
rsta
nden
sind
, geb
ensi
e un
s bitt
e so
schn
ell w
ie m
öglic
h N
achr
icht
. Wir
wer
den
ihn
so sc
hnel
l wie
mög
lich
schi
cken
.
Mit
freu
ndlic
hen
Grü
ßen
Tita
n H
ande
lsco
ntor
Tita
n H
ande
lsco
ntor
Aka
nn d
en R
echn
ungs
betra
g ni
cht g
utsc
hrei
ben.
Bka
nn e
inen
and
eren
Arti
kel l
iefe
rn.
Cm
öcht
e G
ründ
e fü
rdie
Rüc
ksen
dung
wis
sen.
Appendix 1.6
253
19
64D
ie a
usge
schr
iebe
ne S
telle
ver
lang
t
Aei
ne a
bges
chlo
ssen
eBe
rufs
ausb
ildun
g.B
gute
Spr
achk
ennt
niss
e in
zw
ei F
rem
dspr
ache
n.C
kein
e frü
here
n be
rufli
chen
Erfa
hrun
gen.
65D
ie B
ewer
berin
gla
ubt,
dass
Adi
e An
ford
erun
gen
hoch
sin
d.B
sie
den
Anfo
rder
unge
n ge
nügt
.C
sie
die
Anfo
rder
unge
n ni
cht g
enau
ken
nt.
66D
ie F
rem
dspr
ache
nken
ntni
sse
der B
ewer
berin
sin
d
Ain
Eng
lisch
bes
ser a
ls in
Fra
nzös
isch
.B
in F
ranz
ösis
ch b
esse
r als
in It
alie
nisc
h.C
in F
ranz
ösis
ch n
icht
aus
reic
hend
.
67D
ie B
ewer
berin
hat
frühe
r
AD
iens
treis
en fü
r Firm
en o
rgan
isie
rt.B
indi
vidu
elle
Rei
sen
für F
irmen
mita
rbei
ter o
rgan
isie
rt.C
Rei
sen
fürG
rupp
en o
rgan
isie
rt.
68D
ie B
ewer
berin
Ais
t mit
ihre
m A
rbei
tspl
atz
nich
t zuf
riede
n.B
möc
hte
aus
pers
önlic
hen
Grü
nden
nac
h O
snab
rück
.C
woh
nt je
tzt i
n O
snab
rück
und
möc
hte
weg
zieh
en.
69D
ie F
irma,
die
die
Ste
lle a
usge
schr
iebe
n ha
t, is
t
Adi
e D
euts
che
Bund
esba
hn.
Bei
n To
uris
tikun
tern
ehm
en.
Cei
ne F
lugg
esel
lsch
aft.
18
TEIL
EIN
S
Absc
hnitt
Dre
iAu
fgab
en 6
4 –
69
•Le
sen
Sie
den
folg
ende
Tex
t auf
mer
ksam
dur
ch u
nd b
eant
wor
ten
Sie
die
Aufg
aben
64
– 69
auf d
er g
egen
�be
rlieg
ende
n Se
ite.
•M
arki
eren
Sie
bei
den
Auf
gabe
n 64
– 6
9im
Ant
wor
tbog
en d
ie ri
chtig
e lö
sung
A, B
ode
rC
. Es
gib
t nur
ein
e ric
htig
e lö
sung
.
Sehr
gee
hrte
Dam
en, s
ehr g
eehr
te H
erre
n,
Ihre
inde
r Süd
deut
sche
n Ze
itung
ers
chie
nene
Ste
llena
nzei
ge h
at m
ich
sehr
ang
espr
oche
n.Si
e su
chen
ein
e zu
verlä
ssig
e M
itarb
eite
rin, d
ie s
ich
durc
h Fl
exib
ilität
aus
zeic
hnet
, übe
r gut
eita
lieni
sche
Spr
ache
nken
ntni
sse
und
auße
rdem
übe
r Erfa
hrun
gin
der P
lanu
ngvo
nG
rupp
enre
isen
ver
fügt
.
Wei
terh
in w
ünsc
hen
Sie
Erfa
hrun
g im
Um
gang
mit
dem
PC
-Pro
gram
m S
TAR
Tun
d na
türli
chei
ne a
bges
chlo
ssen
e Be
rufs
ausb
ildun
g al
s R
eise
kauf
frau.
Ich
glau
be, d
ass
ich
alle
die
se g
esch
ilder
ten
Anfo
rder
unge
n er
fülle
.
Ich
bin
eins
atzf
reud
ig u
nd e
ngag
iert,
hab
e m
eine
Ital
ieni
sch-
Kenn
tnis
se a
n de
rD
olm
etsc
hers
chul
e in
Hei
delb
erg
erw
orbe
n un
d ve
rfüge
auc
h üb
er s
ehr g
ute
Engl
isch
-Ke
nntn
isse
inW
ort u
nd S
chrif
t sow
ie ü
ber b
rauc
hbar
e Fr
anzö
sisc
h-Ke
nntn
isse
.
Wäh
rend
mei
ner 3
-jähr
igen
Ber
ufst
ätig
keit
war
ich
vor a
llem
mit
der O
rgan
isat
ion
von
Gru
ppen
reis
en b
etra
ut. I
ch h
abe
beso
nder
en S
paß
am U
mga
ngm
it M
ensc
hen,
auf
der
enW
ünsc
he u
nd B
edür
fnis
se ic
h m
ich
gut e
inst
elle
n ka
nn. F
ür a
lle v
on m
ir or
gani
sier
ten
Rei
sen
gab
es b
eson
ders
gut
e R
ückm
eldu
ngen
.
Aus
fam
iliäre
n G
ründ
en p
lane
ich,
nac
h O
snab
rück
um
zuzi
ehen
und
hab
e de
shal
b m
eine
nje
tzig
en A
rbei
tspl
atz
gekü
ndig
t.
An Ih
rem
Unt
erne
hmen
gef
ällt
mir
beso
nder
s, d
ass
es e
in m
ittle
res
Unt
erne
hmen
mit
eine
rüb
ersc
haub
aren
Mita
rbei
terz
ahl i
st. I
ch g
laub
e, m
eine
Fäh
igke
iten
so b
esse
r ent
falte
n zu
könn
en a
ls in
ein
em s
ehr g
roße
n U
nter
nehm
en.
Ger
ne w
ürde
ich
mei
nen
Beru
fsw
eg b
ei Ih
nen
forts
etze
n. F
alls
Sie
mei
ne B
ewer
bung
ansp
richt
, tei
len
Sie
mir
bitte
ein
en T
erm
infü
r ein
per
sönl
iche
s G
espr
äch
mit.
Mit
freun
dlic
hen
Grü
ßen
Ilona
Sch
reib
er
Appendix 1.6
254
21
LESE
VER
STEH
EN u
nd S
PRAC
HK
ENN
TNIS
Teil
Zwei
20
TEIL
EIN
S
Absc
hnitt
Vie
rAu
fgab
en 7
0 –
74
•Le
sen
Sie
den
folg
ende
n Te
xt u
nd�
berle
gen
Sie
sich
f�r j
ede
der L
�ck
en e
in p
asse
ndes
Wor
t.•
Schr
eibe
n Si
e f�
r die
L�
cken
70
– 74
jew
eils
nur
ein
Wor
t in
den
Antw
ortb
ogen
.
Bei
spie
l:
Er i
st s
ehr…
……
.……
…. C
ompu
tern
inte
ress
iert.
Lösu
ng:
Ihre
Anz
eige
in d
er F
rank
furt
er A
llgem
eine
n Ze
itung
Aus
Ihre
r Anz
eige
ent
nehm
e ic
h, (
70) …
……
……
…..
Ihr U
nter
nehm
en ta
lent
ierte
F�hr
ungs
kräf
te s
ucht
. Ic
h ha
be m
ein
Stud
ium
der
Betri
ebsw
irtsc
hafts
lehr
e vo
r kur
zem
abge
schl
osse
n un
d su
che
jetz
t ein
e ve
rant
wor
tung
svol
le T
ätig
keit
(71)
……
……
……
..fr�
hest
mög
liche
n Ze
itpun
kt.
Ich
glau
be, d
ass
ich
den
besc
hrie
bene
n Au
fgab
en a
ufgr
und
(72)
……
……
……
..Q
ualif
ikat
ione
n en
tspr
eche
. W
ähre
nd m
eine
s St
udiu
ms
habe
ich
ein
drei
mon
atig
esPr
aktik
um(7
3) …
……
……
…..
eine
r Firm
a in
den
USA
ver
brac
ht.
Des
halb
ver
f�ge
ich
�be
r seh
r gut
e En
glis
chke
nntn
isse
inW
ort u
nd S
chrif
t.
Neb
en K
reat
ivitä
t geh
ören
Bel
astb
arke
it, V
eran
twor
tung
sbew
usst
sein
und
Team
fähi
gkei
t (7
4) …
……
……
…..
mei
nen
Eige
nsch
afte
n.
0an
Appendix 1.6
255
23
A
Schr
eibe
n Si
e Ih
ren
Bus
ines
spla
n on
line.
Das
ler
nen
Sie
in u
nser
emac
htw
öchi
gen
Wor
ksho
p. W
ir er
läut
ern
Ihne
n, w
elch
e In
halte
das
Pap
ier
habe
n m
uss
und
helfe
n Ih
nen,
Ihre
n in
divi
duel
len
Plan
ein
satz
ferti
g zu
ers
telle
n. N
ach
acht
Woc
hen
verf
ügen
Sie
übe
rIh
ren
Pla
n un
d ha
ben
glei
chze
itig
das
Kno
w h
ow e
rwor
ben,
um
in d
er Z
ukun
ft Ih
re P
läne
selb
stst
ändi
g er
stel
len
zukö
nnen
. Das
alle
s via
Inte
rnet
ohn
e pe
rsön
liche
Zusa
mm
enku
nft.
B
Könn
en S
ie Ih
re e
igen
en F
ähig
keite
n be
im V
orst
ellu
ngsg
espr
äch
richt
igve
rmitt
eln?
Wis
sen
Sie,
nac
h w
elch
en K
riter
ien
Pers
onal
entw
ickl
er B
ewer
ber
ausw
ähle
n? B
eide
s is
t ein
wic
htig
er S
chrit
t zum
Wun
schj
ob. D
iese
Ken
ntni
sse
verm
ittel
t Ihn
en u
nser
dre
itägi
ges
Sem
inar
in M
ünch
en „S
ehen
und
ges
ehen
wer
den“
für S
tude
nten
vor
der
Job
-Suc
he. S
ie k
önne
n w
ähle
n zw
isch
en z
wei
Term
inen
jähr
lich.
Für
wei
tere
Info
rmat
ione
n st
ehen
wir
gern
e zu
r Ver
fügu
ng.
C
Im „Wissensforum“ der Frankfurter Allgemeinen Zeitung setzen
wir unsere Veranstaltungsreihe fort. An zwölf Abenden
referieren Prominente Persönlichkeiten über verschiedene
Aspekte der Themen „Erfolg“ und „lebenslanges Lernen“. Unsere
Veranstaltungen können en bloc oder auch einzeln gebucht
werden. Reservierungen unter 069/ 25 36 08 bei Frau Schröder.
D
Möc
hten
Sie
in e
inem
uns
erer
eur
opäi
sche
n N
achb
arlä
nder
lebe
n un
d do
rtau
ch Ih
ren
Lebe
nsun
terh
alt v
erdi
enen
? D
ann
besu
chen
Sie
uns
er e
inw
öchi
ges S
emin
ar „
fitfü
r Eur
opa“
, das
Ihne
n fu
ndie
rte E
inbl
icke
in u
nser
e N
achb
arlä
nder
ver
mitt
elt.
Von
Woh
nung
ssuc
he ü
ber A
rbei
tsre
cht b
is z
ur P
erso
nalfü
hrun
g,w
erde
n in
uns
erem
Sem
inar
alle
wic
htig
en A
spek
te a
nges
proc
hen.
Wir
wür
den
Sie
gern
e al
s Tei
lneh
mer
beg
rüße
n.
22
TEIL
ZW
EI
Absc
hnitt
Ein
sAu
fgab
en 7
5 –
81
•Le
sen
Sie
die
folg
ende
n Äu
ßeru
ngen
und
die
vie
r Anz
eige
n au
f der
geg
enüb
erlie
gend
enSe
ite a
ufm
erks
amdu
rch.
•W
elch
e An
zeig
e pa
sst z
u w
elch
er d
er Ä
ußer
unge
n 75
– 8
1?•
Mar
kier
en S
ie im
Ant
wor
tbog
en je
wei
lsei
nen
Buch
stab
en A
, B, C
ode
rD fü
r die
rich
tige
Anze
ige.
Bei
spie
l:
0Si
e be
kom
men
im S
emin
ar e
in fe
rtige
s Ko
nzep
t.
Lösu
ng:
0A
B
C
D
75D
iese
s Se
min
arfin
det i
n M
ünch
en s
tatt.
76Si
e w
olle
n im
Aus
land
arb
eite
n un
d su
chen
ein
Sem
inar
zur
Vor
bere
itung
.
77Si
e ha
ben
wen
ig Z
eit u
nd s
uche
n de
shal
bei
n vi
rtuel
les
Sem
inar
.
78D
as S
emin
arfin
det a
n ei
nzel
nen
Aben
den
stat
t.
79D
as S
emin
arfin
det z
wei
mal
im J
ahr s
tatt.
80Si
e su
chen
ein
Sem
inar
, bei
dem
man
nic
ht a
lle T
erm
ine
buch
en m
uss.
81Si
e m
öcht
en le
rnen
, wie
man
sic
h am
best
en p
räse
ntie
rt.
Appendix 1.6
256
25
TEIL
ZW
EI
Absc
hnitt
Dre
iAu
fgab
en 8
7 –
91
•Le
sen
Sie
den
folg
ende
n Te
xt u
nd�
berle
gen
Sie
sich
f�r j
ede
der L
�ck
en e
in p
asse
ndes
Wor
t.•
Schr
eibe
n Si
e f�
r die
L�
cken
87 –
91
jew
eils
nure
in W
ort i
n de
n An
twor
tbog
en.
Bei
spie
l:
Er i
st s
ehr…
……
.……
…. C
ompu
tern
inte
ress
iert.
Lösu
ng:
Ein
Sem
inar
zu
Zeit-
und
Sel
bstm
anag
emen
t
Der
Le
itspr
uch,
de
r si
ch
durc
h da
s ga
nze
Sem
inar
zo
g,w
ar:
Sich
ni
cht
verz
ette
ln,
(87)
……
……
……
.. kl
are
Prio
rität
en
setz
en!
Anha
nd
(88)
……
……
……
.. m
ehrs
eitig
enAr
beits
blat
tes
fand
en
die
Sem
inar
teiln
ehm
er
hera
us,
in
wel
chen
dr
ei
Bere
iche
n ih
res
Zeitm
anag
emen
ts fü
rsie
der
grö
ßte
Han
dlun
gsbe
darf
(89)
……
……
……
.. .
Die
wic
htig
ste
Erke
nntn
is f
ür a
lle T
eiln
ehm
er d
es S
emin
ars:
Nur
(90)
……
……
……
.. si
chda
rübe
r im
Kla
ren
ist,
woh
in e
r in
sein
em B
eruf
und
inse
inem
Priv
atle
ben
lang
frist
ig w
ill,de
rist
auch
in d
er(9
1) …
……
……
…..
, bei
den
tägl
iche
n Ve
rrich
tung
en P
riorit
äten
zu
setz
en.
0an
24
TEIL
ZW
EI
Absc
hnitt
Zw
eiAu
fgab
en 8
2 –
86
•Le
sen
Sie
den
folg
ende
nTe
xt a
ufm
erks
am d
urch
und
wäh
len
Sie
für
jede
Lüc
ke d
aspa
ssen
de W
orta
usde
r Lis
te u
nten
aus
.•
Mar
kier
en S
ie fü
r di
e Lü
cken
82
– 86
im A
ntw
ortb
ogen
die
ric
htig
e Lö
sung
A, B
, C o
der
D. E
s gi
bt n
urei
ne ri
chtig
e Lö
sung
.
Bei
spie
l:
Er m
�ch
te, d
ass
Sie
ihm
den
Gru
nd (0
) ……
……
……
...
0A
ansa
gen
Bsa
gen
Cau
ssag
enD
vers
agen
Lösu
ng:
0A
B
C
D
Die
erf
olgr
eich
eB
espr
echu
ng
Wen
n Si
e zu
ei
ner
Besp
rech
ung
eing
elad
en
wer
den,
so
llten
Si
e(8
2) …
……
……
…..
best
ehen
,da
ss s
ie p
ünkt
lich
begi
nnt
und
sich
mel
den,
wen
n Si
e (8
3) …
……
……
…..
, da
ss d
ie D
isku
ssio
n zu
wei
t vo
m T
hem
aab
schw
eift.
(84
) ……
……
……
.. d
erje
nige
, de
rein
gela
den
hat,
die
Kont
rolle
über
die
Bes
prec
hung
ver
liert,
sol
lten
Sie
ihn
dara
n er
inne
rn,
dass
die
Zei
tw
eglä
uft u
nd S
ie n
och
viel
vor
sic
h ha
ben
oder
nac
h de
r(8
5) …
……
……
…..
Besp
rech
ung
noch
wic
htig
e Au
fgab
en z
u er
ledi
gen
habe
n. M
ache
n Si
e al
soim
mer
kla
r, da
ss S
ie (
86) …
……
……
…..,
si
ch a
n di
eTa
geso
rdnu
ngzu
halte
n.
82A
dage
gen
Bda
mit
Cda
rauf
Dda
rübe
r
83A
beda
uern
Bbe
mer
ken
Cbe
schl
ieße
nD
beto
nen
84A
Nac
hdem
BO
bwoh
lC
Wäh
rend
DW
enn
85A
folg
ende
nB
kom
men
den
Cla
ufen
den
Dnä
chst
en
86A
vorg
eben
Bvo
rgeh
enC
vorh
aben
Dvo
rneh
men
Appendix 1.6
257
27
Her
r Mai
er is
t noc
h in
ein
er B
espr
echu
ng.
……
……
……
..er
Sie
in e
iner
Stu
nde
zurü
ckru
fen?
AKa
nnB
Möc
hte
CM
uss
95
DW�rd
e
Beid
eine
nÜ
berle
gung
en s
ollte
st d
u di
e W
echs
elku
rse
nich
t auß
er A
cht …
……
……
…..
.
Aha
ben
Bla
ssen
Cne
hmen
96
Dse
tzen
Ohn
e se
ine
Hilf
e …
……
……
…..
wir
nie
so s
chne
llfe
rtig
gew
orde
n.
Ahä
tten
Bsi
ndC
wär
en
97
Dw
ürde
n
26
TEIL
ZW
EI
Absc
hnitt
Vie
rAu
fgab
en 9
2 –
97
•W
ähle
n Si
e da
s W
ort,
das
am b
este
n in
die
Lüc
ke im
Sat
z pa
sst.
•M
arki
eren
Sie
bei
den
Auf
gabe
n 92
– 9
7im
Ant
wor
tbog
en d
ie r
icht
ige
Lösu
ng A
, B, C
oder
D.
Es g
ibt n
ur e
ine
richt
ige
Lösu
ng.
Das
im le
tzte
n Ja
hrei
ngef
ührte
Pro
dukt
erfr
eut s
ich
zune
hmen
der
……
……
……
.. .
ABe
deut
ung
BBe
fried
igun
gC
Begr
üßun
g
92
DBe
liebt
heit
Sie
fühl
t sic
h …
……
……
…..
ihre
rneu
en F
irma
sehr
woh
l.
Aau
fB
inC
mit
93
D�
ber
Uns
er C
hef w
eiß
mit
sein
en M
itarb
eite
rn g
ut …
……
……
…..
.
Aum
gehe
nB
umge
hen
soll
Cum
zuge
hen
94
Dw
ie u
mge
hen
Appendix 1.6
258
29
98In
der
Stu
die
wur
de e
rfors
cht,
ob d
ie E
ntse
ndun
g vo
n M
itarb
eite
rn in
s Au
slan
d
Ade
n M
itarb
eite
rn s
chad
et.
Bfü
r die
Firm
a nü
tzlic
h is
t.C
richt
ig g
epla
nt w
ird.
Dzu
vie
l Gel
d ko
stet
.
99Be
i der
Stud
ie h
ande
lt es
sic
h um
eine
Afir
men
inte
rne
Stud
ie.
BU
mfra
ge b
eiBa
nken
und
Pha
rmau
nter
nehm
en.
Cw
isse
nsch
aftli
che
Stud
ie.
DZe
itung
sum
frage
.
100
Die
unt
ersu
chte
n U
nter
nehm
en s
chic
ken
Mita
rbei
ter i
ns A
usla
nd, d
amit
sie
AFr
emds
prac
hen
lern
en.
Bim
Aus
land
Filia
len
aufb
auen
.C
inte
rnat
iona
le E
rfahr
unge
n sa
mm
eln.
Dne
ue K
unde
n w
erbe
n.
101
Die
Mita
rbei
ter h
aben
im A
usla
nd E
rken
ntni
sse
gew
onne
n üb
er
Ada
s D
euts
chla
ndbi
ld d
erau
slän
disc
hen
Partn
er.
Bdi
e Kr
editw
ürdi
gkei
t aus
länd
isch
er P
artn
er.
Cdi
e w
irtsc
haftl
iche
Uns
iche
rhei
t aus
länd
isch
er P
artn
er.
Dpo
tent
ielle
neue
Ges
chäf
tsbe
reic
he fü
r die
Firm
a.
102
Die
zur
ückg
ekeh
rten
Mita
rbei
ter w
aren
frus
trier
t, w
eil s
ie
Ake
ine
Befö
rder
ung
auf e
ine
bess
ere
Stel
le b
ekam
en.
Bm
ehrI
nter
esse
an
ihre
n Er
fahr
unge
n er
war
tet h
atte
n.C
sich
an
meh
r Int
erna
tiona
lität
gew
öhnt
hat
ten.
Dsi
ch in
ihre
r jet
zige
n St
elle
eing
eeng
t füh
lten.
103
Die
Firm
enle
itung
en
Apl
anen
Maß
nahm
en, u
mdi
e Si
tuat
ion
zu v
erbe
sser
n.B
reag
ierte
n ni
cht a
uf d
ie E
rgeb
niss
e de
rStu
die.
Cw
isse
n ni
cht,
wie
sie
die
Situ
atio
n ve
rbes
sern
sol
len.
Dw
olle
n ke
ine
Mita
rbei
ter m
ehr i
ns A
usla
nd s
chic
ken.
28
TEIL
ZW
EI
Absc
hnitt
F�
nfAu
fgab
en 9
8 –
103
•Le
sen
Sie
den
folg
ende
n Te
xt a
ufm
erks
am d
urch
und
löse
n Si
e di
e Au
fgab
en a
uf d
erge
genü
berli
egen
den
Seite
.•
Mar
kier
en S
ie b
ei d
en A
ufga
ben
98 –
103
im A
ntw
ortb
ogen
die
ric
htig
e Lö
sung
A, B
, Cod
erD
. Es
gib
t nur
ein
e ric
htig
e Lö
sung
.
Das
ver
schw
ende
te K
apita
l
Deu
tsch
e Fi
rmen
sch
icke
n zw
arvi
ele
Mita
rbei
terf
ür te
ures
Gel
d in
s Au
slan
d, d
eren
Wis
sen
nutz
en s
ie a
bern
icht
und
ver
schw
ende
n da
mit
wer
tvol
le R
esso
urce
n. D
as is
t das
Erg
ebni
sei
ner S
tudi
e de
s W
isse
nsch
afts
zent
rum
s Be
rlin,
übe
r die
in d
erhe
utig
en A
usga
be v
onW
irtsc
haft
Heu
te b
eric
htet
wird
.
Für d
ie S
tudi
e w
urde
n zw
eigr
oße
deut
sche
Firm
en u
nter
such
t: ei
ne P
harm
a-Fi
rma
mit
20.0
00 M
itarb
eite
rn u
nd e
ine
Bank
mit
30.0
00 A
nges
tellt
en. B
eide
Firm
en h
aben
ein
ige
hund
ert M
itarb
eite
r in
alle
n Te
ilen
derW
elt u
nd d
ie F
irmen
leitu
ngen
sin
d üb
erze
ugt,
dass
inte
rnat
iona
le E
rfahr
ung
wic
htig
ist u
nd d
ie M
itarb
eite
r nac
h ih
rer R
ückk
ehr i
hr W
isse
n in
die
Firm
a ei
nbrin
gen.
Tats
ache
ist a
ber,
dass
die
Firm
en n
ach
der R
ückk
ehr d
erex
patri
ats
dies
e ka
um n
ach
ihre
mW
isse
n un
d ih
ren
Erfa
hrun
gen
frage
n.D
abei
hab
en d
ie M
itarb
eite
r im
Ausl
and
viel
übe
rau
slän
disc
he K
ultu
rund
Ges
chäf
tsge
woh
nhei
ten
gele
rnt.
Sie
kenn
en z
.B. a
uch
die
Kred
itris
iken
, uns
iche
re K
unde
n od
erne
ue G
esch
äfts
bere
iche
und
ver
steh
en v
iel b
esse
r, w
ieAu
slän
der i
hre
deut
sche
n G
esch
äfts
partn
erbe
urte
ilen.
Abe
r in
kein
er d
erbe
iden
Firm
enbe
müh
te m
an s
ich
aktiv
dar
um, d
asW
isse
n de
r Hei
mke
hrer
zu
nutz
en. K
aum
ein
er w
urde
nach
sei
nen
Erfa
hrun
gen
befra
gt u
nd n
ur e
inem
Vier
tel w
urde
ans
chlie
ßend
ein
e St
elle
ange
bote
n, d
ie e
inen
Bez
ug z
u de
m L
and
hatte
, in
dem
sie
gew
esen
war
en. N
ur w
enig
ebe
richt
eten
von
Pro
jekt
en, i
n de
nen
sie
ihrn
eues
Wis
sen
konk
ret u
mse
tzen
kon
nten
.
Es h
ilft a
bern
icht
s, In
tern
atio
nalis
ieru
ng a
nzus
trebe
n, u
nd d
ann
vorh
ande
ne M
öglic
hkei
ten
nich
t zu
nutz
en. D
iese
Ver
schw
endu
ng k
ann
sich
heu
te k
ein
Unt
erne
hmen
meh
r lei
sten
.G
ründ
efü
rdie
se V
ersä
umni
sse
der d
euts
chen
Firm
enle
itung
en z
ufin
den,
ist s
ehr s
chw
er.
Lieg
t es
an z
u st
arre
r Org
anis
atio
n, a
llgem
eine
rIgn
oran
z od
er s
ogar
an
der A
ngst
, das
san
dere
meh
r wis
sen
könn
ten
als
man
sel
ber?
Bei
den
bet
roffe
nen
Mita
rbei
tern
führ
te d
ieH
altu
ng d
er F
irmen
leitu
ngen
mei
sten
s zu
Fru
st u
nd R
esig
natio
n. V
on d
en F
irmen
leitu
ngen
gab
es k
eine
Ste
llung
nahm
e zu
den
Erg
ebni
ssen
der
Bef
ragu
ng.
Appendix 1.6
259
31
LEER
E SE
ITE
30
TEIL
ZW
EI
Absc
hnitt
Sec
hsAu
fgab
en 1
04 –
110
•D
er n
achf
olge
nde
Text
ent
hält
eini
ge F
ehle
r.•
Ist e
in S
atz
ohne
Feh
ler,
mac
hen
Sie
im A
ntw
ortb
ogen
ein
Häk
chen
( √ )
hint
erde
rN
umm
er.
•Is
t ein
Wor
t fal
sch,
sch
reib
en S
ie d
ie k
orre
kte
Form
hin
ter d
ie e
ntsp
rech
ende
Num
mer
inde
n An
twor
tbog
en.
Bei
spie
l:
Es k
onnt
e ke
ine
Übe
rein
kunf
t erz
ielt
wer
den.
0�
Bet�
tigen
Sie
bitt
e de
n Er
halt
derW
are.
00Bes
t�tig
en
Lieb
er H
err V
ollm
er,
104
Dan
ke fü
r Ihr
e An
frage
bez
üglic
h de
r von
uns
ang
ebot
ene
Dru
cker
. Da
105
Sie
die
Dru
cker
für I
hr G
roßr
aum
büro
ben
ötig
en u
nd s
omit
auch
der
106
Ger
äusc
hbel
astu
ng e
ine
Rol
le s
piel
t, m
öcht
e ic
h Si
e di
e An
scha
ffung
107
des
kom
pakt
en D
IN A
3-D
ruck
ers
BJ-2
30 e
mpf
ehle
n. E
r geh
ört z
u de
n
108
leis
este
n Ti
nten
stra
hldr
ucke
rnde
rWel
t und
sin
d de
shal
b fü
r ein
Gro
ßrau
mbü
ro b
este
ns g
eeig
net.
109
Bitte
lass
e Si
e un
s Ih
reEn
tsch
eidu
ng m
öglic
hst b
ald
wis
sen
und
sage
n
110
Sie
uns
auch
, wie
vie
len
Dru
cker
Sie
ben
ötig
en. B
ei e
iner
Bes
tellu
ng
von
min
dest
ens
5 D
ruck
ern
könn
en w
ir Ih
nen
eine
n R
abat
t anb
iete
n.
Mit
freun
dlic
hen
Grü
ßen
H. U
nter
berg
260
LEVELS Listening/Speaking Reading Writing
C2Level 5
CAN advise on/handlecomplex delicate orcontentious issues, suchas legal or financialmatters, to the extentthat he/she has thenecessary specialistknowledge.
CAN understand reportsand articles likely to beencountered duringhis/her work, includingcomplex ideasexpressed in complexlanguage.
CAN make full andaccurate notes andcontinue to participatein a meeting or seminar.
C1Level 4
CAN contributeeffectively to meetingsand seminars withinown area of work andargue for or against acase.
CAN understandcorrespondenceexpressed in non-standard language.
CAN handle a widerange of routine andnon-routine situations inwhich professionalservices are requestedfrom colleague orexternal contacts.
B2Level 3
CAN take and pass onmost messages that arelikely to requireattention during anormal working day.
CAN understand mostcorrespondence, reportsand factual productliterature he/she is likelyto come across.
CAN deal with allroutine requests forgoods or services.
B1Level 2
CAN offer advice toclients within own jobarea on simple matters.
CAN understand thegeneral meaning of non-routine letters andtheoretical articleswithin own work area.
CAN make reasonablyaccurate notes at ameeting or seminarwhere the subject matteris familiar andpredictable.
A2Level 1
CAN state simplerequirements withinown job area, such as ‘Iwant to order 25 of...’.
CAN understand mostshort reports or manualsof a predictable naturewithin his/her own areaof expertise, providedenough time is given.
CAN write a short,comprehensive note ofrequest to a colleague ora known contact inanother company.
A1ALTEbreakthroughlevel
CAN take and pass onsimple messages of aroutine kind, such as‘Friday meeting 10 am’.
CAN understand shortreports or productdescriptions on familiarmatters, if these areexpressed in simplelanguage and thecontents are predictable.
CAN write a simpleroutine request to acolleague, such as ‘CanI have 20X please?’.
APPENDIX 2.1
ALTE Work Typical Abilities
261
APPENDIX 3.1
BEC 1 Sample Paper
Appendic 3.1
262
Appendic 3.1
263
Appendic 3.1
264
Appendic 3.1
265
Appendic 3.1
266
Appendic 3.1
267
Appendic 3.1
268
Appendic 3.1
269
Appendic 3.1
270
Appendic 3.1
271
Appendic 3.1
272
Appendic 3.1
273
Appendic 3.1
274
Appendic 3.1
275
Appendic 3.1
276
277
APPENDIX 3.2
BEC 2 Sample Paper
Appendix 3.2
278
Appendix 3.2
279
Appendix 3.2
280
Appendix 3.2
281
Appendix 3.2
282
Appendix 3.2
283
Appendix 3.2
284
Appendix 3.2
285
Appendix 3.2
286
Appendix 3.2
287
Appendix 3.2
288
Appendix 3.2
289
Appendix 3.2
290
291
APPENDIX 3.3
BEC 3 Sample Paper
Appendix 3.3
292
Appendix 3.3
293
Appendix 3.3
294
Appendix 3.3
295
Appendix 3.3
296
Appendix 3.3
297
Appendix 3.3
298
Appendix 3.3
299
Appendix 3.3
300
Appendix 3.3
301
APPENDIX 4.1
BEC Preliminary Sample Paper
TIME 1 hour 30 minutes
INSTRUCTIONS TO CANDIDATES
Do not open this paper until you are told to do so.
Write your name, Centre number and candidate number in the spaces at the top of this page. Write these
details in pencil on your Answer Sheets if these are not already printed.
Write all your answers in pencil on your Answer Sheets – no extra time is allowed for this.
Read carefully the instructions for each part, and the instructions for completing your Answer Sheets.
Try to answer all the questions.
At the end of the examination hand in both this question paper and your Answer Sheets.
INFORMATION FOR CANDIDATES
There are forty-seven questions on this question paper:
Reading Questions 1 – 45
Writing Questions 46 – 47
UNIVERSITY OF CAMBRIDGE LOCAL EXAMINATIONS SYNDICATE
Examinations in English as a Foreign Language
BUSINESS ENGLISH CERTIFICATE 0351/1,2Preliminary
Test of Reading and Writing Test 023
Wednesday 5 JUNE 2002 Morning 1 hour 30 minutes
Additional materials:Answer Sheets
Candidate
Centre Number Number
Candidate Name
This question paper consists of 17 printed pages.
SP (AT/SLC) S25800/4
302
Appendix 4.1
303
2
AA
ll cu
stom
ers
are
guar
ante
ed n
ext-
day
deliv
ery.
BG
oods
are
del
iver
ed n
ext
day,
dep
endi
ng o
n av
aila
bilit
y.
CA
ny g
oods
out
of
stoc
k ar
e de
liver
ed w
ith y
our
next
ord
er.
3
Acc
ount
s w
ill
Ane
ed t
o ha
ve d
etai
ls o
f ex
pens
es b
efor
e ne
xt m
onth
.
Bse
ttle
all e
xpen
ses
at t
he e
nd o
f th
is m
onth
.
Cre
turn
you
r re
ceip
ts b
y th
e en
d of
the
mon
th.
4
AA
ny p
hoto
copi
es f
rom
Shr
oder
s w
ill c
ost
3p e
ach.
BE
ach
copy
cos
ts 3
p le
ss a
fter
the
first
300
0 co
pies
.
CF
or t
he f
irst
3000
cop
ies,
Shr
oder
s m
ake
no c
harg
e.
5
Sta
ff m
ust
Ain
form
the
ir m
anag
er if
the
y ar
e ab
sent
.
Bge
t pe
rmis
sion
to
take
tim
e of
f.
Cre
min
d th
eir
man
ager
of
whe
n th
ey a
re o
n le
ave.
Sta
ff a
re r
emin
ded
th
at a
ll ar
ran
gem
ents
fo
r le
ave
mu
st b
e ap
pro
ved
in a
dva
nce
by
line
man
ager
s.
SCH
RO
DER
SS
pec
ial
Off
er o
n p
ho
toco
pie
r re
nta
ls.
Fir
st 3
000
cop
ies
free
. A
fter
th
at e
ach
co
py
cost
s ju
st 3
p.
We
gu
aran
tee
del
iver
y to
yo
ur
do
or
wit
hin
24
ho
urs
, if
th
e g
oo
ds
are
in s
tock
.
Acc
ount
s
Sale
s D
epar
tmen
t
Exp
ense
s
Send
in a
ll re
ceip
ts f
or e
xpen
ses
by t
he la
st d
ay o
f th
is m
onth
.
RE
AD
ING
QU
ES
TIO
NS
1 –
45
PA
RT
ON
E
Qu
esti
on
s 1
– 5
•L
oo
k a
t q
ue
stio
ns 1
– 5
.
•In
ea
ch
qu
estio
n,
wh
ich
se
nte
nce
is c
orr
ect?
•F
or
ea
ch
qu
estio
n,
ma
rk o
ne
lett
er
(A,
Bor
C)
on
yo
ur
An
sw
er
Sh
ee
t.
1
Ph
on
e 0
84
5 1
23
4 f
or
Aa
lis
t o
f cu
rre
nt
job
va
ca
ncie
s.
Ba
re
fere
nce
nu
mb
er
for
on
e o
f th
ese
va
ca
ncie
s.
Cd
eta
ils a
bo
ut
job
s o
n t
he
lis
t.
FO
R M
OR
E I
NF
OR
MA
TIO
N A
BO
UT
AN
YO
F T
HE
JO
B V
AC
AN
CIE
S O
N T
HIS
LIS
T,P
HO
NE
0845
123
4, Q
UO
TIN
G T
HE
RE
FE
RE
NC
E N
UM
BE
R
Exa
mp
le:
0
The p
lane a
rriv
es a
t
Aquart
er
to s
even in t
he m
orn
ing.
Bquart
er
past
six
in t
he e
venin
g.
Cquart
er
to s
even in t
he e
venin
g.
The c
orr
ect
answ
er
is C
, so m
ark
your
Answ
er
Sheet
like t
his
:
Do
n’t
fo
rge
t –
flig
ht
BA
69
26
.45
pm
0A
BC
��
��
6A
publis
hin
g c
om
pany w
ants
to s
ele
ct
a c
om
pute
r cours
e f
or
new
sta
ff.
7A
mark
eting d
irecto
r re
quires f
ive h
undre
d leaflets
to p
rom
ote
a n
ew
pro
duct.
8T
he
sa
les m
an
ag
er
of
a s
oft
wa
re c
om
pa
ny h
as t
o a
rra
ng
e s
eve
ral tr
ips a
bro
ad
fo
r h
er
sta
ff.
9T
he m
anager
of
a s
mall
printing c
om
pany r
equires a
dvic
e o
n e
nte
ring f
ore
ign m
ark
ets
.
10A
hu
ma
n r
eso
urc
es m
an
ag
er
ha
s t
o a
pp
oin
t a
n e
xe
cu
tive
se
cre
tary
urg
en
tly f
or
the
Ma
na
gin
g
Directo
r.
PA
RT
TW
O
Qu
es
tio
ns
6 –
10
•Lo
ok a
t th
e bu
sine
ss a
dver
tisem
ents
bel
ow.
•F
or q
uest
ions
6 –
10
on
the
oppo
site
pag
e, c
hoos
e w
hich
com
pany
(A
– H
) ea
ch p
erso
n ne
eds
to c
onsu
lt.
•F
or e
ach
ques
tion,
mar
k on
ele
tter
(A– H
) on
you
r A
nsw
er S
heet
.
•D
o no
t us
e an
y le
tter
mor
e th
an o
nce.
Sellin
g a
bro
ad
?C
all
EX
PO
RT
EX
PE
RT
S!
HE
AD
LIN
E R
EC
RU
ITM
EN
T A
GE
NC
Y
We f
ind
th
e s
taff
yo
u n
eed
.A
E
WE
BM
AS
TE
RS
Bu
sin
ess W
eb
sit
e d
esig
ners
IT TRAINERS
Lead
ers
in
th
e t
rain
ing
fie
ldB
F
PR
ON
TO
SE
CR
ETA
RIA
LS
ER
VIC
ES
Off
ice t
asks q
uic
kly
co
mp
lete
d.
TH
OR
N’S
TR
AV
EL
AG
EN
CY
Exp
erts
in f
orei
gn b
usin
ess
trav
elC
G
- -
- -
RA
PID
PR
INT -
- -
-Fo
r al
l yo
ur p
rint
ing
C LL
AR
KK ’’
S
Tech
nic
al
Pu
blish
ers
DH
Appendix 4.1
304
Appendix 4.1
305
11T
he
re w
as a
co
ntin
ue
d r
ed
uctio
n in
exce
ss s
pe
nd
ing
on
ma
teria
ls a
nd
a d
eclin
e in
th
e
ove
rhe
ad
s f
igu
res.
12E
xce
ss s
pe
nd
ing
on
ma
teria
ls a
nd
la
bo
ur
we
re b
oth
a little
be
low
re
co
rd le
ve
ls,
bu
t sp
en
din
g o
n
ove
rhe
ad
s w
as a
t its lo
we
st.
13T
he
am
ou
nt
for
lab
ou
r d
rop
pe
d,
wh
ile t
he
fig
ure
s f
or
ma
teria
ls a
nd
ove
rhe
ad
s w
ere
un
ch
an
ge
d
on
th
e p
revio
us m
on
th.
14E
xce
ss s
pe
nd
ing
on
ma
teria
ls d
rop
pe
d s
ligh
tly in
th
is m
on
th w
hile
ove
rhe
ad
s s
ho
we
d a
slig
ht
incre
ase,
to a
fig
ure
whic
h w
as e
qualle
d in t
he f
ollo
win
g m
onth
.
15In
this
month
the a
mount
for
mate
rials
was u
nchanged,
while
for
both
labour
and o
verh
eads
the
re w
as a
n im
pro
ve
me
nt
in t
he
fig
ure
s.
PA
RT
TH
RE
E
Qu
esti
on
s 11
– 1
5
•Lo
ok a
t th
e ch
art
belo
w.
It sh
ows
the
amou
nts
by w
hich
a c
ompa
ny’s
exp
endi
ture
on
mat
eria
ls,
labo
ur a
nd o
verh
eads
wer
e ab
ove
budg
et.
•W
hich
mon
th d
oes
each
sen
tenc
e (1
1 –
15)
on t
he o
ppos
ite p
age
desc
ribe?
•F
or e
ach
sent
ence
, m
ark
one
lette
r (A
– H
)on
you
r A
nsw
er S
heet
.
•D
o no
t us
e an
y le
tter
mor
e th
an o
nce.
AB
CD
EF
GH
Month
0
1000
2000
3000
4000
5000
6000
7000
8000£
Exce
ss S
pendin
gM
ate
rials
Labour
Overh
eads
16G
len
mo
re m
ake
s a
ll th
e je
we
llery
th
at
it s
ells
.
AR
igh
tB
Wro
ng
CD
oe
sn
’t sa
y
17G
len
mo
re is t
he
be
st-
kn
ow
n b
ran
d o
f tr
ad
itio
na
l S
co
ttis
h je
we
llery
de
sig
n.
AR
igh
tB
Wro
ng
CD
oe
sn
’t sa
y
18N
evis
ha
s a
la
rge
sto
ck o
f u
nso
ld c
op
ies o
f G
len
mo
re je
we
llery
.
AR
igh
tB
Wro
ng
CD
oe
sn
’t sa
y
19G
ram
pia
n G
ifts
has t
o p
ay a
ll its p
rofits
for
the last
four
years
to H
ighla
nd D
esig
n.
AR
igh
tB
Wro
ng
CD
oe
sn
’t sa
y
20T
he
nu
mb
er
of
me
mb
ers
wh
o c
on
tact
PA
C h
as in
cre
ase
d.
AR
igh
tB
Wro
ng
CD
oe
sn
’t sa
y
21D
esig
ne
rs s
ho
uld
re
po
rt a
ny c
op
ies o
f th
eir w
ork
wh
ich
th
ey s
ee
at
exh
ibitio
ns.
AR
igh
tB
Wro
ng
CD
oe
sn
’t sa
y
22P
AC
ha
s w
on
th
e m
ajo
rity
of
ca
se
s o
f le
ga
l a
ctio
n it
ha
s t
ake
n f
or
its m
em
be
rs.
AR
igh
tB
Wro
ng
CD
oe
sn
’t sa
y
PA
RT
FO
UR
Qu
esti
on
s 16
– 2
2
•R
ead
the
new
spap
er a
rtic
le b
elow
abo
ut a
n or
gani
satio
n w
hich
pro
tect
s its
mem
bers
’orig
inal
desi
gns.
•A
re s
ente
nces
16
– 22
on
the
oppo
site
pag
e ‘R
ight
’or
‘Wro
ng’?
If
ther
e is
not
eno
ugh
info
rmat
ion
to a
nsw
er ‘R
ight
’or
‘Wro
ng’,
choo
se ‘D
oesn
’t sa
y’.
•F
or e
ach
sent
ence
16
– 22
, m
ark
one
lette
r (A
,B
orC
) on
you
r A
nsw
er S
heet
.
Las
t ye
ar,
Sco
tlan
d-ba
sed
Gle
nmor
e Je
wel
lery
be
cam
e th
efo
rtie
th m
embe
r of
the
orga
nisa
tion
PAC
(P
rote
ctio
n A
gain
st C
opyi
ng)
to t
ake
succ
essf
ul l
egal
act
ion
topr
otec
t it
s pr
oduc
ts.
Gle
nmor
e,w
hich
se
lls
hand
-mad
e je
wel
lery
desi
gned
an
d pr
oduc
ed
by
loca
lar
tist
s,
was
an
gry
to
disc
over
a
com
peti
tor,
Nev
is, s
elli
ng c
opie
s of
thes
e de
sign
s.
Nev
is
has
now
agre
ed t
o st
op s
elli
ng t
he p
rodu
cts,
as w
ell
as t
o pa
y G
lenm
ore’
s le
gal
cost
s.A
noth
er
Sco
ttis
h co
mpa
ny,
Hig
hlan
d D
esig
n,
also
as
ked
for
PAC
’s h
elp
last
yea
r; a
s a
resu
lt,
Gra
mpi
an G
ifts
has
agr
eed
to s
top
sell
ing
glas
s de
sign
s si
mil
ar
topr
oduc
ts
mad
e by
H
ighl
and
Des
ign.
G
ram
pian
w
ill
also
pa
y
Hig
hlan
d D
esig
n 50
% o
f th
e pr
ofit
it h
as m
ade
on t
hese
ran
ges
over
the
last
fou
r ye
ars.
‘Rec
entl
y th
ere’
s be
en a
sha
rpri
se
in
the
num
ber
of
repo
rts
rece
ived
fr
om
mem
bers
,’sa
ysJa
mes
E
llis
, PA
C’s
C
hief
Exe
cuti
ve.
‘Thi
s de
mon
stra
tes
that
desi
gner
s cl
earl
y fe
el w
e ca
n he
lpth
em.
If
they
di
scov
er
copi
es
ofth
eir
desi
gns
at
exhi
biti
ons,
fo
rex
ampl
e,
they
ar
e ad
vise
d to
cont
act
us i
mm
edia
tely
.’PA
C,
wit
h ov
er
700
mem
bers
wor
ld-w
ide,
has
an
alm
ost
100%
succ
ess
rate
in le
gal a
ctio
n ta
ken
tode
fend
th
e in
tere
sts
of
the
indi
vidu
als
and
com
pani
es
itre
pres
ents
.
PAC
Win
s A
gain
Appendix 4.1
306
Appendix 4.1
307
23In
Pe
ter
Ma
urice’s
op
inio
n,
wh
at
pre
ve
nts
ma
ny c
usto
me
rs f
rom
ma
kin
g p
urc
ha
se
s?
AT
he
sh
op
ha
sn’t
go
t w
ha
t th
ey w
an
t.
BN
ob
od
y is a
va
ilab
le t
o s
erv
e t
he
m.
CT
he
re is t
oo
mu
ch
to
ch
oo
se
fro
m.
24M
au
rice’s
sta
ff s
ay t
ha
t h
e
Aso
me
tim
es a
nn
oys t
he
m.
Bis
try
ing
to
do
to
o m
uch
to
o q
uic
kly
.
Cn
ee
ds t
o b
e m
ore
se
lf-a
wa
re.
25W
hat
does M
aurice s
ay h
e learn
t fr
om
his
experience in H
ong K
ong?
Ah
ow
to
ru
n a
su
cce
ssfu
l im
po
rt-e
xp
ort
bu
sin
ess
Bth
e im
port
ance o
f bein
g c
onfident
and h
avin
g a
positiv
e a
ttitude
Cth
at
he
ne
ed
ed
to
go
ba
ck t
o b
usin
ess s
ch
oo
l
26W
ha
t is
Ma
urice’s
exh
ibitio
ns c
om
pa
ny,
Big
Eve
nts
, p
lan
nin
g t
o d
o in
20
04
?
Ata
ke
co
ntr
ol o
f th
e C
ap
ita
l B
oa
t S
ho
w
Bm
ove
th
e C
ap
ita
l B
oa
t S
ho
w t
o a
la
rge
r ve
nu
e
Ch
old
an
eve
nt
to c
om
pe
te w
ith
th
e C
ap
ita
l B
oa
t S
ho
w
27W
hic
h o
f th
ese
de
pa
rtm
en
ts h
as a
dire
cto
r w
ho
re
po
rts t
o P
ete
r M
au
rice
?
AF
ina
nce
BM
ark
etin
g
CH
um
an
Re
so
urc
es
28M
au
rice’s
lo
ng
-te
rm a
mb
itio
n f
or
the
ce
ntr
e is t
o
Are
tra
in a
ll th
e s
taff a
t th
e c
en
tre
.
Bm
od
ern
ise
th
e c
en
tre
.
Cb
rin
g e
nte
rta
inm
en
t to
th
e c
en
tre
.
PA
RT
FIV
E
Qu
esti
on
s 23
– 2
8
•R
ead
the
artic
le b
elow
abo
ut a
bus
ines
sman
’s p
lans
for
dev
elop
ing
a sh
oppi
ng c
entr
e.
•F
or e
ach
ques
tion
23 –
28,
on
the
oppo
site
pag
e, c
hoos
e th
e co
rrec
t an
swer
.
•M
ark
one
lette
r (A
,B
orC
) on
you
r A
nsw
er S
heet
.
Sh
akin
g U
p t
he
Bu
sin
ess
Sin
ce b
eco
min
g C
hie
f E
xecu
tive
of
the S
tar
Cit
y s
ho
pp
ing
cen
tre
an
d
ex
hib
itio
n
ha
lls,
Pe
ter
Mau
rice f
eels
he h
as d
on
e a
lo
t.
No
w,
tho
ug
h,
he
w
an
ts
to
ch
an
ge t
he w
ho
le f
eelin
g o
f th
e
bu
sin
ess.
‘Vis
ito
rs
sh
ou
ld
feel
we are
lo
okin
g aft
er
them
,’ h
e
says.
‘Very
o
ften
th
e p
ub
lic g
o
into
a sh
op
an
d fi
nd
so
m
uch
there
th
at
they c
an
’t d
ecid
e w
hat
to
bu
y,
so
th
ey
d
on
’t
bu
y
an
yth
ing
. K
eep
it
sim
ple
, th
at’s
the k
ey t
o r
eta
ilin
g.’
At
Sta
r C
ity,
sta
ff
are
en
co
ura
ge
d
to
tell
m
an
ag
ers
,
inclu
din
g M
au
rice h
imself
, w
hat
they t
hin
k o
f th
em
. ‘T
he t
hin
gs
they say ab
ou
t m
e are
w
hat
I
exp
ect,
b
ecau
se I’m
fa
irly
self
-
aw
are
– I k
no
w w
hat
I’m
lik
e a
nd
that
I can
m
ake p
eo
ple
a litt
le
an
gry
. B
ut
I’m
v
ery
m
uch
in
favo
ur
of
ch
an
ge,
an
d e
very
on
e
kn
ow
s
that
a
lot
need
s
to
be
do
ne.’
He
le
arn
t h
is
ma
na
ge
me
nt
tech
niq
ues t
he h
ard
way.
‘At
23
I w
en
t in
to
bu
sin
ess
an
d
lost
mo
ne
y.
I h
ad
to
le
arn
fa
st.
Th
en
,
at
32,
I w
on
an
exp
ort
co
ntr
act
to
Ho
ng
K
on
g.
I a
dm
ire
th
e
str
en
gth
o
f ch
ara
cte
r an
d
the
am
bit
ion
o
f th
e
peo
ple
th
ere
,
an
d
bro
ug
ht
ba
ck
two
v
ery
sig
nif
ica
nt
wo
rds:
“N
o
pro
ble
m”.
Th
en
I t
oo
k a
co
urs
e
at
Harv
ard
B
usin
ess S
ch
oo
l. It
was v
ery
hard
wo
rk,
bu
t w
ort
h
it.’
As
well
as
run
nin
g
Sta
r C
ity,
Pe
ter
Ma
uri
ce
co
ntr
ols
B
ig
Ev
en
ts,
wh
ich
o
rga
nis
es
exh
ibit
ion
s.
At
the m
om
en
t, B
ig
Even
ts i
s w
ork
ing
on
pla
ns f
or
a
bo
at
sh
ow
to
ri
val
the
Cap
ital
Bo
at
Sh
ow
, w
hic
h
in
2004
is
mo
vin
g f
rom
its
tra
dit
ion
al
sit
e
at
Sta
r C
ity t
o a
new
ven
ue.
Mau
rice h
as c
reate
d a
n u
nu
su
al
co
mp
an
y
str
uctu
re.
‘Th
e
fin
an
cia
l d
ire
cto
r a
nd
co
mm
erc
ial
dir
ecto
r a
re
resp
on
sib
le to
m
e d
irectl
y,
bu
t
in m
y f
irst
week h
ere
, th
e h
ead
of
Mark
eti
ng
resig
ned
. T
hen
th
e
sa
me
h
ap
pe
ne
d
wit
h
Hu
ma
n
Reso
urc
es.
I said
to
bo
th t
eam
s:
“D
o y
ou
wan
t to
self
-man
ag
e?”
Th
at’s w
hat
they d
ecid
ed
to
do
–
it c
an
wo
rk i
f yo
u h
ave p
eo
ple
wh
o w
ork
well t
og
eth
er
an
d c
an
rep
ort
to
yo
u
as
a
team
,’
he
exp
lain
s.
Fo
r th
e
imm
ed
iate
fu
ture
,
Mau
rice w
ill
co
nti
nu
e w
ith
th
e
essen
tial
wo
rk o
f u
pd
ati
ng
th
e
cen
tre.
Aft
er
that,
h
e
pla
ns
to
loo
k a
t w
ays o
f exp
an
din
g S
tar
Cit
y
be
yo
nd
th
e
pre
se
nt
co
nfe
ren
ces a
nd
exh
ibit
ion
s,
to
inclu
de
m
ajo
r sh
ow
s
an
d
co
ncert
s.
‘I w
an
t a l
ively
cen
tre
full o
f excit
ing
even
ts, w
here
my
we
ll-t
rain
ed
sta
ff
are
am
bassad
ors
fo
r th
e c
om
pan
y.’
29A
on
Bin
Cat
30A
will
Bsh
all
Cw
ou
ld
31A
run
Bru
ns
Cru
nn
ing
32A
Th
isB
There
CIt
33A
wh
ich
Bth
at
Ch
ow
34A
ma
ny
Bso
me
Ca
ny
35A
attra
ct
Battra
cting
Cattra
cts
36A
wh
ose
Bw
ho
Cw
ha
t
37A
toB
with
Ca
bo
ut
38A
ou
rsB
our
Cu
s
39A
are
Bh
ave
Cw
ere
40A
pa
ss
Bfa
ilC
mis
s
PA
RT
SIX
Qu
esti
on
s 29
– 4
0
•R
ead
the
adve
rtis
emen
t be
low
abo
ut a
n ex
hibi
tion.
•C
hoos
e th
e co
rrec
t w
ord
to f
ill e
ach
gap,
fro
m A
,B
orC
on t
he o
ppos
ite p
age.
•F
or e
ach
ques
tion
29 –
40,
mar
k on
ele
tter
(A,
Bor
C)
on y
our
Ans
wer
She
et.
Bus
ines
s S
how
of
the
Year
14–1
8 S
epte
mbe
r 20
02Lo
ndon
Bu
sin
ess
2002
is a
key
eve
nt
aim
ed a
t people
inte
ndin
g to
set
up (
29)
......
busi
ness
.
This
year
’s e
xhib
itio
n (
30)
......
pro
vide v
isitors
with v
aluab
le info
rmat
ion o
n p
roduct
s
and s
erv
ices
ess
ential
for
(31)
....
..a
smal
l busi
ness
.(32
) ...
...is
als
o fre
e a
dvi
ce o
n
(33)
....
...to
obta
in fin
ance
,gett
ing
the b
est
out
of In
form
atio
n T
ech
nolo
gy a
nd
deal
ing
with t
he (
34)
......
pro
ble
ms
of heal
th a
nd s
afety
in t
he w
ork
pla
ce.E
ntr
ance
to t
he e
xhib
itio
n is
free a
nd B
usi
nes
s 20
02 h
opes
to (
35)
......
at leas
t 5,0
00 v
isitors
a day
.Com
pan
ies
(36)
....
..to
ok p
art
last
year
sai
d it
was
well
wort
h t
he c
ost
.
If y
ou w
ould
lik
e furt
her
info
rmat
ion (
37)
......
hav
ing
a st
and a
t th
e B
usi
nes
s 20
02exhib
itio
n,c
onta
ct (
38)
......
.on 0
20 5
493 7
721 for
an e
xhib
itors
’ pac
k.
We (
39)
......
.arr
ange
d s
peci
al r
educe
d t
rave
l an
d h
ote
l ra
tes
for
stan
dhold
ers
.So
don’t (
40)
......
the o
pport
unity,
book a
sta
nd n
ow
.
Appendix 4.1
308
Appendix 4.1
309
Tra
velg
o L
tdR
efu
nd
No
: 0055 7
8A
Bookin
g r
efe
rence:
(41)
.......................................
Tic
ket
deta
ils:
Date
of
depart
ure
:22/0
6/0
2
Flig
ht
num
ber:
UA
453 (
UA
TA
IR)
Depart
ure
fro
m:
Heath
row
Destination:
(42)
.......................................
Tic
ket
cla
ss:
(43)
.......................................
Refu
nd d
ue:
£525.0
0
Meth
od o
f R
efu
nd:
(44)
.......................................
Reason f
or
cancella
tion:
clie
nt’s
(45)
.......................................
PA
RT
SE
VE
N
Qu
esti
on
s 41
– 4
5
•R
ea
d t
he
pa
rt o
f th
e le
tte
r a
nd
th
e m
em
o b
elo
w.
•C
om
ple
te t
he
fo
rm o
n t
he
op
po
site
pa
ge
.
•W
rite
a w
ord
or
ph
rase
(in
CA
PIT
AL
LE
TT
ER
S)
or
a n
um
be
ro
n lin
es 4
1 –
45 o
n y
ou
r A
nsw
er
Sh
ee
t. Tra
velg
o L
td
Mem
ora
nd
um
To:
Julia
Fro
m:
Lu
ke
Dat
e:6
Jun
e 20
02
Su
bje
ct:
Ref
un
d
Could
you p
leas
e fill
in a
refu
nd form
for
this
clie
nt?
Note
that
the b
ookin
g re
fere
nce
num
ber
is
actu
ally
01/3
434/A
.It
was
a c
om
pan
y bookin
g fo
r a
Busi
ness
Cla
ss (
not
Fir
st C
lass
)U
ATA
IRflig
ht
from
Heat
hro
w.T
hey
pai
d b
y cr
edit c
ard b
ut
it’s
eas
ier
if w
e s
end t
hem
a c
heque for
£525.0
0.
Wit
h r
egar
d t
o o
ur
con
vers
atio
n o
n t
he
ph
on
e ye
ster
day
, I
can
co
nfi
rm t
hat
, d
ue
to s
erio
us
illn
ess,
Joh
n B
roo
kes
has
to c
ance
l his
flig
ht t
o M
adri
d o
n S
atu
rday
22
Jun
e an
d I
ther
efo
re r
equ
est a
ref
un
d.
I en
clo
se t
he
tick
et:F
irst
Cla
ss, f
ligh
t n
um
ber
UA
453
, bo
oki
ng
refe
ren
ce 3
434/
4.
PA
RT
TW
O
Qu
esti
on
47
•R
ead
this
par
t of
a f
ax f
rom
Mr
Dav
id C
rave
n, o
ne o
f yo
ur c
ompa
ny’s
cus
tom
ers.
•W
rite
a le
tter
to M
r C
rave
n:
•ap
olog
isin
g fo
r w
hat
has
happ
ened
•ex
plai
ning
the
del
ay
•sa
ying
whe
n yo
u w
ill s
end
the
mot
ors
•of
ferin
g a
redu
ctio
n on
the
bill
.
•W
rite
60
– 80
wo
rds.
•W
rite
on
yo
ur
An
swer
Sh
eet.
Do
no
t in
clu
de
any
po
stal
ad
dre
sses
.
We
ord
ered
eig
ht
LP4
elec
tric
mo
tors
fro
m y
ou
las
t m
on
th.
It i
sn
ow
3 J
un
e an
d w
e st
ill h
ave
rece
ived
no
thin
g, e
ven
th
ou
gh
yo
u in
form
ed u
s th
at w
e w
ou
ld h
ave
the
mo
tors
by
30 M
ay. W
en
eed
to k
no
w w
hen
we
will
rec
eive
th
ese
mo
tors
as
ou
r o
wn
wo
rk w
ill b
e h
eld
up
if
they
do
no
t ar
rive
wit
hin
th
e n
ext
few
day
s.
WR
ITIN
G
QU
ES
TIO
NS
46
– 47
PA
RT
ON
E
Qu
esti
on
46
•Y
ou h
ave
notic
ed t
hat
staf
f us
ing
com
pany
car
s ar
e fo
rget
ting
to f
ill t
hem
with
pet
rol o
n th
eir
retu
rn.
•Y
ou a
lso
wan
t to
rem
ind
staf
f of
pro
cedu
res
for
park
ing
com
pany
car
s an
d de
posi
ting
keys
.
•W
rite
a m
emo
to a
ll st
aff:
•ex
plai
ning
wha
t to
do
abou
t pe
trol
afte
r us
ing
a co
mpa
ny c
ar
•sa
ying
whe
re t
o le
ave
the
car
•te
lling
sta
ff w
hat
to d
o w
ith t
he k
eys.
•W
rite
30
– 40
wo
rds.
•W
rite
on
yo
ur
An
swer
Sh
eet.
Mem
ora
nd
um
To:
All
staf
f
Fro
m:
Off
ice
Man
ager
Dat
e:6
Jun
e 20
02
Su
bje
ct:
Co
mp
any
Car
s
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
.....
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
.....
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
.....
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
.....
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
.....
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
.....
Appendix 4.1
310
Appendix 4.1
311
TIME Approx. 40 minutes (including 10 minutes’ transfer time)
INSTRUCTIONS TO CANDIDATES
Do not open this paper until you are told to do so.
Write your name, Centre number and candidate number in the spaces at the top of this page. Write these
details in pencil on your Answer Sheet if these are not already printed.
Listen to the instructions for each part carefully.
Try to answer all the questions.
Write your answers on this question paper.
At the end of the test you will have 10 minutes to copy your answers onto your Answer Sheet.
Read the instructions for completing your Answer Sheet carefully.
Write all your answers in pencil.
At the end of the examination hand in both this question paper and your Answer Sheet.
INFORMATION FOR CANDIDATES
Instructions are given on the tape.
You will hear everything twice.
There are thirty questions on this paper.
UNIVERSITY OF CAMBRIDGE LOCAL EXAMINATIONS SYNDICATE
Examinations in English as a Foreign Language
BUSINESS ENGLISH CERTIFICATE 0351/3Preliminary
Test of Listening Test 023
Wednesday 5 JUNE 2002 Morning Approx. 40 minutes(including 10 minutes’
transfer time)
Additional materials:Answer Sheet
Candidate
Centre Number Number
Candidate Name
This question paper consists of 8 printed pages.
SP (SC/KN) S25767/2
3W
ha
t d
o b
oth
sp
ea
ke
rs t
hin
k t
he
co
mp
an
y n
ee
ds t
o r
evie
w?
Aprices
Bqualit
y
Cd
eliv
ery
tim
es
4W
hic
h g
rap
h is t
he
sp
ea
ke
r re
ferr
ing
to
?
5W
ha
t is
th
e s
ub
ject
of
the
se
min
ar?
AC
om
mu
nic
atio
n s
kill
s
BT
ime
ma
na
ge
me
nt
CR
ecru
itm
en
t str
ate
gie
s
6W
hic
h n
um
be
r sh
ou
ld y
ou
pre
ss t
o d
iscu
ss p
aym
en
t?
AB
C
12
3
050100
150
200
250
AM
J
A
050100
150
200
250
AM
J
B
050100
150
200
250
AM
J
C
Com
pany
sha
re p
rice
s in
pen
ceC
ompa
ny s
hare
pri
ces
in p
ence
Com
pany
sha
re p
rice
s in
pen
ce
PA
RT
ON
E
Qu
esti
on
s 1
– 8
•F
or
qu
estio
ns 1
– 8
yo
u w
ill h
ea
r e
igh
t sh
ort
re
co
rdin
gs.
•F
or
ea
ch
qu
estio
n,
ma
rk o
ne
letter
(A,
Bor
C)
for
the
co
rre
ct
an
sw
er.
Exa
mp
le:
Wh
en
we
re t
he
ma
ch
ine
pa
rts s
en
t?
Mon
day
31Tu
esda
y 1
Thur
sday
3
AB
C
Th
e a
nsw
er
is A
.
•Y
ou
will
he
ar
the
eig
ht
reco
rdin
gs t
wic
e.
1W
he
n s
ho
uld
th
e w
om
an
ph
on
e M
r H
ash
imo
to?
AB
C
2W
hic
h c
ha
rt s
ho
ws g
rad
ua
te r
ecru
itm
en
t fo
r th
is y
ea
r?
RS
E A
R
S E
B
R
SE
C
R=
Researc
hS
= S
ale
sE
= E
ngin
eering
3:00
2:00
11:0
0
Appendix 4.1
312
Appendix 4.1
313
PA
RT
TW
O
Qu
esti
on
s 9
– 15
•Lo
ok a
t th
e no
tes
belo
w.
•S
ome
info
rmat
ion
is m
issi
ng.
•Y
ou w
ill h
ear
am
an le
avin
g an
ans
wer
phon
e m
essa
ge a
bout
ele
ctric
al g
oods
.
•F
or e
ach
ques
tion
9 –
15,
fill i
n th
e m
issi
ng in
form
atio
n in
the
num
bere
d sp
ace
usin
g a
wo
rd,
nu
mb
ers
orle
tter
s.
•Y
ou w
ill h
ear
the
mes
sage
tw
ice.
Ph
on
e M
essa
ge
CO
MPA
NY
NA
ME
:(9
)...
......
......
......
......
......
......
......
......
......
......
......
......
......
......
.Dep
artm
ent
Stor
e
CO
NT
AC
T N
AM
E:
(10)
Rog
er .
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
PH
ON
E N
UM
BE
R:
(11)
0187
3 ...
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
....
UR
GE
NT
– S
EN
D:
(12)
...
......
......
......
......
......
......
......
......
......
......
......
......
frid
ge/fr
eeze
r br
ochu
res
AL
SO
IN
CL
UD
E:
(13)
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
pric
e lis
ts
CH
EC
K A
VA
ILA
BIL
ITY
OF
DIS
HW
AS
HE
R M
OD
EL
NO
:(1
4)
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
CU
ST
OM
ER
NE
ED
S D
ISH
WA
SH
ER
:(1
5)
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
......
.mm
wid
e
7W
hat
does
the
wom
an w
ant
info
rmat
ion
abou
t?
8W
hat
does
the
man
ord
er?
AB
C
Tu
rner
'sL
td
A
AIR
LIN
ET
ICK
ET
SC
AR
HIR
E
BC
HO
TE
L
PA
RT
FO
UR
Qu
esti
on
s 23
– 3
0
•Y
ou w
ill h
ear
an in
terv
iew
with
Mic
hael
Wrig
ht,
a m
anag
ing
dire
ctor
, ta
lkin
g ab
out
his
com
pany
,
ALC
.
•F
or e
ach
ques
tion
23 –
30,
mar
k o
ne
lette
r (A
,B
orC
) fo
r th
e co
rrec
t an
swer
.
•Y
ou w
ill h
ear
the
inte
rvie
w t
wic
e.
23M
icha
el’s
com
pany
firs
t st
arte
d in
bus
ines
s
Apr
oduc
ing
part
s fo
r lo
rrie
s.
Bdo
ing
lorr
y re
pairs
.
Cbu
ildin
g sp
ecia
l lor
ries
for
clie
nts.
24M
icha
el jo
ined
ALC
bec
ause
he
wan
ted
to
Aco
ntin
ue a
s an
eng
inee
r.
Bru
n th
e pr
oduc
tion
depa
rtm
ent.
Cbe
a d
irect
or.
25A
t th
e tim
e M
icha
el b
ecam
e a
dire
ctor
, ALC
was
Ado
ing
good
bus
ines
s.
Bbu
ying
new
equ
ipm
ent.
Cne
gotia
ting
a co
ntra
ct.
26M
icha
el s
ugge
sted
tha
t th
e co
mpa
ny s
houl
d
Ach
ange
its
prod
uctio
n m
etho
ds.
Bai
m t
o ta
ke o
ver
its b
igge
st r
ival
s.
Cin
crea
se it
s pr
oduc
tion
rate
.
27O
ne o
f th
e re
ason
s A
LC s
tart
ed p
rodu
cing
am
bula
nces
was
bec
ause
Aa
cust
omer
had
ask
ed it
to.
Bth
e co
mpa
ny a
lread
y ha
d a
good
des
ign.
Cth
e m
arke
t fo
r am
bula
nces
was
fai
rly s
tabl
e.
PA
RT
TH
RE
E
Qu
esti
on
s 16
– 2
2
•Lo
ok a
t th
e no
tes
from
a s
taff
mee
ting
in a
dep
artm
ent
stor
e.
•S
ome
info
rmat
ion
is m
issi
ng.
•Y
ou w
ill h
ear
a ta
lk b
y th
e st
ore
man
ager
.
•F
or e
ach
ques
tion
16 –
22,
fill
in t
he m
issi
ng in
form
atio
n in
the
num
bere
d sp
ace
usin
g o
ne
or
two
wor
ds.
•Y
ou w
ill h
ear
the
talk
tw
ice.
Staf
f mee
ting
held
on
8 Ju
ne
Nam
e of n
ew A
ssistan
t M
anag
er(1
6)Am
anda
........................................................................
Sportswe
ar D
epar
tmen
tPr
omot
ion
oppo
rtun
ities op
en t
o(1
7)..............................................................................
staf
f
Stoc
kroom New
(18)
........................................
will
impr
ove
stoc
k ha
ndlin
g
Desig
n Te
amSt
aff
trai
ning
– sub
ject:
(19)
........................................................................................
Visit
From
:(2
0)........................................................................................
Date
:(2
1)13
th......................................................................
2002
Subjec
t of t
alk:
(22)
..............................................................................
polic
y
Appendix 4.1
314
Appendix 4.1
315
28W
hen
it st
arte
d pr
oduc
ing
ambu
lanc
es, A
LC’s
mai
n pr
oble
m w
as
Aco
mpe
ting
with
ano
ther
com
pany
.
Bha
ving
to
cut
its p
rices
.
Clo
sing
a c
ontr
act.
29M
icha
el f
eels
ALC
is u
nusu
al b
ecau
se t
he d
irect
ors
Aar
e re
spon
sibl
e fo
r m
arke
ting
the
prod
ucts
.
Bha
ve li
ttle
cont
act
with
man
ager
s.
Cw
ork
on t
he p
rodu
ctio
n lin
e.
30N
ow A
LC is
now
aim
ing
to
Aim
prov
e its
cas
h flo
w.
Bre
crui
t m
ore
staf
f.
Cdo
mor
e tr
aini
ng.
Yo
u n
ow
hav
e 10
min
ute
s to
tra
nsf
er y
ou
r an
swer
s to
yo
ur
An
swer
Sh
eet.
CO
NT
ENT
S
Intr
oduc
tion,
BEC
Can
dida
ture
, Aw
ards
, Not
ifica
tion
of R
esul
ts, U
sefu
l Doc
umen
tatio
n . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. .
3
Exam
inat
ion
Rep
ort,
Rea
ding
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
5
Mar
king
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. .
5
Com
men
ts o
n ca
ndid
ate
perf
orm
ance
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. .
5
Rec
omm
enda
tions
for
cand
idat
e pr
epar
atio
n . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
6
Prep
arin
g fo
r B
EC P
relim
inar
y R
eadi
ng (a
sum
mar
y) .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. 7
Wri
ting
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. .
8
Mar
king
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. .
8
Com
men
ts o
n ca
ndid
ate
perf
orm
ance
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. .
8
Rec
omm
enda
tions
for
cand
idat
e pr
epar
atio
n . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
8
Prep
arin
g fo
r B
EC P
relim
inar
y W
ritin
g (a
sum
mar
y) .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. .
11
List
enin
g . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
12
Mar
king
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. .
12
Com
men
ts o
n ca
ndid
ate
perf
orm
ance
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. .
12
Rec
omm
enda
tions
for
cand
idat
e pr
epar
atio
n . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
13
Prep
arin
g fo
r B
EC P
relim
inar
y Li
sten
ing
(a s
umm
ary)
. .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. 14
BEC
Pre
limin
ary
June
200
2
Cop
yrig
ht.
The
cont
ents
of
this
pub
licat
ion
are
cove
red
byin
tern
atio
nal c
opyr
ight
law
. All
righ
ts r
eser
ved;
no
part
of t
his
docu
men
t may
be
repr
oduc
ed, s
tore
d in
a r
etri
eval
sys
tem
, or
tran
smitt
ed
in
any
form
or
by
an
y m
eans
, el
ectr
onic
,m
echa
nica
l, ph
otoc
opyi
ng,
reco
rdin
g or
oth
erw
ise,
with
out
the
prio
r w
ritte
n co
nsen
t of t
he U
nive
rsity
of C
ambr
idge
ESO
LEx
amin
atio
ns.
Indi
vidu
als
and
inst
itut
ions
w
ho
do
not
obse
rve
thes
e co
pyri
ght
rest
rict
ions
w
ill
be
liabl
e to
pros
ecut
ion.
In u
sing
the
key
s an
d m
arki
ng n
otes
in
prac
tice
test
ing,
it
shou
ld b
e re
mem
bere
d th
at s
core
s in
ind
ivid
ual
pape
rs o
rqu
estio
ns
are
subj
ect
to
wei
ghtin
g du
ring
th
e ac
tual
exam
inat
ion
proc
essi
ng.
Appendix 4.1
316
Appendix 4.1
317
Appendix 4.1
318
PA
RT
2:
Min
i-p
resen
tati
on
s f
or
2can
did
ate
s(a
bo
ut
5m
inu
tes)
[Inte
rlocuto
rhands e
ach
candid
ate
a d
iffe
rent
topic
card
,and
som
epaper
and
apencil
for
note
s.]
[Allo
w1
min
ute
'spre
para
tion tim
e.
Both
candid
ate
spre
pare
their
talk
s a
tth
esam
etim
e,
separa
tely
.]
[Allo
w45
seconds
to1
min
ute
.O
nly
use b
ack-u
pquestions if
the
candid
ate
is u
nable
to
speak
without
pro
mpting.]
[Allo
w10
seconds.
Do n
ot
use
back-u
p q
uestions.]
[Allo
w45
seconds
to1
min
ute
.O
nly
use b
ack-u
pquestions if
the
candid
ate
is u
nable
to
speak
without
pro
mpting.]
[Allo
w10
seconds.
Do n
ot
use
back-u
p q
uestions.]
[Retr
ieve
mate
rials
.]
*US
EC
AN
DID
AT
ES
’ N
AM
ES
TH
RO
UG
HO
UT
TH
ET
ES
T
BL
AN
K P
AG
E
I'm
goin
gto
giv
e e
ach
of
you a
card
with 2
topic
s.
I'd
like
you t
ochoose
on
eto
pic
and
talk
about
itfo
rabout
a m
inute
.Y
ou h
ave o
ne m
inute
topre
pare
for
this
.Y
ou a
re a
llow
ed t
om
ake n
ote
s.
All
rig
ht?
Here
are
your
topic
s.
Choose o
ne
ofth
e topic
s a
nd
pre
pare
to
talk
about
it.
Rem
em
ber
you c
an
make n
ote
s if
you w
ish.
All
rig
ht.
Now
,*B
, w
hic
h t
opic
have y
ou c
hosen,
A o
rB
?[c
andid
ate
answ
ers
]W
ould
you lik
eto
talk
about
what
you t
hin
kis
import
ant
when (
inte
rlocuto
rsta
tes c
andid
ate
’s c
hosen t
opic
)?
Thank y
ou.
Now
,*A
,w
hic
h d
o y
ou t
hin
k is m
ost
import
ant
(bulle
t1,
2or
3)?
Thank y
ou.
All
rig
ht.
Now
,*A
,w
hic
h t
opic
have y
ou c
hosen,
Aor
B? [
candid
ate
answ
ers
]W
ould
you lik
e to
talk
about
what
you thin
k is im
port
ant
when (
inte
rlocuto
rsta
tes c
andid
ate
’schosen t
opic
)?
Thank y
ou.
Now
,*B
,w
hic
h d
o y
ou t
hin
k is m
ostim
port
ant
(bulle
t1,
2or
3)?
Thank y
ou.
That's
the e
nd
ofth
e f
irst
part
of
the t
est.
In t
he
next
part
you
are
each
goin
gto
talk
on
your
ow
n.
Thank y
ou.
Appendix 4.1
319
Task C
ard
11 –
Exam
iner’
sC
op
y
Are
care
er
op
po
rtu
nit
ies
the m
ost
import
antth
ing a
bout
anew
job? (
Why/W
hy
not?
)Is
th
elo
cati
on
ofth
ejo
b im
port
ant?
(W
hy/W
hy
not?
)H
ow
import
ant
is t
he
opport
unity
tom
ake b
usin
ess t
rip
s? (
Why/W
hy
not?
)
Sele
ct
from
the
follo
win
g a
dditio
nal pro
mpts
(if
the a
bove h
ave
alr
eady
been
covere
d):
Is s
ala
ryth
e m
ost
import
ant th
ing a
bout
a job?
(Why/W
hy
not?
)A
re t
rain
ing
op
po
rtu
nit
ies im
port
ant?
(Why/W
hy n
ot?
)Is
afr
ien
dly
wo
rkin
g e
nvir
on
men
tim
port
ant
in a
new
job? (
Why/W
hy
not?
)
Are
fli
gh
td
ep
art
ure
tim
es
im
port
ant?
(W
hy/W
hy
not?
)H
ow
import
ant
is t
he
serv
ice o
na
flig
ht?
(Why/W
hy
not?
)Is
it
essentialto
have
good
tran
sp
ort
toan
dfr
om
th
eair
po
rts
? (
Why/W
hy
not?
)
Sele
ct
from
the
follo
win
g a
dditio
nal pro
mpts
(if
the a
bove h
ave
alr
eady
been
covere
d):
How
import
ant
is t
he
foo
don
a p
lane? (
Why/W
hy
not?
)H
ow
import
ant
is t
he
am
ou
nt
of
lug
gag
epassengers
can
take o
na
busin
ess
flig
ht?
(Why/W
hy
not?
)W
hic
hair
po
rtfa
cil
itie
sare
usefu
lfo
rbusin
ess
travelle
rs? (
Why?)
B:
WH
AT
IS I
MP
OR
TA
NT
WH
EN
…?
TR
AV
EL
LIN
GB
Y A
IR F
OR
BU
SIN
ES
S
FL
IGH
TD
EP
AR
TU
RE
TIM
ES
IN-F
LIG
HT
SE
RV
ICE
TR
AN
SP
OR
TT
OA
ND
FR
OM
AIR
PO
RT
S
A:
WH
AT
IS I
MP
OR
TA
NT
WH
EN
…?
LO
OK
ING
FO
R A
NE
W J
OB
CA
RE
ER
OP
PO
RT
UN
ITIE
S
LO
CA
TIO
N O
FJO
B
PO
SS
IBIL
ITY
OF
MA
KIN
G B
US
INE
SS
TR
IPS
Appendix 4.1
320
Task C
ard
14 –
Exam
iner’
sC
op
y
Are
co
urs
em
ate
ria
lsth
em
ost
import
ant
thin
gto
consid
er?
(Why/W
hy
not?
)Is
it
import
ant
who
the
train
er
is? (
Why/W
hy
not?
)H
ow
import
ant
is t
he
nu
mb
er
of
part
icip
an
tsin
the
gro
up?
(Why/W
hy
not?
)
Sele
ct
from
the
follo
win
g a
dditio
nal pro
mpts
(if
the a
bove h
ave
alr
eady
been
covere
d):
Are
opport
unitie
s to
get
qu
ali
ficati
on
sessential?
(W
hy/W
hy
not?
)H
ow
import
ant
is c
ost?
(W
hy/W
hy
not?
)W
hy
isit
import
ant
toconsid
er
the
tim
ean
d len
gth
ofeach
lesson?
Is s
peed
of
serv
ice
the m
ost
import
ant
thin
gto
consid
er?
(W
hy/W
hy
not?
)H
ow
import
ant
is c
ost?
(W
hy/W
hy
not?
)Is
it im
port
antto
ask
oth
er
people
for
their
rec
om
me
nd
ati
on
s? (
Why/W
hy
not?
)
Sele
ct
from
the
follo
win
g a
dditio
nal pro
mpts
(if
the a
bove h
ave
alr
eady
been
covere
d):
Isit e
ssential fo
rth
e d
eliv
ery
com
pany
toin
clu
de i
ns
ura
nce? (
Why/W
hy
not?
)H
ow
import
ant
is t
he
siz
e o
fth
e d
eliv
ery
com
pany? (
Why/W
hy
not?
)H
ow
import
ant
are
dis
co
un
tsfr
om
the c
om
pany? (
Why/W
hy
not?
)
A:
WH
AT
IS I
MP
OR
TA
NT
WH
EN
…?
JO
ININ
G A
CO
MP
UT
ER
SK
ILL
S C
OU
RS
E
CO
UR
SE
MA
TE
RIA
LS
TR
AIN
ER
NU
MB
ER
OF
PA
RT
ICIP
AN
TS
IN
GR
OU
P
B:
WH
AT
IS I
MP
OR
TA
NT
WH
EN
…?
CH
OO
SIN
G A
DE
LIV
ER
YC
OM
PA
NY
SP
EE
D O
FS
ER
VIC
E
CO
ST
PE
RS
ON
AL
RE
CO
MM
EN
DA
TIO
N
Task C
ard
13 –
Exam
iner’
sC
op
y
Is th
ety
pe o
ftr
an
sp
ort
import
ant?
(Why/W
hy
not?
)H
ow
import
ant
is t
he
acco
mm
od
ati
on
availa
ble
? (
Why/W
hy
not?
)Is
itim
port
antto
have a
co
nta
ct
pers
on
abro
ad? (
Why/W
hy
not?
)
Sele
ct
from
the
follo
win
g a
dditio
nal pro
mpts
(if
all
the a
bove a
re c
overe
d):
How
import
ant
is t
he
co
st
ofth
e tri
p? (
Why/W
hy
not?
)Is
itim
port
antto
have c
oll
eag
ues
wit
h y
ou
on a
busin
ess tri
p?
(Why/W
hy
not?
)Is
it e
ssential to
sp
eak t
he lan
gu
ag
e?(W
hy/W
hy
not?
)
Is lo
cati
on
the m
ost
import
ant
thin
gto
consid
er?
(Why/W
hy
not?
)W
hatkin
ds o
f fa
cil
itie
sin
the o
ffic
es
are
essential?
(W
hy?)
How
import
ant
is c
ost?
(W
hy/W
hy
not?
)
Sele
ct
from
the
follo
win
g a
dditio
nal pro
mpts
(if
the a
bove h
ave
alr
eady
been
covere
d):
Why
isit
import
ant
toconsid
er
the
siz
eofth
e p
rem
ises?
How
import
ant
is t
he
co
nd
itio
nofth
e b
uild
ing
?(W
hy/W
hy
not?
)A
repark
ing
spaces
for
sta
ffessen
tial?
(Why/W
hy
not?
)
A:
WH
AT
IS I
MP
OR
TA
NT
WH
EN
…?
AR
RA
NG
ING
AB
US
INE
SS
TR
IP A
BR
OA
D
TY
PE
OF
TR
AN
SP
OR
T
AC
CO
MM
OD
AT
ION
CO
NT
AC
TP
ER
SO
N
B:
WH
AT
IS I
MP
OR
TA
NT
WH
EN
…?
CH
OO
SIN
G O
FF
ICE
ST
O R
EN
T
LO
CA
TIO
N
FA
CIL
ITIE
S
CO
ST
Appendix 4.1
321
Appendix 4.1
322
Wri
tin
g S
am
ple
Sc
rip
ts
Part
1
Scri
pt
A
Dea
r co
lleag
ues,
Aft
erus
ing
the
com
pany
car
s, p
leas
e fi
ll th
em u
p w
ith
petr
olan
d lea
ve t
he
cars
in f
ront
of
our
bui
ldin
g on
our
com
pany
car
par
king
pla
ces.
Then
give
the
car
key
to o
ur t
eam
sec
reta
ry.
Man
y th
anks
.
Scri
pt
A
All
conte
nt
po
ints
ach
ieved
within
th
e s
pecifie
d w
ord
lim
it w
ith n
o a
mbig
uity
and m
inim
al eff
ort
fro
mth
ere
ader.
Som
e s
imple
lin
kin
g d
evic
es
are
use
d. T
he s
tyle
is f
rie
ndly
but
not
info
rmal.
Ba
nd
5
Scri
pt
B
I’ll r
em
ind y
outh
at y
ou m
ust
fill t
he c
ompa
ny c
ar w
ith p
etr
olon
their
retu
rn.
Aft
erth
at y
oufo
llow
the p
roce
dur
es
for
park
ing
as y
ou k
now
and
depo
sit
the
keys
in
the
blu
bas
ket.
Scri
pt
B
The f
irst and thir
dconte
nt
poin
ts a
reachie
ved b
ut th
e r
eader
wou
ld n
ot kno
w‘w
here
to lea
ve
the c
ar’
as s
pecifie
d in t
he q
uestion.
Oth
erw
ise th
e q
ua
lity
of
the
lang
uag
e a
nd s
tyle
is a
deq
uate
for
this
level
and n
o m
isunders
tan
din
gs a
re c
rea
ted.
It w
ould
be a
ccepte
d t
hat
peo
ple
work
ing w
ith
in t
he c
om
pan
yw
ould
know
wha
t ‘the b
lue b
asket’ w
as a
nd
wh
ere
it
was
locate
d.
Ba
nd
3
Scri
pt
C
Pleas
e f
ill th
e c
ompa
ny c
ars
hav
e b
een
dri
ved in
case
the n
ext
one c
ould
n’t
dri
ve
it a
gain
.
And
als
o fo
llow
ing
the
inst
ruct
ions
of
the
park
ing
order
. Ple
ase
leav
e th
e dep
osit
ofth
eke
yis
wel
l.T
han
k yo
u
Scri
pt
C
Althou
gh t
he c
andid
ate
has tri
ed
to a
ddre
ss
the c
onte
nt
po
ints
, n
on
e o
f th
em
has b
ee
n a
ch
ieved
,eith
er
beca
use o
fin
suff
icie
nt
langu
age
contr
ol (p
oin
t 1)
or
because t
he instr
uctions a
reto
o v
agu
e(p
oin
ts2 a
nd 3
).B
and 1
is a
ward
ed r
ath
er
than B
an
d 0
as the
cand
idate
has
wri
tten 3
8 w
ord
s a
nd
th
ere
sponse is n
ot to
tally
irre
leva
nt.
An
sw
er
Ke
ys
Read
ing
Part
1
1C
2B
3A
4C
5F
Part
4
16
B17
C18
C19
B20
A21
A22
A
Part
6
29
B30
A31
C32
B33
C34
A35
A36
BP
art
2
6F
7D
8G
9A
10
E
Part
5
23
C24
A25
B26
C27
A28
C
37
C38
C39
B40
C
Part
3
11
F12
D13
C14
B15
H
Part
7
41 0
1/3
434/A
42 M
adri
d43
Busin
ess (
Cla
ss)
44 (
a)/
(by)
cheq
ue
45 illn
ess/ sic
kness/ sic
k/ unw
ell/
not
well
Lis
ten
ing
Part
1
1B
2C
3A
4A
5A
6B
7C
8
A
Part
3
16 H
ill17 jun
ior
sale
s18 (
ne
w)
soft
ware
(pro
gra
mm
e(s
))19 w
ind
ow
/sho
p d
ispla
y20 (
the
/our)
Chair
man/(
Com
pan
y)
Ch
air
man
21 (
13
th o
f) J
uly
22 e
xp
ort
(s)/
export
ing (
po
licy)
Part
2
9 N
augh
ton
10 W
oodes
11 4
533
39
12 2
50
13 8
14 J
Q165
15 5
90 (
mm
)
Part
4
23
B24
B25
C26
A27
C28
A29
A30
B
Appendix 4.1
323
Scri
pt
F
Dea
r M
rC
rave
n:M
y ap
olog
is f
or t
he
del
ay, I
’m r
ealy
sor
ry b
ut t
he
del
ay w
as b
ecau
se t
he
mot
ors
arri
ved lat
er in
my
com
pany
, I’
m g
oing
to
send
the
mot
orto
mor
ro a
t 9
.00
am
, I
like
to g
ive
to y
ou a
dis
coun
t in
the
tota
l pr
ice
the
dis
coun
t is
20
% o
ff in
the
tota
l.S
orry
for
allt
he
unco
nven
ence
.Y
our
fait
hfu
ly
Igna
cio
Daz
ziM
anag
er
Scri
pt
F
The letter
isw
ell
laid
out
with
an a
ppro
pri
ate
sa
luta
tion a
nd c
lose.
All
4 c
onte
nt p
oin
tsha
ve b
ee
nachie
ve
d b
ut th
ere
are
a n
um
ber
of
err
ors
in s
pelli
ng, pu
nctu
ation a
nd s
ente
nce s
tructu
re. T
he t
ext is
quite s
hort
. T
here
is a
bare
ly a
de
qua
te r
ange o
f str
uctu
re a
nd
vocabu
lary
dis
pla
ye
d f
or
this
leve
l.
Ba
nd
3
Part
2
Scri
pt
D
Dea
r M
rC
rave
n,T
han
k yo
u fo
ryo
urfa
x o
n th
e 3
Jun
e.I
apol
ogis
e ab
out
the
del
ay o
f ou
rpr
oduc
t. I
reg
ret
to inf
orm
you
that
the
mot
ors
are
out
of s
tock
righ
t no
w. W
ew
ill o
bta
in in
2 d
ays.
In
addit
ion,
we
will
send
the
mot
ors
to y
ou t
his
frid
ay o
n th
e 7
Jun
e. I
wou
ld lik
eto
giv
e yo
u 5
%dis
coun
t.If
you
req
uire
mor
e in
form
atio
n, p
leas
don
’t b
e so
hes
itat
e to
cont
act
us. T
han
kyo
u fo
ryo
ur c
o-op
erat
ion.
You
rsi
ncer
ely
J S
char
renb
erg
Scri
pt
D
All
conte
nt
po
ints
have b
een a
chie
ved. D
espite s
om
e inaccura
cie
s,
the m
essage is
well
org
anis
ed a
nd
is a
lwa
ys
cle
ar.
Som
e g
ood f
unction
al la
ngu
age
is
use
d a
nd t
he t
one o
f th
e lett
er
isa
polo
ge
tic a
nd
respectf
ul to
the c
usto
mer.
Ba
nd
5
Scri
pt
E
Dear
Mr
Cra
ven,
We
hav
e j
ust
rece
ived y
our
fax d
ated 3
Jun
e a
nd w
e a
polo
gise
for
what
has
hap
pene
d.
We
mus
t cl
ose o
urof
fice
for
one w
eek
beca
use w
e h
ave b
een
stol
len.
You
willre
ceiv
e th
ese
mot
ors
by
15 J
une a
nd w
e h
ave
reduc
ed a
t 10
% o
n th
e b
ill.
If
you
need
mor
e inf
orm
atio
n don
’t h
esi
tade t
o co
ntac
t ou
r dri
ver
on 7
77
.39
.77
.
You
rssi
ncere
ly,
Han
na V
isco
nti
Scri
pt
E
There
is a
go
od r
ange
of
tenses, a
n a
ppro
pri
ate
sty
le a
nd s
om
every
good
functiona
l la
ngu
age
. T
he
info
rmation is
we
ll org
anis
ed
bu
t th
e c
and
idate
co
uld
ha
ve
made m
ore
eff
ort
to lin
k the s
ente
nces
and
para
gra
phs to
geth
er.
The s
econd
conte
nt p
oin
t re
quir
es
co
nsid
era
ble
inte
rpre
tation o
n t
he p
art
of
the
reader.
Ba
nd
4
Appendix 4.1
324
Superv
isor:
PR
EL
IM
IN
AR
Y
BE
C P
relim
inary
Read
ing
An
sw
er
Sh
eet
00
00 1 2 3 4 5 6 7 8 9
1 2 3 4 5 6 7 8 9
1 2 3 4 5 6 7 8 9
1 2 3 4 5 6 7 8 9
Can
did
ate
Nam
eIf
no
t a
lre
ad
y p
rin
ted
, w
rite
na
me
in C
AP
ITA
LS
an
d c
om
ple
te t
he
Can
did
ate
No
. g
rid
(in
pen
cil).
Can
did
ate
’s S
ign
atu
re
Exam
inati
on
Tit
le
Cen
tre
If t
he
ca
nd
ida
te is A
BS
EN
To
r h
asW
ITH
DR
AW
Nsh
ad
e h
ere
Can
did
ate
No
.
Ce
ntr
e N
o.
Exam
inati
on
Deta
ils
21 43
6
5
87 109
Part
1
Instr
ucti
on
sU
se
a P
EN
CIL
(B
or
HB
).
Ru
b o
ut
an
y a
nsw
er
yo
u w
ish
to
ch
an
ge
with
an
era
se
r.
Fo
r P
art
s 1
to
6:
Ma
rk o
ne
bo
x f
or
ea
ch
an
sw
er.
Fo
r e
xa
mp
le:
If y
ou
th
ink C
is t
he
rig
ht
an
sw
er
to t
he
qu
estio
n,
ma
rk y
ou
r a
nsw
er
sh
ee
t lik
e t
his
:
Part
2
�
Tu
rn o
ver
for
Part
s 3
- 7
Fo
r P
art
7:
Write
yo
ur
an
sw
er
cle
arly in
CA
PIT
AL
LE
TT
ER
S.
Write
on
e le
tte
r o
r n
um
be
r in
ea
ch
bo
x.
If t
he
an
sw
er
ha
s m
ore
th
an
on
e w
ord
, le
ave
on
e b
ox e
mp
ty b
etw
ee
n w
ord
s.
Fo
r e
xa
mp
le:
AB
C0
0
A
BE
C P
- R
DP
45
3/3
53
BC
AB
C
AB
C
AB
C
AB
C
AB
CD
EF
GH
AB
CD
EF
GH
AB
CD
EF
GH
AB
CD
EF
GH
AB
CD
EF
GH
Appendix 4.1
325
Superv
isor:
This
section for
use b
y E
xam
iner
only
PR
EL
IM
IN
AR
Y
BE
C P
relim
inary
Wri
tin
g A
nsw
er
Sh
eet
00
00 1 2 3 4 5 6 7 8 9
1 2 3 4 5 6 7 8 9
1 2 3 4 5 6 7 8 9
1 2 3 4 5 6 7 8 9
Can
did
ate
Nam
eIf
no
t a
lre
ad
y p
rin
ted
, w
rite
na
me
in C
AP
ITA
LS
an
d c
om
ple
te t
he
Can
did
ate
No
. g
rid
(in
pen
cil).
Can
did
ate
’s S
ign
atu
re
Exam
inati
on
Tit
le
Cen
tre
If t
he
ca
nd
ida
te is A
BS
EN
To
r h
asW
ITH
DR
AW
Nsh
ad
e h
ere
Can
did
ate
No
.
Ce
ntr
e N
o.
Exam
inati
on
Deta
ils
Part
1:
Wri
te y
ou
r an
sw
er
in t
he b
ox b
elo
w.
�
Wri
te y
ou
r an
sw
er
to P
art
2 o
n t
he o
ther
sid
e o
f th
is s
heet
Part
10
12
34
5
BE
C P
- W
DP
45
4/3
54
Part
7
14
13
15
41
01
41
29
31
30
32P
art
6 AB
C
AB
C
AB
C
AB
C
AB
CD
EF
GH
AB
CD
EF
GH
AB
CD
EF
GH
33
35
34
36
AB
C
AB
C
AB
C
AB
C
37
39
38
40
AB
C
AB
C
AB
C
AB
C
Part
3
16
AB
C2
3A
BC
12
11
AB
CD
EF
GH
AB
CD
EF
GH
Part
4P
art
5
18
17
19
AB
C
AB
C
AB
C
20
21
AB
C
AB
C
22
AB
C
25
24
26
AB
C
AB
C
AB
C
27
28
AB
C
AB
C
42
01
42
43
01
43
44
01
44
45
01
45
Appendix 4.1
326
Superv
isor:
0
PR
EL
IM
IN
AR
Y
BE
C P
relim
inary
Lis
ten
ing
An
sw
er
Sh
eet
00
00 1 2 3 4 5 6 7 8 9
1 2 3 4 5 6 7 8 9
1 2 3 4 5 6 7 8 9
1 2 3 4 5 6 7 8 9
Can
did
ate
Nam
eIf
no
t a
lre
ad
y p
rin
ted
, w
rite
na
me
in C
AP
ITA
LS
an
d c
om
ple
te t
he
Can
did
ate
No
. g
rid
(in
pen
cil).
Can
did
ate
’s S
ign
atu
re
Exam
inati
on
Tit
le
Cen
tre
If t
he
ca
nd
ida
te is A
BS
EN
To
r h
asW
ITH
DR
AW
Nsh
ad
e h
ere
Can
did
ate
No
.
Ce
ntr
e N
o.
Exam
inati
on
Deta
ils
21 43 5Part
1
Instr
ucti
on
sU
se
a P
EN
CIL
(B
or
HB
).
Ru
b o
ut
an
y a
nsw
er
yo
u w
ish
to
ch
an
ge
with
an
era
se
r.
Fo
r P
art
s 1
an
d 4
:
Ma
rk o
ne
bo
x f
or
ea
ch
an
sw
er.
Fo
r e
xa
mp
le:
If y
ou
th
ink C
is t
he
rig
ht
an
sw
er
to t
he
qu
estio
n,
ma
rk y
ou
r a
nsw
er
sh
ee
t lik
e t
his
:
Part
2
�
Tu
rn o
ver
for
Part
s 3
an
d 4
Fo
r P
art
s 2
an
d 3
:
Write
yo
ur
an
sw
er
cle
arly in
CA
PIT
AL
LE
TT
ER
S.
Write
on
e le
tte
r in
ea
ch
bo
x.
If t
he
an
sw
er
ha
s m
ore
th
an
on
e w
ord
, le
ave
on
e b
ox e
mp
ty b
etw
ee
n w
ord
s.
Fo
r e
xa
mp
le:
AB
C0
6 7 8
109
01
01
12
11
01
01
13
01
9 10
11
12
13
14
01
15
01
14
15
A
BE
C P
- L
DP
45
6/3
56
BC
AB
C
AB
C
AB
C
AB
C
AB
C
AB
C
AB
C
This
section for
use b
y E
xam
iner
only
Part
2:
Wri
te y
ou
r an
sw
er
in t
he b
ox b
elo
w.
Part
20
1.1
1.2
Exam
iner
Num
ber
Exam
iner’
s S
ignatu
re
2.1
2.2
3.1
3.2
4.1
4.2
5.1
5.2
87
65
43
21
09
87
65
43
21
09
87
65
43
21
09
87
65
43
21
09
Appendix 4.1
327
Fe
ed
ba
ck
Fo
rm
BE
C P
reli
min
ary
Exam
inati
on
Rep
ort
Ju
ne 2
002
We a
re inte
reste
d in h
eari
ng y
our
vie
ws o
n h
ow
usefu
l th
is r
eport
has b
een.
We w
ou
ld b
e m
ost gra
tefu
l if
yo
u c
ould
bri
efly a
nsw
er
the f
ollo
win
g q
uestions
and r
etu
rn a
ph
oto
co
py
of
this
pa
ge t
o th
e f
ollo
win
g a
ddre
ss:
BE
CR
ep
ort
s C
o-o
rdin
ato
rC
am
bri
dge E
SO
L1 H
ills R
oad
Cam
bri
dge
CB
1 2
EU
UK
Fax: 44 1
223
46
027
8
1.
Ple
ase d
escri
be y
our
situa
tio
n: (e
.g.
EF
L t
eacher,
Dir
ecto
rof
Stu
die
s,
Exam
inatio
ns O
ffic
er,
Local
Secre
tary
, etc
)
2.
Have y
ou p
repare
d c
an
did
ate
s f
or
BE
C P
relim
inary
?
3.
Do y
ou p
lan to
pre
pare
cand
idate
s f
or
BE
C P
relim
inary
in t
he f
utu
re?
4.
Ho
wh
ave
yo
uused t
his
rep
ort
? (
e.g
. to
pro
vid
e f
eed
back
to o
ther
teach
ers
, fo
rexam
ination
pra
ctice, e
tc.)
5.
Whic
h p
art
s o
fth
is r
eport
did
you f
ind m
ost usefu
l?
6.
Whic
h p
art
s w
ere
not so u
sefu
l?
7.
What extr
a info
rmation w
ould
yo
u lik
e to s
ee inclu
de
d in
th
is r
ep
ort
?
8.
Your
nam
e: (O
ption
al)
Centr
e/S
cho
ol:
Tha
nk y
ou
.
Part
3
16
01
16
23
AB
C
Part
4
25
24
26
AB
C
AB
C
AB
C
27
28
AB
C
AB
C
29
AB
C
17
01
17
18
01
18
19
01
19
20
01
20
21
01
21
22
01
22
30
AB
C
328
TIME 1 hour
INSTRUCTIONS TO CANDIDATES
Do not open this paper until you are told to do so.
Write your name, Centre number and candidate number in the spaces at the top of this page. Write these
details in pencil on your Answer Sheet if these are not already printed.
Write all your answers in pencil on your Answer Sheet – no extra time is allowed for this.
Read carefully the instructions for each part and the instructions for completing your Answer Sheet.
Try to answer all the questions.
At the end of the examination hand in both this question paper and your Answer Sheet.
INFORMATION FOR CANDIDATES
There are forty-five questions on this question paper.
UNIVERSITY OF CAMBRIDGE LOCAL EXAMINATIONS SYNDICATE
Examinations in English as a Foreign Language
BUSINESS ENGLISH CERTIFICATE 0352/1Vantage
Test of Reading Test 023
Thursday 6 JUNE 2002 Morning 1 hour
Additional materials:Answer Sheet
Candidate
Centre Number Number
Candidate Name
This question paper consists of 10 printed pages.
SP (SLC) S26502/3
APPENDIX 4.2
BEC Vantage Sample Paper
Appendix 4.2
329
Ch
ang
es i
n P
erfo
rman
ce F
eed
bac
k
A B C D
So
, h
ow
is t
he
ne
w f
ee
db
ack c
ultu
re l
ike
ly t
o a
ffe
ct
yo
u?
It
co
uld
fo
rm t
he
ba
sis
of
yo
ur
pe
rso
na
l d
eve
lop
me
nt
pro
gra
mm
e,
pro
vid
ing
po
inte
rs t
o y
ou
r str
en
gth
s a
nd
als
o t
o t
ho
se
are
as y
ou
ne
ed
to
de
ve
lop
mo
re.
Or
fee
db
ack c
ou
ld b
e u
se
d f
or
‘su
cce
ssio
n p
lan
nin
g’,
where
co
mp
an
ies u
se
th
e i
nfo
rma
tio
n t
o s
pe
cu
late
on
wh
o h
as t
he
rig
ht
skill
s t
o m
ove
in
to m
ore
se
nio
r p
ositio
ns. A
s y
et,
fe
w o
rga
nis
atio
ns h
ave
str
etc
he
d th
e r
ole
of fe
ed
ba
ck s
o fa
r a
s to
lin
k
it to s
ala
ries. B
ut one thin
g is
cle
ar:
the futu
re w
ill b
ring e
ven w
ider
part
icip
ation b
y a
ll m
em
bers
Sa
rah
Ra
ins,
fro
m t
he
ph
arm
ace
utica
l co
mp
an
y O
pte
c,
sa
id,
‘No
w f
ee
db
ack i
s a
va
ilab
le o
n
ou
r n
etw
ork
, w
e e
nco
ura
ge
ma
na
ge
rs to
ch
oo
se
ho
w th
ey u
se
it.
It is
a fle
xib
le to
ol a
nd
we
te
ll
them
that
waitin
g f
or
the a
nnual event
of
a f
orm
al appra
isal needn’t
apply
.’A
t th
e e
ng
ine
erin
g
co
mp
an
y N
T, 2
50
te
ch
nic
al m
an
ag
ers
ha
ve
be
en
th
rou
gh
th
e fe
ed
ba
ck p
roce
ss. Ja
ck P
alm
er,
a
se
nio
r m
an
ag
er
the
re,
sa
id,
‘We
n
ee
de
d
to
de
ve
lop
th
e
inte
rpe
rso
na
l skill
s
of
the
se
technic
ally
-min
ded
people
. In
part
icula
r,
we
wante
d
to
build
on
their
team
-work
ing
and
coachin
g s
kill
s.’
PC
s lin
ke
d t
o t
he
co
mp
an
y I
Tn
etw
ork
are
se
t to
be
co
me
th
e f
ee
db
ack m
ach
ine
s.
Ma
ny f
irm
s
intr
od
ucin
g
36
0-d
eg
ree
fe
ed
ba
ck
are
u
sin
g
Pe
rso
na
l D
eve
lop
me
nt
Pla
nn
er
so
ftw
are
.
Feedback o
n a
n indiv
idual, w
hic
h is b
ased o
n a
questionnaire r
ela
ting to a
ttribute
s n
eeded for
tha
t p
ers
on’s
role
in t
he c
om
pany,
is c
olle
cte
d u
sin
g t
his
ele
ctr
onic
syste
m. A
ll th
e info
rmation
ga
the
red
is a
na
lyse
d a
nd
th
e e
nd
re
su
lt is a
su
gg
este
d d
eve
lop
me
nt
pla
n.
Th
e a
dva
nta
ge
is
tha
t in
div
idu
als
ma
ke
re
qu
ests
fo
r th
e f
ee
db
ack t
he
mse
lve
s a
nd
re
ce
ive
th
e r
esu
lts d
ire
ctly.
In t
he
pa
st,
fe
ed
ba
ck a
bo
ut
yo
ur
pe
rfo
rma
nce
use
d t
o m
ea
n a
qu
iet
ch
at
with
th
e b
oss.
Bu
t
no
w 3
60
-de
gre
e fe
ed
ba
ck –
th
e s
yste
m w
he
re e
mp
loye
es a
re a
lso
giv
en
fe
ed
ba
ck fro
m p
ee
rs
an
d fr
om
th
e p
eo
ple
th
ey m
an
ag
e –
is ta
kin
g ro
ot
in co
rpo
rate
cu
ltu
re.
Th
e syste
m is
chara
cte
rised b
y g
reate
r part
icip
ation a
nd h
as g
row
n o
ut
of
the d
esire o
f com
panie
s t
o c
reate
mo
re o
pe
n w
ork
ing
en
viro
nm
en
ts w
he
re p
eo
ple
wo
rk b
ett
er
tog
eth
er
an
d id
ea
s a
nd
op
inio
ns
are
exch
an
ge
d b
etw
ee
n t
ea
ms a
nd
acro
ss le
ve
ls o
f se
nio
rity
.
PA
RT
ON
E
Qu
esti
on
s 1
– 7
•Lo
ok a
t th
e st
atem
ents
bel
ow a
nd t
he in
form
atio
n on
the
opp
osite
pag
e ab
out
feed
back
on
staf
f
perf
orm
ance
.
•W
hich
sec
tion
(A,
B,
Cor
D)
does
eac
h st
atem
ent
1 –
7re
fer
to?
•F
or e
ach
stat
emen
t 1
– 7,
mar
k on
e le
tter
(A,
B,
Cor
D)
on y
our
Ans
wer
She
et.
•Y
ou w
ill n
eed
to u
se s
ome
of t
hese
lette
rs m
ore
than
onc
e.
Exa
mp
le:
0th
e re
luct
ance
of
com
pani
es t
o ba
se p
ay o
n st
aff
feed
back
1st
aff
bein
g re
min
ded
that
it is
not
ess
entia
l to
rest
rict
feed
back
to
once
a y
ear
2th
e w
ay in
whi
ch f
eedb
ack
coul
d id
entif
y pe
ople
sui
tabl
e fo
r pr
omot
ion
3th
e ai
m o
f im
prov
ing
staf
f co
mm
unic
atio
n th
roug
hout
an
orga
nisa
tion
4th
e fe
edba
ck o
btai
ned
on a
n em
ploy
ee b
eing
link
ed t
o re
quire
men
ts f
or a
par
ticul
ar jo
b
5as
pect
s of
a g
roup
of
empl
oyee
s’w
ork
that
wer
e id
entif
ied
as r
equi
ring
impr
ovem
ent
6fe
edba
ck in
dica
ting
both
pos
itive
and
neg
ativ
e as
pect
s of
an
indi
vidu
al’s
wor
k
7th
e pa
rtic
ipat
ion
of le
ss s
enio
r pe
rson
nel i
n a
mem
ber
of s
taff’
s fe
edba
ck
0A
BC
D
Appendix 4.2
330
Exa
mp
le:
AIf
an
org
an
isa
tio
n is ju
st
be
gin
nin
g t
o w
ork
glo
ba
lly a
nd
ha
s o
nly
re
ce
ntly c
rea
ted
inte
rnational te
am
s,
it o
ften
un
de
restim
ate
s t
he
le
ve
l o
f su
pp
ort
ne
ed
ed
by t
ea
ms.
BIt
is n
ow
we
ll-e
sta
blis
he
d t
ha
t a
ny t
ea
m
will
ha
ve
a g
rea
ter
ch
an
ce
of
su
cce
ss if
it
ha
s c
lea
r g
oa
ls,
a s
tro
ng
se
nse
of
co
mm
itm
en
t, a
pp
rop
ria
te le
ad
ers
hip
an
d
go
od
in
terp
ers
on
al re
latio
nsh
ips.
CT
he
re
co
gn
itio
n o
f th
is h
as c
rea
ted
ma
ny
mo
re k
no
wle
dg
e w
ork
ers
, th
at
is,
pe
op
le
wh
o c
rea
te,
exch
an
ge
an
d b
roa
dca
st
info
rma
tio
n a
s k
no
wle
dg
e.
DO
rga
nis
atio
ns m
ust
un
de
rsta
nd
th
at
op
era
tin
g g
lob
ally
affe
cts
eve
ry a
sp
ect
of
bu
sin
ess,
an
d t
he
y c
an
no
t sim
ply
se
t u
p
inte
rna
tio
na
l te
am
s a
nd
assu
me
th
at
eve
ryth
ing
els
e c
an
re
ma
in u
nch
an
ge
d.
ET
he f
irst
majo
r im
pact
of
this
is t
hat
sa
telli
te t
ech
no
log
y is in
cre
asin
gly
allo
win
g
tea
m m
em
be
rs t
o p
art
icip
ate
in
dis
cu
ssio
ns w
he
reve
r th
ey a
re,
at
an
y t
ime
the
y c
ho
ose
.
FU
nfo
rtu
na
tely
, h
ow
eve
r, f
ew
org
an
isa
tio
ns
un
til n
ow
ha
ve
be
en
pre
pa
red
to
ma
ke
th
e
ne
ce
ssa
ry in
ve
stm
en
t to
ga
in t
he
po
ten
tia
l
be
ne
fits
th
at
inte
rna
tio
na
l te
am
s o
ffe
r.
GU
nlik
e m
ost
na
tio
na
l te
am
s,
inte
rna
tio
na
l
tea
ms o
fte
n w
ork
ap
art
an
d a
cro
ss
cu
ltu
res a
nd
tim
e z
on
es,
for
exte
nd
ed
pe
rio
ds o
f tim
e.
AB
CD
EF
G0
PA
RT
TW
O
Qu
esti
on
s 8
– 12
•R
ea
d t
he
art
icle
be
low
ab
ou
t w
ork
ing
in
in
tern
atio
na
l te
am
s.
•C
ho
ose
th
e b
est
se
nte
nce
fro
m t
he
op
po
site
pa
ge
to
fill
ea
ch
of
the
ga
ps.
•F
or
ea
ch
ga
p 8
– 1
2, m
ark
one letter
(A–
G)
on
yo
ur
An
sw
er
Sh
ee
t.
•D
o n
ot
use
an
y le
tte
r m
ore
th
an
on
ce
.
•T
he
re is a
n e
xa
mp
le a
t th
e b
eg
inn
ing
, (0
).
An
in
tern
atio
na
l te
am
ca
n b
e d
efin
ed
as a
gro
up
of
pe
op
le w
ho
co
me
fro
m d
iffe
ren
t n
atio
na
litie
s
an
d
wo
rk
tog
eth
er
tow
ard
s
a
co
mm
on
g
oa
l.
(0).
.......G ...
..
Th
e fa
ct
tha
t th
ey a
re sp
rea
d o
ut
pre
se
nts
a r
an
ge
of o
pp
ort
un
itie
s a
nd
ch
alle
ng
es
tha
t te
am
s w
ork
ing
in
th
e sa
me
p
lace
d
o n
ot
exp
erie
nce
.
On
e tre
nd
in p
art
icu
lar
wh
ich
is c
rea
tin
g th
e n
ee
d
for
mo
re in
tern
atio
na
l te
am
s is t
ha
t w
e a
re in
th
e
mid
dle
o
f a
d
ram
atic
info
rma
tio
n
revo
lutio
n.
(8).
....
....
...
. T
hu
s,
the
se
te
am
s c
an
no
w s
pe
nd
as m
uch
tim
e w
ork
ing
ap
art
as t
og
eth
er.
Th
ey
ca
n
acce
ss
an
d
sh
are
in
form
atio
n
as
ne
ve
r
be
fore
. B
usin
ess w
ill i
ncre
asin
gly
be
do
ne
in
an
‘info
rma
tio
n s
pa
ce’,
with
in
form
atio
n b
eco
min
g a
pro
du
ct
in i
ts o
wn
rig
ht.
(9)
............ .
Doin
g t
his
thro
ug
h th
e in
tern
et
an
d e
-ma
il is
in
exp
en
siv
e
an
d
rela
tive
ly
ea
sy,
in
b
oth
te
ch
no
log
ica
lly
de
ve
lop
ed
an
d d
eve
lop
ing
co
un
trie
s.
Aq
ue
stio
n
co
mm
on
ly
aske
d
by
ma
na
ge
rs
is
wh
eth
er
the
se
te
am
s a
ctu
ally
w
ork
. C
an
th
ey
de
live
r im
pro
ve
d p
erf
orm
an
ce
? A
fte
r a
de
ca
de
of
wo
rk e
xp
erie
nce
an
d r
ese
arc
h w
ith
in
tern
atio
na
l
tea
ms,
I b
elie
ve
th
e
an
sw
er
is
po
sitiv
e.
(10)
....
....
....
.
Wh
at’s
m
ore
, m
an
y
of
tho
se
co
mp
an
ies
wh
ich
h
ave
a
ctu
ally
intr
od
uce
d
inte
rna
tio
na
l te
am
s h
ave
fo
cu
se
d o
nly
o
n th
e
pe
rfo
rma
nce
o
f th
e te
am
s,
with
ou
t ta
kin
g in
to
acco
un
t th
e c
on
text in
wh
ich
th
ey a
re in
tro
du
ce
d.
Co
nte
xt
pla
ys a
ke
y r
ole
in
th
e lik
elih
oo
d o
f th
eir
su
cce
ss.
Cre
atin
g t
he
rig
ht
co
nte
xt
for
inte
rna
tio
na
l te
am
s
ne
ed
s m
ore
th
an
a q
uic
k fix
, th
ou
gh
. It
re
qu
ire
s a
long-t
erm
com
mitm
ent.
(11)
............
. O
n
the
co
ntr
ary
, co
mp
an
ies n
ee
d t
o f
ocu
s o
n t
he
wa
y
the
y o
pe
rate
, a
nd
p
ossib
ly in
itia
te a
co
mp
lete
revie
w o
f th
eir p
ractice
s b
efo
re in
tro
du
cin
g a
n
inte
rna
tio
na
l te
am
.
Giv
en
th
ese
ch
alle
ng
es,
wh
at
sh
ou
ld
org
an
isa
tio
ns
do
to
m
ake
su
re
tha
t th
eir
inte
rna
tio
na
l te
am
s a
re su
cce
ssfu
l? M
uch
h
as
be
en
writt
en
ab
ou
t e
ffe
ctive
te
am
pro
ce
sse
s i
n
ge
ne
ral, a
nd
th
e f
irst
thin
g t
o s
ay is t
ha
t m
ost
of
the
se
g
uid
elin
es a
pp
ly e
qu
ally
to
in
tern
atio
na
l
tea
ms.
Exp
erie
nce
ha
s s
ho
wn
th
at
inte
rna
tio
na
l
tea
ms
are
sim
ply
m
ore
co
mp
lex
ve
rsio
ns
of
na
tio
na
l te
am
s.
(12)
....
....
....
.
Wh
ile
th
ese
ele
ments
may h
ave a
variety
of in
terp
reta
tions in
diffe
ren
t cu
ltu
res,
the
y
are
a
s
imp
ort
an
t to
inte
rna
tio
na
l te
am
s
as
the
y
are
to
n
atio
na
l
tea
ms.
Inte
rn
ati
on
al
Te
am
s
Appendix 4.2
331
13In
th
e f
irst
pa
rag
rap
h,
the
write
r sa
ys t
ha
t p
oo
r le
ad
ers
Ad
o n
ot
wa
nt
to lis
ten
to
criticis
m.
Bd
o n
ot
de
se
rve
to
be
ta
ke
n s
erio
usly
.
Ca
re e
asie
r to
id
en
tify
th
an
go
od
on
es.
Da
re m
ore
wid
esp
rea
d t
ha
n p
eo
ple
th
ink.
14W
hy d
oe
s t
he
write
r b
elie
ve
th
ere
is d
isa
gre
em
en
t a
bo
ut
wh
at
effe
ctive
le
ad
ers
hip
is?
AD
efinitio
ns o
f successfu
l le
aders
hip
vary
accord
ing t
o t
he s
ituation.
BT
here
are
few
exam
ple
s o
f outs
tandin
g leaders
availa
ble
to s
tudy.
CLeaders
are
unable
to g
ive c
lear
descriptions o
f th
eir q
ualit
ies.
DT
he
re
su
lts o
f re
se
arc
h o
n t
he
su
bje
ct
ha
ve
co
nclu
de
d little
.
15T
he
pu
blis
hin
g e
xe
cu
tive’s
prio
ritie
s f
or
lea
de
rsh
ip f
ocu
se
d o
n
Asig
nific
an
t a
nd
lo
ng
-te
rm a
ims.
Bin
tern
al o
rga
nis
atio
na
l a
sp
ects
.
Cpro
fessio
nal skill
s a
nd a
bili
ties.
Do
ve
rall
bu
sin
ess c
on
texts
.
16A
cco
rdin
g t
o t
he
write
r, t
he
fin
an
ce
dire
cto
r w
as u
nsu
cce
ssfu
l b
eca
use
Asta
ff w
ere
un
co
mfo
rta
ble
with
his
sty
le.
Be
xis
tin
g c
lien
ts w
ere
su
sp
icio
us o
f ch
an
ge
.
Cco
mp
etito
rs h
ad
a m
ore
dyn
am
ic a
pp
roa
ch
.
Dco
llea
gu
es g
ave
little
su
pp
ort
to
his
id
ea
s.
17S
taff a
t th
e a
ccounta
ncy f
irm
who w
ere
pro
mote
d w
ere
required t
o
Aco
rre
ct
mis
take
s.
Bh
ave
a h
igh
le
ve
l o
f kn
ow
led
ge
.
Cm
ain
tain
dis
cip
line
with
in t
he
org
an
isa
tio
n.
Da
dvis
e c
lien
ts o
n r
esp
on
din
g t
o u
nce
rta
inty
.
18T
he
exa
mp
le o
f th
e m
an
ufa
ctu
rin
g m
an
ag
er
is g
ive
n t
o e
mp
ha
sis
e t
ha
t
Am
an
ag
ers
ne
ed
su
pp
ort
fro
m t
he
ir e
mp
loye
rs.
Ble
ad
ers
sh
ou
ld n
ot
be
afr
aid
of
be
ing
un
po
pu
lar.
Ce
ffe
ctive
le
ad
ers
mu
st
be
se
nsitiv
e t
o s
taff n
ee
ds.
Dm
an
ag
ers
do
no
t a
lwa
ys u
nd
ers
tan
d t
he
att
itu
de
s o
f sta
ff.
PA
RT
TH
RE
E
Qu
esti
on
s 13
– 1
8
•R
ea
d t
he
art
icle
be
low
ab
ou
t le
ad
ers
hip
in
bu
sin
ess a
nd
th
e q
ue
stio
ns o
n t
he
op
po
site
pa
ge
.
•F
or
ea
ch
qu
estio
n 1
3 –
18,
mark
one letter
(A,
B,
Cor
D)
on
yo
ur
An
sw
er
Sh
ee
t fo
r th
e a
nsw
er
yo
u c
ho
ose
.
Th
e E
ffe
cti
ve
L
ea
de
r
Fro
m w
orkp
lace
sur
veys
I h
ave
foun
d th
at m
ost
peop
le w
ant
to b
e –
and
feel
the
y co
uld
be –
mor
eef
fect
ive
lead
ers.
Cer
tain
ly t
hey
wan
t th
eir
lead
ers
tobe
mor
e ef
fect
ive.
But
wha
t do
we
mea
n by
eff
ecti
vele
ader
ship
in
busi
ness
? It
wou
ld a
ppea
r a
sim
ple
ques
tion
. U
nfor
tuna
tely
, ef
fect
iven
ess
is m
ore
easi
lyre
cogn
isab
le w
hen
it i
s ab
sent
. L
eade
rs w
ho a
ttem
ptto
use
bus
ines
s ja
rgon
and
try
out
the
lat
est
idea
s ar
eto
o of
ten
seen
as
fi
gure
s of
fu
n.
Whi
lst
peop
lefr
eque
ntly
agr
ee o
n w
hat
inef
fect
ive
lead
ersh
ip i
s,cl
earl
y kn
owin
g w
hat
not
to d
o is
har
dly
help
ful
inpr
acti
ce.
Hug
e am
ount
s of
res
earc
h ha
ve b
een
done
on
this
very
wid
e su
bjec
t. W
hen
you
look
at
lead
ersh
ip i
ndi
ffer
ent
way
s,
you
see
diff
eren
t th
ings
. W
hile
desc
ript
ions
of
lead
ersh
ip a
re a
ll d
iffe
rent
, the
y ar
e al
ltr
ue
– an
d th
is
is
whe
re
disa
gree
men
t ar
ises
.H
owev
er,
lead
ersh
ip i
s sp
ecif
ic t
o a
give
n co
ntex
t.T
he
effe
ctiv
enes
s of
yo
ur
acti
ons
is
asse
ssed
in
rela
tion
to
the
cont
ext
and
to t
he c
ondi
tion
s un
der
whi
ch y
ou t
ook
them
.F
or
a m
agaz
ine
arti
cle
I w
rote
re
cent
ly,
Iin
terv
iew
ed a
pub
lish
ing
exec
utiv
e, a
utho
r of
sev
eral
wel
l-kn
own
publ
icat
ions
, ab
out
wha
t ef
fect
ive
lead
ersh
ip is
. It w
as s
igni
fica
nt th
at, a
t fir
st, h
e di
d no
tm
enti
on h
is o
wn
com
pany
. He
talk
ed a
t le
ngth
abo
utw
hat
was
hap
peni
ng i
n th
e in
dust
ry –
the
mer
gers
,ta
keov
ers
and
glob
al n
atur
e of
the
busi
ness
. Bef
ore
hew
as a
ble
to d
escr
ibe
his
own
obje
ctiv
es f
or t
he n
ewpu
blis
hing
org
anis
atio
n he
was
set
ting
up,
he
had
tose
e a
clea
r fi
t be
twee
n th
ese
prop
osal
s an
d th
e la
rger
situ
atio
n ou
tsid
e. O
bvio
us?
Of
cour
se. B
ut I
hav
e lo
stco
unt
of t
he n
umbe
r of
lea
ders
I h
ave
coac
hed
who
beli
eved
tha
t th
eir
idea
s w
ere
vali
d w
hate
ver
the
situ
atio
n.A
t th
is
poin
t I
shou
ld
also
m
enti
on
anot
her
exam
ple,
tha
t of
a f
inan
ce d
irec
tor
who
se p
lan
ofac
tion
was
not
wel
l re
ceiv
ed.
The
com
pany
he
had
join
ed h
ad g
row
n st
eadi
ly f
or t
wen
ty y
ears
, se
rvin
g
clie
nts
who
w
ere
in
the
mai
n di
stru
stfu
l of
an
ypr
oduc
t th
at
was
to
o re
volu
tion
ary.
T
he
fina
nce
dire
ctor
saw
pot
enti
al c
hall
enge
s fr
om c
ompe
tito
rsan
d w
ante
d hi
s or
gani
sati
on t
o m
ove
wit
h th
e ti
mes
.U
nfor
tuna
tely
, m
ost
staf
f be
low
him
wer
e un
wil
ling
to
chan
ge.
I co
nclu
ded
that
al
thou
gh
ther
e w
ere
cert
ainl
y so
me
pers
onal
ski
lls
he c
ould
impr
ove
upon
,w
hat
he m
ost
need
ed t
o do
was
to
com
mun
icat
eef
fect
ivel
y w
ith
his
subo
rdin
ates
, so
tha
t th
ey a
ll f
elt
at e
ase
wit
h hi
s di
ffer
ent
appr
oach
.S
ome
effe
ctiv
e le
ader
s be
liev
e th
ey c
an c
ontr
olun
cert
aint
y be
caus
e th
ey k
now
wha
t th
e or
gani
sati
onsh
ould
be
do
ing
and
how
to
do
it
. W
ithi
n th
eor
gani
sati
on it
self
, exp
erti
se is
usu
ally
gre
atly
val
ued,
and
exec
utiv
es a
re e
xpec
ted,
as
they
ris
e w
ithi
n th
esy
stem
, to
kno
w m
ore
than
tho
se b
enea
th t
hem
and
,th
eref
ore,
to
man
age
the
oper
atio
n. A
good
exa
mpl
eof
this
wou
ld b
e a
firm
of
acco
unta
nts
I vi
site
d. T
heir
busi
ness
was
bui
lt o
n se
llin
g re
liab
le e
xper
tise
to
the
clie
nt,
who
na
tura
lly
wan
ts
unce
rtai
nty
to
beso
met
hing
onl
y ot
her
com
pani
es h
ave
to f
ace.
Wit
hin
this
fir
m, g
ivin
g th
e ri
ght
answ
er w
as g
reat
ly v
alue
d,an
d m
ista
kes
wer
e cl
earl
y to
be
avoi
ded.
I am
par
ticu
larl
y in
tere
sted
in
wha
t ai
ms
lead
ers
have
and
wha
t th
eir
role
sho
uld
be i
n he
lpin
g th
eor
gani
sati
on
to
achi
eve
its
stra
tegi
c ai
ms.
S
ome
lead
ers
are
high
ly in
effe
ctiv
e w
hen
the
aim
doe
sn’t
fit
wit
h th
e ne
ed,
such
as
the
man
ufac
turi
ng m
anag
erw
ho
was
en
cour
aged
by
he
r bo
sses
to
m
ake
revo
luti
onar
y ch
ange
s.
She
di
d,
and
was
ve
rysu
cces
sful
. H
owev
er,
whe
n sh
e m
oved
to
a di
ffer
ent
part
of
the
busi
ness
, she
car
ried
on
her
prog
ram
me
ofch
ange
. U
nfor
tuna
tely
, th
is p
art
of t
he b
usin
ess
had
alre
ady
suff
ered
ba
dly
from
tw
o m
ism
anag
edat
tem
pts
at c
hang
e. M
y po
int
is t
hat
wha
t he
r pe
ople
need
ed a
t tha
t mom
ent w
as a
ste
ady
hand
, not
fur
ther
chan
ges
– sh
e sh
ould
ha
ve
reco
gnis
ed
that
. T
heou
tcom
e w
as th
at w
ithi
n si
x m
onth
s st
aff
wer
e ca
llin
gfo
r he
r re
sign
atio
n.
Appendix 4.2
332
Exa
mp
le: A
exte
nsio
nB
expa
nsio
nC
accu
mul
atio
nD
infla
tion
19A
outp
utB
yiel
dC
tota
lD
mas
s
20A
budg
etB
inco
me
Cac
coun
tD
fund
21A
forw
ard
Btr
ansf
erC
adva
nce
Dde
liver
22A
repl
acem
ent
Bre
arra
ngin
gC
relo
catio
nD
redi
strib
utio
n
23A
aim
edB
imag
ined
Cac
cept
edD
expe
cted
24A
gues
sB
judg
emen
tC
estim
ate
Dco
nvic
tion
25A
focu
sB
obje
ctC
inte
ntio
nD
purp
ose
26A
conc
entr
atin
gB
plan
ning
Cat
tend
ing
Ddi
rect
ing
27A
runn
ing
Bim
plem
entin
gC
exec
utin
gD
orga
nisi
ng
28A
appo
int
Bre
sult
Cem
ploy
Dcr
eate
29A
scho
lars
hips
Bgr
ants
Cal
loca
tions
Dgi
fts
30A
finis
hB
clos
ure
Cen
ding
Dco
nclu
sion
31A
disp
atch
edB
prom
pted
Cla
unch
edD
effe
cted
32A
attit
ude
Bap
proa
chC
outlo
okD
feed
back
33A
prac
tice
Bpr
ogre
ssC
proc
ess
Dpr
oced
ure
AB
CD
0
PA
RT
FO
UR
Qu
esti
on
s 19
– 3
3
•R
ea
d t
he
extr
act
be
low
fro
m t
he
an
nu
al re
po
rt o
f a
co
mp
an
y w
ith
ma
nu
factu
rin
g in
tere
sts
aro
un
d
the
wo
rld
.
•C
ho
ose
th
e b
est
wo
rd t
o f
ill e
ach
ga
p f
rom
A,
B,
Cor
Do
n t
he
op
po
site
pa
ge
.
•F
or
ea
ch
qu
estio
n 1
9 –
33,
mark
one letter
(A,
B,
Cor
D)
on
yo
ur
An
sw
er
Sh
ee
t.
•T
he
re is a
n e
xa
mp
le a
t th
e b
eg
inn
ing
, (0
).
Manufa
ctu
ring
Str
ate
gy
During t
he last
year, w
e a
nnounced t
he s
ignific
ant (0
)…..B …
....
of
our
pla
stic s
heeting p
lant
in M
ala
ysia
, w
hic
h,
togeth
er
with t
he a
cquis
itio
n
of
the
Ja
va
ne
se
fa
cto
ry,
will
a
pp
roxim
ate
ly
do
ub
le
the
G
rou
p’s
manufa
ctu
ring
(19)
……
……
. T
he c
ost
of
this
develo
pm
ent
is w
ithin
(20)
……
……
and w
ill b
e a
ppro
xim
ate
ly $
5.6
m,
of
whic
h $
2.7
m w
as
incurr
ed d
uring t
he p
revio
us y
ear. I
t is
on s
chedule
to (
21)…
……
…in
cre
asin
g v
olu
mes f
rom
Octo
ber
2002.
Fo
llow
ing
th
e
(22)
……
……
of
pla
stic
tub
ing
m
an
ufa
ctu
re
fro
m
Germ
any t
o T
haila
nd,
we h
ave e
ffectively
double
d t
he c
apacity o
f th
is
facili
ty a
t an (2
3)…
……
…cost of $12m
. T
he p
roje
ct is
set to
cost le
ss
tha
n
the
o
rig
ina
l (2
4)…
……
…a
nd
is
o
n
targ
et
for
incre
ase
d
pro
duction b
y J
une 2
003.
In F
ebru
ary
, w
e a
nnounced o
ur
(25)
……
……
to s
ell
our
facto
ry i
n
Ire
lan
d.
Th
is
de
cis
ion
is
in
lin
e
with
th
e
Gro
up
’s
str
ate
gy
of
(26)
……
……
on o
ur
core
cate
gories o
f bra
nded p
roducts
.
In
June,
we
announced
investm
ent
in
a
new
sta
te-o
f-th
e-a
rt
UK
manufa
ctu
ring facili
ty for
specia
list pla
stic c
om
ponents
. T
his
facili
ty w
ill
be
(27)
……
……
by m
id-2
003 a
nd w
ill incre
ase t
he G
roup’s
capacity
to m
anufa
ctu
re p
roducts
effic
iently i
n-h
ouse.
At
the s
am
e t
ime i
t w
ill
(28)
……
……
about
200 n
ew
jobs i
n a
n a
rea o
f hig
h u
nem
plo
ym
ent.
The f
acto
ry is t
o c
ost
appro
xim
ate
ly $
24m
, to
ward
s w
hic
h g
overn
ment
(29)
……
……
of up to $
4m
are
already a
vaila
ble
. S
adly
, as p
art
of th
is
move,
we
announced
the (3
0)…
……
…of
our
Bla
ckburn
fa
cili
ty,
whic
h is d
ue t
o t
ake p
lace in t
he e
arly p
art
of
2003.
As p
art
of our
com
mitm
ent to
effective e
xte
rnal c
om
munic
ations w
ith a
ll
our
sta
kehold
ers
, in
Octo
ber
we (
31)…
……
…th
e c
orp
ora
te w
ebsite,
whic
h i
s n
ow
pro
vid
ing u
p-t
o-d
ate
info
rmation o
n t
he G
roup,
and w
e
look forw
ard
to r
eceiv
ing (3
2)…
……
…fr
om
users
of th
e s
ite. E
xis
ting
pro
duct w
ebsites a
re n
ow
in the (3
3)…
……
…of bein
g r
edesig
ned a
s
part
of
the g
lobal re
bra
ndin
g s
trate
gy.
Appendix 4.2
333
PA
RT
FIV
E
Qu
esti
on
s 34
– 4
5
•R
ea
d t
he
art
icle
be
low
ab
ou
t m
ark
et
rese
arc
h.
•In
mo
st
of
the
lin
es 3
4 –
45th
ere
is o
ne e
xtr
a w
ord
. It is e
ither
gra
mm
atically
incorr
ect
or
does
not
fit
in w
ith t
he m
eanin
g o
f th
e t
ext. S
om
e lin
es,
how
ever, a
re c
orr
ect.
•If a
lin
e is c
orr
ect, w
rite
CO
RR
EC
To
n y
ou
r A
nsw
er
Sh
ee
t.
•If
th
ere
is a
n e
xtr
a w
ord
in
th
e lin
e,
write
th
e ex
tra
wo
rdin
CA
PIT
AL
LE
TT
ER
S o
n y
ou
r A
nsw
er
Sh
ee
t.
•T
he
exe
rcis
e b
eg
ins w
ith
tw
o e
xa
mp
les,
(0)
and (
00).
Exa
mp
les:
0M
ark
et re
se
arc
h in
vo
lve
s in
co
llectin
g a
nd
so
rtin
g fa
cts
an
d o
pin
ion
s fro
m s
pe
cific
gro
up
s
00o
f p
eo
ple
. T
he
pu
rpo
se
of re
se
arc
h c
an
va
ry fro
m d
isco
ve
rin
g th
e p
op
ula
rity
of a
po
litic
al
34p
art
y t
o a
sse
ssin
g w
he
the
r is
a p
rod
uct
ne
ed
s c
ha
ng
ing
or
rep
lacin
g.
Mo
st
wo
rk i
n
35co
nsu
me
r re
se
arc
h i
nvo
lve
s i
nte
rvie
we
rs e
mp
loye
d b
y m
ark
et
rese
arc
h a
ge
ncie
s,
bu
t
36ce
rta
in in
du
str
ial a
nd
so
cia
l re
se
arc
h is
ca
rrie
d o
ut b
y a
ny s
pe
cia
list a
ge
ncie
s. In
terv
iew
s
37m
ay
be w
ith in
div
iduals
or
gro
ups
and c
an la
st a
nyt
hin
g a
s fr
om
a few
min
ute
s to
an h
our
or
38m
ore
. In
som
e i
nte
rvie
ws,
people
may b
e a
sked t
o e
xam
ine o
r tr
y o
ut
pro
ducts
befo
re
39giv
ing u
p their o
pin
ion. S
uccessfu
l inte
rvie
wers
tend to li
ke m
eeting p
eople
and s
hould
not
40o
nly
b
e
sh
y
of
ad
dre
ssin
g
str
an
ge
rs.
Inte
rvie
we
rs
are
u
su
ally
e
xp
ecte
d
to
wo
rk
41u
nsu
pe
rvis
ed
, o
rga
nis
ing
th
eir o
wn
wo
rklo
ad
. S
elf-d
iscip
line
is a
bso
lute
ly e
sse
ntia
l, a
nd
42a
s a
re m
otiva
tio
n a
nd
en
erg
y. T
he
re a
re n
o s
pe
cific
ag
e l
imits f
or
su
ch
a w
ork
, th
ou
gh
43m
an
y a
ge
ncie
s p
refe
r to
em
plo
y o
lde
r a
pp
lica
nts
with
exp
erie
nce
of
me
etin
g p
eo
ple
.
44M
ark
et re
searc
h a
gencie
s w
hic
h fre
quently o
rganis
e tra
inin
g, w
here
tra
inees le
arn
how
to
45re
cognis
e s
ocio
-econom
ic g
roups a
nd p
ractise a
ppro
achin
g to the p
ublic
. F
or
info
rmation
on
ma
rke
t re
se
arc
h tra
inin
g a
nd
qu
alif
ica
tio
ns, co
nta
ct th
e M
ark
et R
ese
arc
h A
sso
cia
tio
n.
0I
N
00C
OR
RE
CT
Ma
rke
t R
ese
arc
h
Appendix 4.2
334
TIME 45 minutes
INSTRUCTIONS TO CANDIDATES
Do not open this paper until you are told to do so.
Write your name, Centre number and candidate number in the spaces at the top of this page and on each
sheet of answer paper used.
Read the instructions carefully.
Answer both questions.
Write your answers on the separate answer paper provided.
Write clearly in pen, not pencil. You may make alterations but make sure that your work is easy to read.
If you use more than one sheet of paper, fasten the sheets together.
At the end of the examination hand in both this question paper and your answer paper.
INFORMATION FOR CANDIDATES
Part 2 carries twice as many marks as Part 1.
UNIVERSITY OF CAMBRIDGE LOCAL EXAMINATIONS SYNDICATE
Examinations in English as a Foreign Language
BUSINESS ENGLISH CERTIFICATE 0352/2Vantage
Test of Writing Test 023
Thursday 6 JUNE 2002 Morning 45 minutes
Additional materials:Answer Paper
Candidate
Centre Number Number
Candidate Name
This question paper consists of 3 printed pages.
SP (SLC/KS) S26977/2
Appendix 4.2
335
PA
RT
TW
O
•Y
ou a
re t
he M
ark
eting M
anager
for
a r
eta
il gro
up.
You h
ave b
een a
sked t
o w
rite
a s
hort
report
for
the
Bo
ard
of
Dire
cto
rs o
n t
he
co
mp
an
y’s
pe
rfo
rma
nce
be
twe
en
19
98
an
d 2
00
1.
•L
oo
k a
t th
e c
ha
rts b
elo
w,
on
wh
ich
yo
u h
ave
alre
ad
y m
ad
e s
om
e h
an
dw
ritt
en
no
tes.
•T
he
n,
usin
g a
llyo
ur
ha
nd
writt
en
no
tes,
write
yo
ur
rep
ort
.
•W
rite
12
0 –
14
0 w
ord
s.
•W
rite
on
th
e s
ep
ara
te a
nsw
er
pa
pe
r p
rovid
ed
.
–2–10123456
ele
ctr
ical
sto
res
book
sto
res
clo
thin
gsto
res
pharm
acie
s
Sale
s g
row
th (
2001)
4.2
5%
–1.2
5%
2.5
%
5.5
%
05
10
15
20
25
’01
’00
’99
’98
Ad
vert
isin
g
9%
% of total expenditure
17%
10%
23%
Pre
-tax p
rofi
ts (
£m
)
’98
’99
’00
’01
520
520
726
554
result
of h
igh a
dvert
ising
budg
et in
1999
and
low
inter
est ra
tes
hope t
o see
positi
veres
ults o
f this
in20
02
bigge
st gr
owth
!–
say w
hy
all to
be s
old
in 2
002
–say
why
but w
e stil
l hav
ebig
gest
share
ofma
rket
PA
RT
ON
E
•Y
ou a
re t
he m
anag
er o
f th
e C
usto
mer
Ser
vice
s D
epar
tmen
t. Y
ou a
re p
lann
ing
to in
trod
uce
a
new
com
pute
r sy
stem
in y
our
depa
rtm
ent.
•W
rite
an e
-mai
lto
all s
taff
in y
our
depa
rtm
ent:
•te
lling
the
m w
hen
the
new
sys
tem
will
be
intr
oduc
ed
•ex
plai
ning
wha
t th
e ad
vant
ages
of
the
new
sys
tem
will
be
•ad
visi
ng t
hem
who
to
cont
act
if th
ey h
ave
any
ques
tions
.
•W
rite
40
– 50
wo
rds.
•W
rite
on t
he s
epar
ate
answ
er p
aper
pro
vide
d.
Com
pute
r S
yste
m
All
Sta
ff
Appendix 4.2
336
TIME Approx. 40 minutes (including 10 minutes’ transfer time)
INSTRUCTIONS TO CANDIDATES
Do not open this paper until you are told to do so.
Write your name, Centre number and candidate number in the spaces at the top of this page. Write these
details in pencil on your Answer Sheet if these are not already printed.
Listen to the instructions for each part carefully.
Try to answer all the questions.
Write your answers on this question paper.
At the end of the test you will have 10 minutes to copy your answers onto your Answer Sheet.
Read the instructions for completing your Answer Sheet carefully.
Write all your answers in pencil.
At the end of the examination hand in both this question paper and your Answer Sheet.
INFORMATION FOR CANDIDATES
Instructions are given on the tape.
You will hear everything twice.
There are thirty questions on this paper.
UNIVERSITY OF CAMBRIDGE LOCAL EXAMINATIONS SYNDICATE
Examinations in English as a Foreign Language
BUSINESS ENGLISH CERTIFICATE 0352/3Vantage
Test of Listening Test 023
Thursday 6 JUNE 2002 Morning Approx. 40 minutes(including 10 minutes’
transfer time)
Additional materials:Answer Sheet
Candidate
Centre Number Number
Candidate Name
This question paper consists of 7 printed pages.
SP (CW) S22995/2
Appendix 4.2
337
Co
nve
rsat
ion
Tw
o
(Qu
esti
on
s 5
– 8)
•Lo
ok a
t th
e no
te b
elow
.
•Y
ou w
ill h
ear
a w
oman
cal
ling
abou
t a
job
appl
icat
ion.
MES
SAG
E
For:
Jill
Sara
(5)
......
......
......
......
......
......
......
......
......
......
......
.....
calle
d th
is m
orni
ng a
bout
the
post of
(6)...
......
......
......
......
......
......
......
......
......
......
......
.. .
They
’d li
ke y
ou t
o at
tend
a (
7)...
......
......
......
......
......
......
......
......
......
......
......
.. on
the
28th
; th
ey’ll c
onfi
rm t
his by
lett
er.
In t
he m
eant
ime,
can
you
send
her
deta
ils of
your
(8)...
......
......
......
......
......
......
......
......
......
......
......
.. .
PA
RT
ON
E
Qu
esti
on
s 1
– 12
•Y
ou w
ill h
ear
thre
e te
leph
one
conv
ersa
tions
or
mes
sage
s.
•W
rite
on
e o
r tw
o w
ord
s o
r a
nu
mb
erin
the
num
bere
d sp
aces
on
the
note
s or
for
ms
belo
w.
•Y
ou w
ill h
ear
each
rec
ordi
ng t
wic
e.
Co
nve
rsat
ion
On
e
(Qu
esti
on
s 1
– 4)
•Lo
ok a
t th
e no
te b
elow
.
•Y
ou w
ill h
ear
a m
an c
allin
g hi
s of
fice.
Tele
ph
on
e m
essa
ge
Mar
tin
Hay
es ph
oned
from
the
(1)
......
......
......
......
......
......
......
......
......
......
......
.....
.
Ther
e’s a
problem
: th
e (2
)...
......
......
......
......
......
......
......
......
......
......
......
.. ha
ven’
tar
rive
d! (
They
wer
e sen
t by
air la
st w
eek.)
Ano
ther
thi
ng: he
nee
ds m
ore
(3)...
......
......
......
......
......
......
......
......
......
......
......
.. .
He’s
att
endi
ng a
(4)
......
......
......
......
......
......
......
......
......
......
......
.....
this
morni
ng, so
call
him
bac
k ar
ound
lunc
htim
e.
Appendix 4.2
338
PA
RT
TW
O
Qu
esti
on
s 13
– 2
2
Sec
tio
n O
ne
(Qu
esti
on
s 13
– 1
7)
•Y
ou w
ill h
ear
five
shor
t re
cord
ings
.
•F
or e
ach
reco
rdin
g, d
ecid
e w
hich
asp
ect
of w
orki
ng c
ondi
tions
the
spe
aker
is t
alki
ng a
bout
.
•W
rite
one
lette
r (A
– H
) ne
xt t
o th
e nu
mbe
r of
the
rec
ordi
ng.
•D
o no
t us
e an
y le
tter
mor
e th
an o
nce.
•Y
ou w
ill h
ear
the
five
reco
rdin
gs t
wic
e.
13...
......
......
......
......
......
......
......
......
......
14...
......
......
......
......
......
......
......
......
......
15...
......
......
......
......
......
......
......
......
......
16...
......
......
......
......
......
......
......
......
......
17...
......
......
......
......
......
......
......
......
......
Sec
tio
n T
wo
(Qu
esti
on
s 18
– 2
2)
•Y
ou w
ill h
ear
anot
her
five
reco
rdin
gs.
•F
or e
ach
reco
rdin
g, d
ecid
e w
hat
each
spe
aker
is t
ryin
g to
do.
•W
rite
one
lette
r (A
– H
) ne
xt t
o th
e nu
mbe
r of
the
rec
ordi
ng.
•D
o no
t us
e an
y le
tter
mor
e th
an o
nce.
•Y
ou w
ill h
ear
the
five
reco
rdin
gs t
wic
e.
18...
......
......
......
......
......
......
......
......
......
19...
......
......
......
......
......
......
......
......
......
20...
......
......
......
......
......
......
......
......
......
21...
......
......
......
......
......
......
......
......
......
22...
......
......
......
......
......
......
......
......
......
Aca
reer
pro
spec
ts
Bhe
alth
and
saf
ety
Cw
orki
ng h
ours
Dho
liday
allo
wan
ce
Etr
aini
ng c
ours
es
Fdi
scip
linar
y pr
oced
ures
Gjo
b se
curit
y
Hpa
y in
crea
ses
Ano
min
ate
a su
pplie
r
Bpr
esen
t sa
les
figur
es
Csu
ppor
t a
prop
osal
Dre
fuse
an
incr
emen
t
Eag
ree
to e
xpen
ditu
re
Fcl
aim
dam
ages
Gne
gotia
te a
con
trac
t
Hre
ques
t a
post
pone
men
t
Co
nve
rsat
ion
Th
ree
(Qu
esti
on
s 9
– 12
)
•Lo
ok a
t th
e no
te b
elow
.
•Y
ou w
ill h
ear
a m
an p
honi
ng a
bout
som
e ar
rang
emen
ts f
or a
mee
ting.
WH
ILE
YO
U W
ER
E O
UT
Mes
sag
e fo
r:La
uren
O’N
eil
Fro
m:
Chris Da
rcy
Mes
sag
e
Chris (H
R) p
hone
d ab
out
mee
ting
of
(9)...
......
......
......
......
......
......
......
......
......
......
......
..
next
wee
k. T
here
’s go
ing
to b
e an
noun
cem
ent
abou
t
(10)
......
......
......
......
......
......
......
......
......
......
......
.....
. MD
want
s yo
u to
mak
e
prese
ntat
ion
on (
11)...
......
......
......
......
......
......
......
......
......
......
......
.. of t
he n
ew c
ompa
ny.
Coul
d yo
u al
so co
ver
(12)
......
......
......
......
......
......
......
......
......
......
......
.....
in y
our
prese
ntat
ion?
Appendix 4.2
339
28H
e th
inks
he
was
app
oint
ed c
hairm
an o
f LB
I be
caus
e th
e co
mpa
ny
Akn
ew o
f hi
s su
cces
ses
with
fai
ling
com
pani
es.
Bfe
lt he
had
a p
ositi
ve im
age
with
the
pub
lic.
Clik
ed h
is f
earle
ss a
ppro
ach
to p
robl
em-s
olvi
ng.
29A
ccor
ding
to
Phi
lip S
penc
er,
succ
essf
ul m
anag
ers
are
dist
ingu
ishe
d by
the
ir
Aco
ncer
n fo
r de
tail.
Bde
sire
to
mak
e m
oney
.
Cst
rong
lead
ersh
ip.
30H
is f
inal
adv
ice
to p
eopl
e st
artin
g in
bus
ines
s is
to
Am
ake
ever
y ef
fort
to
prev
ent
mis
take
s.
Bfin
d th
e be
st s
ourc
es o
f in
form
atio
n.
Cm
aint
ain
a po
sitiv
e at
titud
e at
all
times
.
Yo
u n
ow
hav
e 10
min
ute
s to
tra
nsf
er y
ou
r an
swer
s to
yo
ur
An
swer
Sh
eet.
PA
RT
TH
RE
E
Qu
esti
on
s 23
– 3
0
•Y
ou w
ill h
ear
a ra
dio
inte
rvie
w w
ith a
lead
ing
indu
stria
list
and
busi
ness
con
sulta
nt,
Phi
lip
Spe
ncer
.
•F
or e
ach
ques
tion
23 –
30,
mar
k on
ele
tter
(A,
Bor
C)
for
the
corr
ect
answ
er.
•Y
ou w
ill h
ear
the
reco
rdin
g tw
ice.
23W
hen
visi
ting
com
pani
es,
Phi
lip S
penc
er’s
obj
ectiv
e is
to
Aim
prov
e st
aff
prod
uctiv
ity.
Bid
entif
y pr
oble
m a
reas
.
Cre
trai
n w
eak
man
agem
ent.
24P
robl
ems
at M
anso
n’s
had
cont
inue
d af
ter
Spe
ncer
’s f
irst
visi
t be
caus
e of
Apo
or d
istr
ibut
ion
syst
ems.
Bin
adeq
uate
mar
ket
rese
arch
.
Cou
tdat
ed p
rodu
ctio
n m
etho
ds.
25D
iffic
ultie
s at
Crit
erio
n G
lass
ste
mm
ed f
rom
lack
of
atte
ntio
n to
Aco
mpe
titor
s’de
sign
s.
Bqu
ality
of
mer
chan
dise
.
Cco
nsum
er d
eman
d.
26P
hilip
Spe
ncer
bla
mes
his
ear
ly b
usin
ess
diffi
culti
es o
n
Ain
expe
rienc
e w
ith n
ew c
ompa
nies
.
Bla
ck o
f kn
owle
dge
of t
he f
inan
cial
sec
tor.
Cba
d ad
vice
fro
m e
stab
lishe
d or
gani
satio
ns.
27H
e de
fend
s hi
s un
usua
l per
sona
l sty
le b
y sa
ying
tha
t
Ait
is im
port
ant
in b
usin
ess
to m
ake
a st
rong
impr
essi
on.
Bhi
s bu
sine
ss id
eas
are
mor
e im
port
ant
than
his
app
eara
nce.
Cm
ost
busi
ness
peo
ple
are
too
serio
us a
nd t
radi
tiona
l.
Appendix 4.2
340
Appendix 4.2
341
Appendix 4.2
342
Appendix 4.2
343
PA
RT
2 :
Min
i-p
res
en
tati
on
s f
or
2 c
an
did
ate
s (
ab
ou
t 6
min
ute
s)
[Inte
rlocuto
rhands e
ach c
andid
ate
a d
iffe
rent
topic
card
,and s
om
e p
aper
and a
pencil
for
note
s.]
[Allo
w1
min
ute
's p
repara
tion t
ime.
Both
candid
ate
s p
repare
their
talk
s a
tth
e s
am
e t
ime,
separa
tely
.]
[Allo
w45 s
econds t
o1
min
ute
. O
nly
use b
ack-u
p q
uestions if
the c
andid
ate
is u
nable
to
speak w
ithout
pro
mpting.]
[Allo
w10 s
econds.
Do n
ot
use b
ack-u
p q
uestions.]
[Allo
w45 s
econds t
o1
min
ute
. O
nly
use b
ack-u
p q
uestions if
the c
andid
ate
is u
nable
to
speak w
ithout
pro
mpting.]
[Allo
w10 s
econds.
Do n
ot
use b
ack-u
p q
uestions.]
[Re
trie
ve
ma
teri
als
.]
*US
E C
AN
DID
AT
ES
’N
AM
ES
TH
RO
UG
HO
UT
TH
E T
ES
T
I'm
goin
gto
giv
e e
ach o
fyou a
choic
e o
f3
topic
s.
I'd
like
you t
ochoose o
ne
of
the t
opic
s a
nd g
ive
a s
hort
pre
senta
tion o
n it
for
about
am
inute
.Y
ou w
ill h
ave a
bout
am
inute
to
pre
pare
for
this
and
you c
an m
ake
note
s if
you w
ish w
hile
you p
repare
. A
fter
you
have
finis
hed y
our
talk
,your
part
ner
will
ask
you a
question.
All
rig
ht?
H
ere
are
your
topic
s.
Choose o
ne
of
the t
opic
s t
ota
lkabout.
Y
ou c
an m
ake
note
s.
All
rig
ht.
N
ow
, *B
,w
hic
h t
opic
have y
ou c
hosen,
A,
Bor
C? [
candid
ate
answ
ers
]W
ould
you lik
eto
talk
about
what
you t
hin
kis
im
port
ant
when (
inte
rlocuto
rsta
tes c
andid
ate
’s c
hosen t
opic
).
Thank
you.
Now
, *A
,ple
ase a
sk
*Byour
question a
bout
his
/her
talk
.
Thank
you.
All
rig
ht.
N
ow
, *A
,w
hic
h t
opic
have y
ou c
hosen,
A,
Bor
C? [
candid
ate
answ
ers
]W
ould
you lik
e t
ota
lkabout
what
you t
hin
kis
im
port
ant
when (
inte
rlocuto
rsta
tes c
andid
ate
’schosen t
opic
).
Thank
you.
Now
, *B
,ple
ase a
sk
*Ayour
question a
bout
his
/her
talk
.
Thank
you.
That's t
he
end o
fth
e f
irst
part
of
the t
est.
In
the
next
part
you
are
each g
oin
gto
giv
e a
short
pre
senta
tion.
Thank
you.
Appendix 4.2
344
Task C
ard
17 –
Ex
am
ine
r’s C
op
y
Whatty
pes
of
acti
vit
ies
are
import
antto
consid
er?
(W
hy?)
Is it e
sse
ntialto
co
nsid
er
co
st?
(W
hy/W
hy
not?
)
Sele
ct fr
om
the
fo
llow
ing a
dd
itio
na
l pro
mpts
(if th
e a
bove h
ave a
lrea
dy b
ee
n c
overe
d):
How
import
ant
is t
he
ven
ue? (W
hy/W
hy
not?
)Is
it
import
ant
whic
hco
mp
an
yp
ers
on
nel
are
involv
ed in e
nte
rtain
ing c
lients
?(W
hy/W
hy
not?
)
Is lo
cati
on
the m
ost
import
ant
thin
gto
consid
er?
(W
hy/W
hy
not?
)W
hy
is t
he
len
gth
of
the
co
ntr
ac
tim
port
ant?
Sele
ct fr
om
the
fo
llow
ing a
dd
itio
na
l pro
mpts
(if th
e a
bove h
ave a
lrea
dy b
ee
n c
overe
d):
How
import
ant
is c
ost?
(W
hy/W
hy n
ot?
)H
ow
import
ant
is it
toconsid
er
the
co
nd
itio
nof
the
pre
mis
es?(W
hy/W
hy
not?
)
Why
isim
ag
e im
port
ant
?H
ow
import
ant
is it
toconsid
er
the
pro
du
cti
on
pro
cess? (
Why?)
Sele
ct fr
om
the
fo
llow
ing a
dd
itio
na
l pro
mpts
(if th
e a
bove h
ave a
lrea
dy b
ee
n c
overe
d):
How
import
ant
is t
he
co
st
ofth
e p
ackag
ing
? (
Why/W
hy
not?
)Is
it e
sse
ntialto
ca
rry o
ut
mark
et
researc
hbefo
re d
ecid
ing
on the
packag
ing
? (
Why/W
hy
not?
)
A:
WH
AT
IS
IMP
OR
TA
NT
WH
EN
…?
EN
TE
RT
AIN
ING
CL
IEN
TS
TY
PE
S O
FA
CT
IVIT
IES
CO
ST
B:
WH
AT
IS
IMP
OR
TA
NT
WH
EN
…?
CH
OO
SIN
G R
ET
AIL
PR
EM
ISE
ST
O R
EN
T
LO
CA
TIO
N
LE
NG
TH
OF
CO
NT
RA
CT
C:
WH
AT
IS IM
PO
RT
AN
T W
HE
N…
?
DE
CID
ING
ON
PA
CK
AG
ING
FO
R P
RO
DU
CT
S
IMA
GE
PR
OD
UC
TIO
N P
RO
CE
SS
Appendix 4.2
345
Task C
ard
20 –
Ex
am
ine
r’s C
op
y
Is o
fferi
ng
an
ap
olo
gy t
he
most
import
ant
thin
g? (W
hy/W
hy n
ot?
)Is
it e
ssential to
su
gg
est
aso
luti
on
to the
pro
ble
m? (
Why/W
hy
not?
)
Sele
ct fr
om
the
fo
llow
ing a
dd
itio
na
l pro
mpts
(if th
e a
bove h
ave a
lrea
dy b
ee
n c
overe
d):
How
import
ant is
it t
o o
ffer
co
mp
en
sati
on
? (
Why/W
hy
not?
)W
hy
is it
import
ant
to investig
ate
the
cau
se
of
the
pro
ble
m?
Are
pro
du
cti
on
co
sts
the
most
import
antth
ing t
oconsid
er?
(W
hy/W
hy
not?
)W
hy
isit
essential to
consid
er
co
mp
eti
tors
’p
rices?
Sele
ct fr
om
the
fo
llow
ing a
dd
itio
na
l pro
mpts
(if th
e a
bove h
ave a
lrea
dy b
ee
n c
overe
d):
How
import
ant
is it
toconsid
er
avera
ge s
pen
din
g levels
of
the
targ
etm
ark
ets
? (W
hy?)
Why
isit
import
ant
toconsid
er
pro
du
ct
imag
e?
Are
fin
an
cia
lin
cen
tives
the
most
import
ant th
ing
toconsid
er?
(W
hy/W
hy
not?
)H
ow
import
ant
is it
tohave a
care
er
str
uctu
re f
or
em
plo
yees?
(Why?)
Sele
ct fr
om
the
fo
llow
ing a
dd
itio
na
l pro
mpts
(if th
e a
bove h
ave a
lrea
dy b
ee
n c
overe
d):
How
import
ant
is o
rganis
ing
so
cia
l even
tsfo
r sta
ff? (W
hy/W
hy n
ot?
)Is
it im
port
antto
invo
lve s
taff
in d
ecis
ion m
akin
gin
the
com
pany? (
Why/W
hy
not?
)
B:
WH
AT
IS
IMP
OR
TA
NT
WH
EN
…?
SE
TT
ING
PR
ICE
S F
OR
NE
WP
RO
DU
CT
S
PR
OD
UC
TIO
N C
OS
TS
CO
MP
ET
ITO
RS
’ P
RIC
ES
C:
WH
AT
IS IM
PO
RT
AN
T W
HE
N…
?
AIM
ING
TO
RE
DU
CE
ST
AF
F T
UR
NO
VE
R
FIN
AN
CIA
L I
NC
EN
TIV
ES
CA
RE
ER
ST
RU
CT
UR
E
A:
WH
AT
IS
IMP
OR
TA
NT
WH
EN
…?
DE
AL
ING
WIT
H C
OM
PL
AIN
TS
FR
OM
CL
IEN
TS
OF
FE
RIN
G A
NA
PO
LO
GY
SU
GG
ES
TIN
GA
SO
LU
TIO
N T
OT
HE
PR
OB
LE
M
Task C
ard
18 –
Ex
am
ine
r’s C
op
y
Isth
e e
mplo
yee’s
att
itu
de
tow
ork
the m
ost
import
ant
thin
gto
consid
er?
(W
hy/W
hy
not?
)Is
it e
ssential to
consid
er
an e
mplo
yee’s
cu
rren
t p
erf
orm
an
ce? (
Why/W
hy
not?
)
Sele
ct fr
om
the
fo
llow
ing a
dd
itio
na
l pro
mpts
(if th
e a
bove h
ave a
lrea
dy b
ee
n c
overe
d):
How
import
ant
is it
toconsid
er
the
em
plo
yee’s
am
bit
ion
? (
Why/W
hy
not?
)H
ow
import
ant
is it fo
rth
e c
andid
ate
to h
ave a
pp
rop
riate
skills
for
the
new
post?
(W
hy?)
Why
isit
import
ant
toconsid
er
furt
her
stu
dy
or
train
ing
?Is
it im
port
antto
consid
er
op
po
rtu
nit
ies
for
furt
her
pro
mo
tio
n? (
Why/W
hy
not?
)
Sele
ct fr
om
the
fo
llow
ing a
dd
itio
na
l pro
mpts
(if th
e a
bove h
ave a
lrea
dy b
ee
n c
overe
d):
Is it e
sse
ntialto
co
nsid
er
fin
an
cia
lre
ward
s? (
Why/W
hy
not?
)H
ow
import
ant
are
flexib
le w
ork
ing
arr
an
gem
en
tsw
hen c
onsid
eri
ng a
care
er
chang
e?
(Why/W
hy
not?
)
How
import
ant
is it
tocarr
yout
mark
et
researc
h? (W
hy/W
hy
not?
)Is
sele
cti
ng
the a
pp
rop
riate
med
iath
e m
ost
import
ant
thin
g? (
Why/W
hy
not?
)
Sele
ct fr
om
the
fo
llow
ing a
dd
itio
na
l pro
mpts
(if th
e a
bove h
ave a
lrea
dy b
ee
n c
overe
d):
How
import
ant
is it
tob
ud
get
eff
ecti
vely
? (
Why?)
Isit
essentialfo
rth
eadvert
isin
gcam
paig
nto
su
pp
ort
the
imag
eofth
epro
duct?
(Why/W
hy
not?
)
B:
WH
AT
IS
IMP
OR
TA
NT
WH
EN
…?
CO
NS
IDE
RIN
G A
CA
RE
ER
CH
AN
GE
FU
RT
HE
R S
TU
DY
OR
TR
AIN
ING
OP
PO
RT
UN
ITIE
S F
OR
FU
TU
RE
PR
OM
OT
ION
A:
WH
AT
IS
IMP
OR
TA
NT
WH
EN
…?
SE
LE
CT
ING
ST
AF
FF
OR
PR
OM
OT
ION
AT
TIT
UD
E T
OW
OR
K
CU
RR
EN
T P
ER
FO
RM
AN
CE
C:
WH
AT
IS IM
PO
RT
AN
T W
HE
N…
?
PL
AN
NIN
G A
NA
DV
ER
TIS
ING
CA
MP
AIG
N
MA
RK
ET
RE
SE
AR
CH
SE
LE
CT
ING
AP
PR
OP
RIA
TE
ME
DIA
Appendix 4.2
346
PA
RT
3 :
Co
lla
bo
rati
ve
ta
sk
an
d d
isc
us
sio
n f
or
2 c
an
did
ate
s (
ab
ou
t 5
min
ute
s)
[Poin
t to
the
card
show
ing t
he t
ask w
hile
giv
ing t
he instr
uctions b
elo
w.]
[Pla
ce t
he c
ard
in f
ront
of
the c
andid
ate
s s
o t
hat
they c
an b
oth
see it.
]
[If
ne
ce
ssa
ry,
giv
e c
larifica
tio
n.
Th
en
allo
w 3
0 s
eco
nd
s f
or
ca
nd
ida
tes t
o a
bso
rb t
he
in
form
atio
n a
nd
to
thin
k h
ow
tobegin
.]
[Aft
er
the c
andid
ate
s h
ave f
inis
hed s
peakin
g,
the inte
rlocuto
r asks q
uestions a
nd f
inis
hes t
he s
peakin
gte
st
as d
irecte
d o
n t
he e
xam
iner’s
copy o
fth
e t
ask c
ard
.]
Now
, in
this
part
of
the
test
you
are
goin
g t
o d
iscuss s
om
eth
ing
tog
eth
er.
You h
ave a
bout
30 s
econds t
ore
ad t
his
task
care
fully
,and t
hen
about
3m
inute
s t
odis
cuss a
nd
decid
e a
bout
it t
og
eth
er.
Y
ou s
hould
giv
ere
asons f
or
your
decis
ions a
nd o
pin
ions.
You d
on't
need
to w
rite
anyth
ing
. I
s t
hat
cle
ar?
I'm
just
goin
gto
liste
n a
nd t
hen a
sk
you t
osto
p a
fter
about
3m
inute
s.
Ple
ase s
peak
so t
hat
we c
an
hear
you.
Appendix 4.2
347
Task 3
3 –
Exam
iner’
sC
op
y
Fo
r th
ree
ca
nd
ida
tes
[Ret
riev
e m
ate
rials
.]
Work
Exp
erie
nce P
rogra
mm
e
The
man
ufa
cturi
ng
com
pan
yyou w
ork
for
has
dec
ided
to o
ffer
a tw
o-
wee
k w
ork
ex
per
ience
pro
gra
mm
e fo
r a
smal
l gro
up o
f st
uden
ts f
rom
a
loca
l busi
nes
s co
lleg
e.
You h
ave
been
ask
ed t
o h
elp w
ith t
he
pre
par
atio
ns
for
this
pro
gra
mm
e.
Dis
cuss
the
situ
atio
n t
oget
her
, an
d d
eci
de:
what
kin
ds
of
work
ex
per
ien
ce t
he
com
pan
yco
uld
off
er
how
the
par
tici
pan
ts s
hould
be
sele
cted
what
fee
db
ack a
nd e
val
uat
ion s
hould
tak
e pla
ce a
fter
the
pro
gra
mm
e has
fin
ished
Than
k y
ou. T
hat
is
the
end o
f th
e sp
eakin
gte
st.
Inte
rlo
cuto
r: [
Sel
ect
one
or
more
of
the
foll
ow
ing q
ues
tions
as
appro
pri
ate
, to
red
ress
any
imbala
nce
or
to b
roaden
the
dis
cuss
ion.]
Wh
atoth
er p
rep
arati
on
s w
ould
the
com
pan
y n
eed
to m
ake
bef
ore
rece
ivin
g w
ork
ex
per
ience
stu
den
ts? (W
hy?)
What
are
the
ad
va
nta
ges
to a
com
pan
y o
f off
erin
ga
wo
rk e
xper
ience
pro
gra
mm
e to
busi
nes
s st
uden
ts?
What
do y
ou t
hin
k i
s th
e m
ost
use
ful
kin
d o
f w
ork
ex
per
ience
fo
rbusi
nes
s st
uden
ts? (W
hy?)
Wh
ath
elp
would
you g
ive
a st
uden
t on t
hei
r fi
rst
day
of
work
exper
ience
?(W
hy?)
Whic
h a
reas
of
busi
nes
s w
ould
you
lik
e to
hav
e m
ore
ex
per
ien
ce o
f?(W
hy?)
Inw
hat
ways
can b
usi
nes
ses
dev
elop c
lose
lin
ks
wit
h t
he
co
mm
un
ity
?
Appendix 4.2
348
Wri
tin
g S
am
ple
Sc
rip
ts
Part
1
Scri
pt
A
To:
All
Sta
ff
c.c.
Sub
ject
:C
ompu
ter
Sys
tem
I w
ould
lik
e t
oin
form
you
abou
t ne
w c
ompu
ter
syst
em
whic
h is
schedul
ed t
o
prov
ide
in t
he
begi
nnin
g of
next
mon
th.
This
new
syst
em
is
inst
alle
d a
fle
xib
le p
rogr
amm
ew
hic
h is
able
to
obta
in
info
rmat
ion
abou
t va
riou
s cu
stom
ers’
need
s.
Ifyo
uhav
ean
y qu
est
ions
, pl
eas
eco
ntac
t to
Mr.
Sm
ith,IT
Sys
tem
sM
anag
er.
Scri
pt
A
All
conte
nt
po
ints
are
achie
ved t
hou
gh p
oin
t 2
(expla
natio
n o
fth
e a
dva
nta
ges o
fth
e n
ew
syste
m)
isslig
htly
uncle
ar.
There
is a
n a
dequ
ate
ra
ng
e o
f la
ngu
age t
ho
ugh s
om
e a
wkw
ard
ness.
Ba
nd
3
Scri
pt
B
Dea
rS
taff
To
all st
aff
in C
usto
mer
Ser
vice
s D
epar
tmen
t w
e w
ill int
roduc
ing
new
com
pute
rsy
stem
in
our
dep
artm
ent
on 2
Jul
y 2
00
2.
The
new
syst
em.e
xel
ii w
ill h
elp
usto
calc
olat
eal
l th
efi
gars
by
givi
ng t
he
very
lett
le inf
orm
atio
n an
d b
y th
is n
ewsy
stem
we
cons
umin
g ou
rti
me.
Plea
se d
on’t
has
itat
e if
you
hav
e qu
acti
on o
r in
form
atio
n an
d a
dvi
ce c
all M
ary
Bla
ck o
n 0
20
78
85
92
07
/
All
the
bes
t
Scri
pt
B
This
answ
er
attem
pts
to a
ddre
ss
all
the c
on
tent
po
ints
but
is s
eriously
aff
ecte
d b
ya lack
ofcontr
ol
over
gra
mm
ar
and v
ocab
ula
ryexhib
itin
g f
requ
ent
basic
err
ors
. T
hese e
rrors
obscure
the m
eanin
g a
nd
have a
very
nega
tive e
ffect on the t
arg
et re
ader.
This
answ
er
isover
the
pre
scri
be
d leng
th f
or
this
part
of
the test b
ut
was n
ot
pe
nalis
ed d
irectly
for
this
.B
an
d 1
An
sw
er
Ke
ys
Read
ing
Part
1
1C
2D
3A
4B
5C
6D
7A
Part
2
8E
9C
10
F11
D12
B
Part
3
13
C14
A15
D16
A17
B18
C
Part
4
19
A20
A21
D22
C23
D24
C25
C26
A27
A28
D29
B30
B31
C32
D33
C
Part
5
34
is35 C
OR
RE
CT
36
an
y37 as
38 C
OR
RE
CT
39 u
p40 o
nly
41 a
nd
42 a
43 C
OR
RE
CT
44
whic
h45 to
Lis
ten
ing
Part
1
1(r
eta
il)exh
ibitio
n2 s
tands
3price
(-)
lists
4 (
pre
ss)
confe
rence
5 M
idd
lem
iss
/Mid
dle
Mis
s6 S
ale
sE
xec(u
tive)
7 (
com
pan
y)
pre
senta
tion
8 r
efe
rees
9 (
the)
share
ho
lders
10 (
pla
nned)
merg
er
11 (
com
bin
ed)
sale
s12 e
xp
ecte
d s
avin
gs
Part
2
13
E14
B15
H16
F17
C
18
D19
H20
C21
A22
F
Part
3
23
B24
C25
C26
B27
B28
C29
A30
C
Appendix 4.2
349
Scri
pt
D
All
conte
nt
po
ints
are
ad
dre
ssed a
nd t
here
is g
ood o
rgan
isa
tion t
hou
gh c
ohesio
n a
nd lin
kin
g is
som
etim
es p
oor.
The r
egis
ter
and f
orm
at is
reasonab
le a
nd
there
is
evid
ence o
f ade
qu
ate
ra
nge
.T
here
are
som
e e
rrors
and a
wkw
ard
ness
of
phra
sin
g, e.g
. ‘g
oin
g to s
end w
ith th
is g
rou
p’, ‘is
be
ing
specia
lised in t
his
kin
d o
f sto
res’. O
vera
ll th
e targ
et re
ader
wou
ldbe info
rmed.
Ba
nd
3
Scri
pt
E
Repo
rt o
n: “
FCB” pe
rfor
manc
e b
etw
een
1998 a
nd 2
001
IN
TR
OD
UC
TIO
N
This
repo
rt a
ims
to s
how
the s
itua
tion
of
our
com
pany
abou
t pr
e-t
ax p
rofi
ts,
adve
rtis
ing
and s
ales
grow
th b
etw
een
199
8 a
nd 2
00
1.
FINDINGS
In 1
99
8 a
nd 1
99
9 p
re-t
ax p
rofi
ts r
emai
ned s
table
at
£m
52
0, th
en
they
shar
ply
rose
in
20
00
than
ks t
o hig
h a
dve
rtis
ing
and low
int
ere
st r
ates
in 1
99
9.
Unf
ortu
nate
ly p
rofi
ts f
ell a
t £
m5
54
las
t ye
ar.
Adve
rtis
ing
situ
atio
n has
flu
ctua
ted d
urin
g th
e f
our
year
s an
d I
hop
e t
he
com
pany
will hav
e p
osit
ive r
esu
lts
in 2
00
2.
Sal
es
of c
loth
ing
stor
es h
ad a
lit
tle d
ecr
eas
e las
t ye
ar, but
F.C
.B s
till h
as t
he
big
gest
shar
e o
f m
arke
t. P
har
mac
ies
perf
orm
ance
was
goo
d a
nd e
lect
rica
l st
ores
regi
ster
ed a
big
gro
wth
in
sale
s due
to
aneff
icie
nt a
dve
rtis
ement
.
I hav
e d
eci
ded t
o se
ll a
ll t
he b
ook
stor
esin
20
02
to
sell a
ll t
he b
ook
stor
es
in
20
02
to
conc
ent
rate
the c
ompa
ny e
ffor
ts o
n th
e c
loth
ing
ones.
CO
NC
LU
SIO
N
It is
conc
luded t
hat
the s
itua
tion
of
our
com
pany
is
not
bad
, an
d F
CB
will be a
ble
to a
chie
ve p
osit
ive r
esu
lts
in 2
00
2.
RE
CO
MM
EN
DA
TIO
N
It is
reco
mm
end
ed t
hat
this
year
FC
B s
hou
ld inc
reas
e pr
e-ta
x p
rofi
t an
d
clot
hin
g st
ores
sale
s an
d t
ry t
o re
duc
e a
dve
rtis
ing
cost
s.
Scri
pt
E
The a
nsw
er
iseff
ective
lyorg
anis
ed
with g
oo
d u
se o
f cohesiv
e d
evic
es a
nd h
ead
ings a
nd
on
lym
inor
err
ors
. U
nfo
rtunate
ly p
oin
t 2, re
gard
ing
the
expecte
d p
ositiv
eim
pact of
the e
xtr
a s
pe
nd
ing o
nadvert
isin
g in 2
00
1, is
not
ade
qua
tely
addre
ssed
, an
d s
o t
he t
arg
et re
ad
er
wou
ldnot
be f
ully
info
rmed.
This
keeps this
answ
er
in b
and 2
.
Ba
nd
2
Scri
pt
C
I am
wri
ting
to
info
rm y
ou t
hat
a n
ewco
mpu
ter
syst
emw
ill b
e in
trod
uced
as
from
Jun
e 13
th. I
tw
ill e
nable
all
staf
f to
acc
ess
on-l
ine
cour
ses
avai
lable
on
our
intr
anet
.F
oran
y qu
esti
ons,
plea
seco
ntac
t M
rsB
row
n,E
DP
Dep
artm
ent.
Bes
t re
gard
s.
Scri
pt
C
A c
oncis
e a
nd v
ery
eff
ective a
nsw
er
wh
ich c
overs
all
con
tent p
oin
ts c
learl
y a
nd e
ffectively
. It
has a
very
positiv
e e
ffect on the targ
et re
ad
er.
Ba
nd
5
Part
2
Scri
pt
D
As
mar
keti
ng m
anag
erI
hav
eto
info
rmab
out
our
com
pany
’spe
rfor
man
cesi
nce
198
8 t
o 2
00
1, a
nd b
ecau
se o
f th
isI
hav
e to
dea
l w
ith t
he
pre-
tax p
rofi
ts,
adve
rtis
ing
and s
ales
gro
wth
.*1
. Pr
e-ta
x p
rofi
ts (
£m
)In
19
98
, ou
rpr
ofit
s w
ere
£5
20
m, i
n 19
99
£5
20
m, in
20
00
£7
26
m a
nd in
20
01
£5
59
m. W
e ca
n ap
prec
iate
that
the
year
wit
h t
he
big
gest
pro
fit
was
in
20
00
;th
is is
the
resu
lt o
f ah
igh a
dve
rtis
ing
bud
get
in 1
99
9 a
nd low
inte
rest
rat
es. T
he
rest
of
the
year
s hav
e bee
n lo
w,s
o I
thin
kw
e hav
e to
do
the
sam
e as
in
199
9n
next
year
.*2
.A
dve
rtis
ing
In 1
99
8 w
as a
9%
of
tota
l ex
pend
itur
e, in
199
9 a
17
%, 2
00
0 1
0%
and
in
20
01
a2
3%
. This
las
t on
e has
bee
n th
e hig
hes
t pe
r ce
nt o
f to
tal ex
pend
itur
e an
d I
hop
e to
see
ther
e po
siti
ve r
esul
ts in
20
02
and
the
follo
win
g ye
ars.
* 3
We
hav
e to
dea
l w
ith d
iffe
rent
iss
ues,
as
wag
es, th
eir
grow
th w
as a
bou
t4
.25
%;
our
clot
hin
g st
ores
was
in
1.2
5%
, but
this
is n
ot a
big
pro
ble
m b
ecau
se w
est
ill h
ave
big
gest
shar
e of
mar
ket.
A 2
.5%
in
boo
kst
ores
, but
all
hav
e to
be
sold
in 2
00
2 b
ecau
se w
e ar
e go
ing
to e
nd w
ith t
his
gro
up.
The
big
gest
gro
wth
was
the
elec
tric
al s
tore
s w
ith
a 5
.5%
; th
is is
bec
ause
our
reta
il gr
oup
is b
eing
spe
cial
ised
in
this
kin
d o
f st
ores
and
we
hav
e to
car
ry o
nw
ith t
his
gro
wth
..T
han
k yo
uve
ry m
uch f
oryo
ur a
tten
tion
and
Iam
sur
e th
at t
he
follo
win
g ye
ars
will
be
bet
ter
than
thes
e ar
e.
Appendix 4.2
350
Superv
isor:
VA
NT
AG
E
BE
C V
an
tag
e R
ead
ing
An
sw
er
Sh
eet
00
00 1 2 3 4 5 6 7 8 9
1 2 3 4 5 6 7 8 9
1 2 3 4 5 6 7 8 9
1 2 3 4 5 6 7 8 9
Ca
nd
ida
te N
am
eIf
no
t a
lre
ad
y p
rin
ted
, w
rite
na
me
in C
AP
ITA
LS
an
d c
om
ple
te t
he
Can
did
ate
No
. g
rid
(in
pen
cil).
Can
did
ate
’s S
ign
atu
re
Exam
inati
on
Tit
le
Cen
tre
If t
he
ca
nd
ida
te is A
BS
EN
To
r h
asW
ITH
DR
AW
Nsh
ad
e h
ere
Can
did
ate
No
.
Ce
ntr
e N
o.
Exam
inati
on
Deta
ils
21 43
8
5
109 12
11
Part
1
Ins
tru
cti
on
sU
se a
PE
NC
IL (B
or
HB
).
Rub o
ut any a
nsw
er
you w
ish to c
hange w
ith a
n e
raser.
Fo
r P
art
s 1
to
4:
Mark
one b
ox for
each a
nsw
er.
Fo
r e
xa
mp
le:
If y
ou thin
k C
is the r
ight answ
er
to the q
uestion, m
ark
your
answ
er
sheet lik
e this
:
Pa
rt 2
�
Tu
rn o
ve
r fo
r P
art
s 3
- 5
Fo
r P
art
5:
Write
your
answ
er
cle
arly in C
AP
ITA
L L
ET
TE
RS
.
Write
one lett
er
in e
ach b
ox.
Fo
r e
xa
mp
le:
AB
C0
6 7
0 A
BE
C V
- R
DP
45
8/3
58
BC
AB
C
AB
C
AB
C
AB
C
AB
CD
EF
G
AB
CD
EF
G
AB
CD
EF
G
AB
CD
EF
G
AB
CD
EF
G
AB
C
AB
C
D D D D D D D
Scri
pt
F
COM
PANY’S
PERFORM
ANCE B
ETW
EEN 1
998 –
2001
Pre-t
ax p
rofi
ts:
In 1
99
8 a
nd 1
99
9, ou
rpr
e-t
ax b
ene
fits
wer
eov
er
£5
20
million
,
how
ever
in
20
00
, ou
r pr
e-ta
x p
rofi
ts inc
reas
ed a
roun
d £
20
6 m
illion
whic
h w
as
the r
esu
lt o
f hig
h a
dve
rtis
ing
bud
get
in t
he p
revi
ous
year
and
low
int
ere
st r
ates.
On
the o
ther
han
d, in
20
01,
the p
re-t
ax b
ene
fits
decr
eas
ed o
ver
£17
2 m
illion
wit
h r
ega
rd t
o 2
00
0.
Adve
rtis
ing
Com
pare
d w
ith
199
8, in
19
99
the e
xpe
ndit
ure
was
8%
hig
her
, th
at is
to s
ay, 17
% w
hilst
in
20
00
the e
xpe
ndit
ure w
asju
st 1
0%
.
It m
ust
be a
dded
, how
eve
rth
at in
20
01
the e
xpe
ndit
ure w
as 2
3%
but
we h
ope
to s
ee p
osit
ive r
esu
lts
of t
his
in
20
02
.
Sal
es
grow
th in
20
01
We h
ave h
ad a
4.2
5%
in
phar
mac
ies
but
, on
the o
ther
han
d,
we h
ave s
uffe
red los
ses
of 1
.25
% in
the c
loth
ing
stor
es a
lthou
gh w
e s
till
hav
e
big
gest
shar
e o
f m
arke
t.W
ith r
ega
rd t
o boo
k s
tore
s an
d e
lect
rica
l st
ores,
our
grow
th h
as b
een
arou
nd 2
.5 a
nd 5
.5%
resp
ect
ively
.
In a
ny c
ase, al
l ou
r boo
kst
ores
will be s
old in
20
02
beca
use o
f ou
r gr
eat
cam
paig
n on
radio
and
tele
visi
on. M
oreo
ver,
the b
igge
st g
row
th h
as b
een
in
ele
ctri
cal st
ores
beca
use o
urgr
oup
has
im
prov
ed o
urse
rvic
es
and
cost
s
com
pare
d w
ith o
ut c
ompe
tito
rs.
In
conc
lusi
on, I s
ugge
st w
e k
eep
the s
ame leve
lof
expe
ndit
ure
on a
dve
rtis
ing
beca
use a
s yo
u ca
n no
tice
, it
giv
es
us g
reat
res
ults
and
bene
fits
and
as
far
as
impr
ovem
ent
is
conc
ern
ed w
e a
re is
good
way
to
catc
h n
ew
cus
tom
ers
and c
lient
s
alth
ough
we m
ust
impr
ove
our
boo
k d
epa
rtm
ent
.
Scri
pt
F
All
conte
nt
po
ints
are
de
alt w
ith. T
he o
rga
nis
atio
n o
f th
e a
nsw
er
isvery
goo
d. T
here
is a
wid
e r
ange
of
lan
gua
ge a
nd e
ffective u
se o
flin
kin
g d
evic
es. T
he c
an
did
ate
has
bee
n a
mbitio
us in t
he
ir u
se
of
lan
gua
ge a
nd t
his
had le
d to s
om
e e
rrors
and s
om
e a
wkw
ard
ness
in t
he c
onclu
sio
n. O
vera
lla p
ositiv
eeff
ect on the r
ea
der.
Ba
nd
4
Appendix 4.2
351
VA
NT
AG
E
BE
C V
an
tag
e L
iste
nin
g A
nsw
er
Sh
eet
00
00 1 2 3 4 5 6 7 8 9
1 2 3 4 5 6 7 8 9
1 2 3 4 5 6 7 8 9
1 2 3 4 5 6 7 8 9
Can
did
ate
Nam
eIf
no
t a
lre
ad
y p
rin
ted
, w
rite
na
me
in C
AP
ITA
LS
an
d c
om
ple
te t
he
Can
did
ate
No
. g
rid
(in
pen
cil).
Can
did
ate
’s S
ign
atu
re
Exam
inati
on
Tit
le
Cen
tre
If t
he
ca
nd
ida
te is A
BS
EN
To
r h
asW
ITH
DR
AW
Nsh
ad
e h
ere
Can
did
ate
No
.
Ce
ntr
e N
o.
Exam
inati
on
Deta
ils
Instr
ucti
on
sU
se
a P
EN
CIL
(B
or
HB
).
Ru
b o
ut
an
y a
nsw
er
yo
u w
ish
to
ch
an
ge
with
an
era
se
r.
Continue o
n the o
ther
sid
e o
f th
is s
heet
�
Fo
r P
art
s 2
an
d 3
:
Ma
rk o
ne
bo
x f
or
ea
ch
an
sw
er.
Fo
r e
xa
mp
le:
If y
ou
th
ink C
is t
he
rig
ht
an
sw
er
to t
he
qu
estio
n,
ma
rk y
ou
r a
nsw
er
sh
ee
t lik
e t
his
:
AB
C0
For
Pa
rt 1
:
Write
yo
ur
an
sw
er
cle
arly in
CA
PIT
AL
LE
TT
ER
S.
Write
on
e le
tte
r o
r n
um
be
r in
ea
ch
bo
x.
If t
he
an
sw
er
ha
s m
ore
th
an
on
e w
ord
, le
ave
on
e b
ox e
mp
ty b
etw
ee
n w
ord
s.
Fo
r e
xa
mp
le:
Part
1 -
Co
nvers
ati
on
On
e
1
01
1
2
01
2
3
01
3
4
01
4
0
Su
pe
rvis
or:
BE
C V
- L
DP
46
0/3
60
Part
3
Part
5
14
13
16
15
17
35
34
01
01
37
36
01
01
38
01
34
35
36
37
38
AB
CD
AB
CD
19
21
20
23
22P
art
4
25
24
27
26
28
29
30
18
AB
CD
AB
CD
AB
CD
AB
CD
31
32
33
AB
CD
AB
CD
AB
CD
AB
CD
AB
CD
AB
CD
AB
CD
AB
CD
AB
CD
AB
CD
AB
CD
AB
CD
AB
CD
AB
CD
AB
CD
39
01
40
01
39
40
41
01
42
01
41
42
43
01
44
01
43
44
45
01
45
Appendix 4.2
352
Part
1 -
Co
nvers
ati
on
Tw
o
Part
1 -
Co
nvers
ati
on
Th
ree
16
15
175
01
5
AB
CD
EF
GH
AB
CD
EF
GH
AB
CD
EF
GH
Part
2 -
Secti
on
On
e
23
AB
C
14
13
AB
CD
EF
GH
AB
CD
EF
GH
Part
3
25
24
26
AB
C
AB
C
AB
C
27
28
AB
C
AB
C
29
AB
C
6
01
6
7
01
7
8
01
8
9
01
9
10
01
10
11
01
11
12
01
12
Part
2 -
Secti
on
Tw
o
21
20
22
AB
CD
EF
GH
AB
CD
EF
GH
AB
CD
EF
GH
19
18
AB
CD
EF
GH
AB
CD
EF
GH
30
AB
C
Part
1 -
Co
nvers
ati
on
Tw
o
Part
1 -
Co
nvers
ati
on
Th
ree
16
15
175
01
5
AB
CD
EF
GH
AB
CD
EF
GH
AB
CD
EF
GH
Part
2 -
Secti
on
On
e
23
AB
C
14
13
AB
CD
EF
GH
AB
CD
EF
GH
Part
3
25
24
26
AB
C
AB
C
AB
C
27
28
AB
C
AB
C
29
AB
C
6
01
6
7
01
7
8
01
8
9
01
9
10
01
10
11
01
11
12
01
12
Part
2 -
Secti
on
Tw
o
21
20
22
AB
CD
EF
GH
AB
CD
EF
GH
AB
CD
EF
GH
19
18
AB
CD
EF
GH
AB
CD
EF
GH
30
AB
C
353
Time 1 hour
INSTRUCTIONS TO CANDIDATES
Do not open this paper until you are told to do so.
Write your name, Centre number and candidate number in the spaces at the top of this page. Write these
details in pencil on your Answer Sheet if these are not already printed.
Write all your answers in pencil on your Answer Sheet – no extra time is allowed for this.
Read carefully the instructions for each part and the instructions for completing your Answer Sheet.
Try to answer all the questions.
At the end of the examination hand in both this question paper and your Answer Sheet.
INFORMATION FOR CANDIDATES
There are fifty-two questions on this question paper.
This question paper consists of 11 printed pages.
SP (NF) S31428© UCLES 2002
UNIVERSITY OF CAMBRIDGE LOCAL EXAMINATIONS SYNDICATE
Examinations in English as a Foreign Language
BUSINESS ENGLISH CERTIFICATE 0353/1HigherTest of Reading Test 023
Wednesday 5 JUNE 2002 Morning 1 hour
Additional materials:Answer Sheet
CandidateCentre Number Number
Candidate Name
APPENDIX 4.3
BEC Higher Sample Paper
Appendix 4.3
354
A B C D E
We
are
an i
nter
nati
onal
pac
kagi
ng a
nd p
rint
ing
grou
p an
d ha
ve
ambi
tiou
s pl
ans
for
futu
reex
pans
ion
both
thr
ough
org
anic
gro
wth
and
by
acqu
isit
ion.
We
are
now
see
king
a s
ucce
ssor
to o
urpr
esen
t M
anag
ing
Dir
ecto
r w
ho i
s du
e to
ret
ire
inth
ree
mon
ths’
tim
e.T
hesu
cces
sful
cand
idat
e w
ill
have
to
di
spla
y te
chni
cal
com
pete
nce
in
the
indu
stry
and
wil
l hav
e a
dem
onst
rabl
e tr
ack
reco
rdof
man
agin
g a
high
tech
nolo
gy b
usin
ess.
The
new
MD
wil
l be
exp
ecte
d to
bui
ld o
n ou
r en
viab
lebl
ue-c
hip
cust
omer
ba
se
thro
ugh
secu
re
and
prof
itab
le b
usin
ess
deve
lopm
ent
acti
viti
es.
We
are
se
eki
ng
to
ap
po
int
a P
roje
ct M
an
ag
er
to w
ork
on
a t
wo
-ye
ar
con
tra
ct w
ith th
e fi
sh p
roce
ssin
g in
du
stry
. Ap
plic
an
ts m
ust
ha
ve a
de
gre
e in
foo
d
scie
nce
o
r b
usi
ne
ss
dis
cip
line
w
ith
a
min
imu
m
of
thre
e
yea
rs’
exp
erie
nce
, p
refe
rab
ly
in
the
se
afo
od
in
du
stry
, a
nd
m
ust
b
e
ab
le
tod
em
on
stra
te
an
u
nd
ers
tan
din
g
of
the
cu
rre
nt
issu
es
faci
ng
th
e
fish
pro
cess
ing
in
du
stry
. D
utie
s w
ill
incl
ud
e
lea
din
g
a
sma
ll te
am
o
fre
sea
rch
ers
, a
sse
ssin
g t
he
ne
ed
s o
f cl
ien
t co
mp
an
ies
an
d p
rovi
din
g t
he
mw
ith s
up
po
rt,
prim
arily
th
rou
gh
th
e o
rga
nis
atio
n o
f te
chn
ica
l wo
rksh
op
s.
OP
ER
AT
ION
S M
AN
AG
ER
We
are
a m
ajo
r st
atio
ne
ry c
om
pa
ny.
Afte
r ye
ars
of
imp
ress
ive
su
cce
ss in
gro
un
d-b
rea
kin
g n
ew
pro
du
cts
an
d c
ust
om
er
rela
tion
ship
de
velo
pm
en
t, o
ur
pre
sen
t ob
ject
ive
is to
drive
ma
nu
fact
urin
gp
roce
sse
s h
igh
er
up
th
e a
ge
nd
a a
nd
we
are
no
w c
om
mitt
ed
to
ma
nu
fact
urin
g i
nn
ova
tion
. W
ew
ish
, th
ere
fore
, to
a
pp
oin
t a
S
en
ior
Op
era
tion
s M
an
ag
er
to
imp
art
th
e
very
la
test
in
ma
nu
fact
urin
g d
eve
lop
me
nt. D
rive
, e
nth
usi
asm
an
d a
pa
ssio
n f
or
exc
elle
nce
are
re
qu
ire
d,
as
isth
e a
bili
ty t
o w
in a
sim
ilar
resp
on
se f
rom
co
llea
gu
es
at
all
leve
ls.
Suc
cess
for
this
£40
mil
lion
foo
d pr
oduc
tion
pla
nt h
as c
ome
as a
res
ult o
f cl
ear
nati
onal
mar
ket
focu
s, c
oupl
ed w
ith
qual
ity
prod
ucts
com
man
ding
pre
miu
m p
rice
s. D
eman
d co
ntin
ues
toou
tstr
ip th
e ab
ilit
y to
pro
duce
and
new
pro
duct
line
s ha
ve b
een
enth
usia
stic
ally
rec
eive
d by
the
mar
ket p
lace
. To
ensu
re th
at th
e bu
sine
ss m
eets
its
dem
andi
ng c
usto
mer
req
uire
men
ts in
a w
ell-
cont
roll
ed a
nd p
rofe
ssio
nal
fash
ion,
a H
ead
of P
rodu
ctio
n is
now
nee
ded
to i
nsta
ll g
ood
man
ufac
turi
ng p
ract
ices
and
to e
nsur
e th
at th
e pr
oduc
tion
sta
ff a
re m
ould
ed m
ore
posi
tive
ly in
toa
cohe
sive
and
res
pons
ive
unit
.HEA
D O
F PR
OD
UC
TIO
N
Pro
ject
Man
ager
Man
agin
g D
irec
tor
We
are
look
ing
for
som
eone
with
tea
m m
anag
emen
t, da
taba
se a
nd p
roce
ss d
evel
opm
ent
skill
s to
hea
dou
r cu
stom
er s
ervi
ce d
epar
tmen
t. Th
e pe
rson
app
oint
ed w
ill b
e re
spon
sibl
e fo
r m
anag
ing
oper
atio
nal
deliv
ery
and
perf
orm
ance
. H
e or
she
will
hav
e to
dem
onst
rate
exp
erie
nce
in t
he m
anag
emen
t of
fluct
uatin
g su
pply
and
dem
and
situ
atio
ns.
The
com
pany
, a
lead
er i
n th
e pr
ovis
ion
of s
ervi
ces
tobu
sine
sses
in
the
tele
com
mun
icat
ions
sec
tor,
has
a st
rong
rep
utat
ion
for
qual
ity a
nd s
peed
of
solu
tion
deliv
ery.
We
are
pois
ed t
o im
plem
ent
an e
xplo
sive
gro
wth
pla
n an
d ar
e ta
rget
ed t
o tr
eble
in
size
by
2004
.
HEA
D O
F C
UST
OM
ER S
ERV
ICE
PA
RT
ON
E
Qu
esti
on
s 1
– 8
•L
oo
k a
t th
e s
tate
me
nts
be
low
an
d a
t th
e e
xtra
cts
fro
m f
ive
job
ad
vert
ise
me
nts
on
th
e o
pp
osi
te
pa
ge
.
•W
hic
h a
dve
rtis
em
en
t(A
,B
,C
,D
or
E)
do
es
ea
ch s
tate
me
nt
1 –
8re
fer
to?
•F
or
ea
ch s
tate
me
nt
1 –
8, m
ark
one le
tter
(A,
B,
C,
Dor
E)
on
yo
ur
An
swe
r S
he
et.
•Y
ou
will
ne
ed
to
use
so
me
of
the
se le
tte
rs m
ore
th
an
on
ce.
•T
he
re is
an
exa
mp
le a
t th
e b
eg
inn
ing
, (0
).
Exa
mp
le:
0T
his
co
mp
an
y is
offe
rin
g a
job
on
ly o
n a
te
mp
ora
ry b
asi
s.
0
1T
his
co
mp
an
y is
kn
ow
n f
or
de
alin
g w
ith p
rob
lem
s q
uic
kly.
2T
his
co
mp
an
y p
lan
s to
ta
ke o
ver
oth
er
com
pa
nie
s.
3T
his
co
mp
an
y h
as
as
its c
lien
ts s
om
e o
f th
e c
ou
ntr
y’s
lea
din
g c
om
pa
nie
s.
4T
his
co
mp
an
y is
no
t tu
rnin
g o
ut
as
ma
ny
go
od
s a
s it
cou
ld s
ell.
5T
he
pe
rso
n a
pp
oin
ted
to
th
is p
ost
will
de
live
r a
ssis
tan
ce t
o o
the
r co
mp
an
ies.
6T
his
co
mp
an
y w
an
ts t
o c
ha
ng
e t
he
ma
in f
ocu
s o
f its
att
en
tion
.
7T
his
co
mp
an
y’s
go
od
s a
re e
xpe
nsi
ve.
8T
he
pe
rso
n a
pp
oin
ted
to
th
is jo
b w
ill h
ave
to
sh
ow
an
ab
ility
to
de
al w
ith e
ver-
cha
ng
ing
ma
rke
t
con
diti
on
s.
AB
CB
E
Appendix 4.3
355
AA
t th
e s
am
e t
ime
, th
e e
con
om
ic c
lima
te s
ug
ge
sts
tha
t th
ere
is a
ve
ry r
ea
l pro
spe
ct o
f le
an
er
bu
dg
ets
in f
utu
re.
BW
e a
re in
stin
ctiv
ely
incl
ine
d t
o f
ee
l mo
re p
osi
tive
to
pe
op
le w
ho
are
sim
ilar
to u
s.
CH
ow
eve
r, s
om
e e
xpe
rts
fee
l th
at
the
ma
in a
dva
nta
ge
of
on
-scr
ee
n in
terv
iew
ing
is t
ha
t it
ad
dre
sse
s fla
ws
in t
he
co
nve
ntio
na
l in
terv
iew
.
DF
or
this
re
aso
n t
he
re w
ill a
lwa
ys b
e a
pla
ce f
or
it.
EF
or
som
e p
ost
s, a
pp
lica
nts
ma
y b
e a
ble
to
ch
oo
se b
etw
ee
n a
fa
ce-t
o-f
ace
inte
rvie
w a
nd
an
on
-scr
ee
n in
terv
iew
.
FS
om
e p
eo
ple
in in
du
stry
co
nsi
de
r th
is t
o b
e a
fa
r m
ore
re
liab
le a
pp
roa
ch t
o s
ele
ctio
n t
ha
n a
hig
h-t
ech
inte
rvie
w.
GB
ut
for
it to
be
co
mp
lete
ly f
air,
th
e o
n-s
cre
en
me
tho
d w
ou
ld h
ave
to
be
use
d w
ith a
ll
can
did
ate
s.
HE
mp
loye
rs n
ow
ha
ve a
t th
eir d
isp
osa
l a r
an
ge
of
com
mu
nic
atio
n t
oo
ls.
AB
CD
EF
GH
Exa
mp
le:
0
PA
RT
TW
O
Qu
esti
on
s 9
– 14
•R
ea
d t
his
te
xt f
rom
an
art
icle
ab
ou
t jo
b in
terv
iew
s.
•C
ho
ose
th
e b
est
se
nte
nce
fro
m t
he
op
po
site
pa
ge
to
fill
ea
ch o
f th
e g
ap
s.
•F
or
ea
ch g
ap
9–
14,
mark
one le
tter
(A–
H)
on
yo
ur
An
swe
r S
he
et.
•D
o n
ot
use
an
y le
tte
r m
ore
th
an
on
ce.
•T
he
re is
an
exa
mp
le a
t th
e b
eg
inn
ing
, (0
).
The
pr
oble
ms
of
glob
al
recr
uitm
ent
are
disa
ppea
ring
rap
idly
. The
rea
son
for
this
lies
in th
ete
chno
logy
tha
t co
uld
rede
fine
the
tra
diti
onal
job
inte
rvie
w.
The
se g
ive
them
acc
ess
to th
egl
obal
re
crui
tmen
t m
arke
t, en
abli
ng
them
to
inte
rvie
w a
nd a
sses
s th
eir
choi
ce o
f ca
ndid
ates
on
scre
en,
for
exam
ple
via
vide
o-co
nfer
ence
li
nk,
CD
-Rom
dis
play
or
elec
tron
ic f
ile
tran
sfer
.T
he d
evel
opm
ent
of t
he u
se o
f te
chno
logy
as
am
etho
d of
rec
ruit
men
t ha
s br
ough
t co
nsid
erab
lebe
nefi
ts t
o re
crui
tmen
t pr
acti
ces.
For
exa
mpl
e, i
tm
eans
gre
at s
avin
gs i
n te
rms
of b
oth
tim
e an
dth
e tr
avel
bu
dget
.O
ne
prob
lem
w
ith
face
-to-
face
int
ervi
ews
is t
hat
body
lan
guag
e is
boun
d to
play
an
im
port
ant
part
in
th
em.
Thi
s ne
cess
aril
y le
ads
to
an
inhe
rent
unfa
irne
ss
in
such
in
terv
iew
s.
Put
ting
di
stan
cebe
twee
n ca
ndid
ate
and
inte
rvie
wer
wit
h th
e us
e of
a vi
deo
cam
era
can
help
to o
verc
ome
this
pro
blem
as b
ody
lang
uage
wil
l be
less
obv
ious
.It
coul
d pr
ove
an
unfa
ir
adva
ntag
e,
or
poss
ibly
disa
dvan
tage
, if
use
d on
ly w
ith
thos
e un
able
to
atte
nd a
fac
e-to
-fac
e in
terv
iew
.A
grea
t de
al h
as b
een
mad
e in
rec
ent
year
s of
NL
P(n
euro
-lin
guis
tic
prog
ram
min
g),
whi
ch
incl
udes
the
sci
ence
of
body
lan
guag
e, a
nd i
tsva
lue
in
job
inte
rvie
ws.
Oth
ers,
how
ever
, re
ject
th
e ne
w
tech
nolo
gy
sim
ply
beca
use
they
ar
e af
raid
of
it
. T
he
bene
fits
of
tech
nolo
gy,
thou
gh,
are
too
grea
t to
ign
ore,
whe
non
e co
nsid
ers
that
the
bes
t pe
rson
for
a p
arti
cula
rjo
b m
ay d
ecid
e no
t to
att
end
for
inte
rvie
w i
f he
or
she
has
to t
rave
l a
cons
ider
able
dis
tanc
e.A
ppoi
ntin
g se
nior
ex
ecut
ives
is
in
crea
sing
lyse
en a
s a
glob
al b
usin
ess.
Com
pani
es w
hich
inte
ndto
sel
ect
cand
idat
es f
or j
obs
from
a w
ider
poo
l w
ill
have
lit
tle
choi
ce b
ut t
o be
ar t
he c
ost
ofov
erse
as t
rave
l to
con
vent
iona
l in
terv
iew
s, o
r to
embr
ace
the
new
tech
nolo
gy.
Inev
itab
ly,
com
pani
es
wil
l be
se
ekin
g m
ore
cost
-eff
ecti
vew
ays
of r
ecru
itin
g qu
alit
y ca
ndid
ates
, and
for
this
,vi
rtua
l in
terv
iew
ing
may
off
er a
sol
utio
n.T
he g
reat
est v
alue
of
face
-to-
face
inte
rvie
ws
is a
tth
e st
age
of f
inal
sel
ecti
on.
N
ever
thel
ess,
ther
e ar
e m
any
posi
tive
as
pect
s of
us
ing
tech
nolo
gy
as
a re
crui
tmen
t to
ol.
Tim
es
are
chan
ging
, an
d un
less
the
die
-har
ds w
ho i
gnor
ene
w t
echn
olog
y ch
ange
wit
h th
em,
they
may
fin
dth
emse
lves
lef
t be
hind
.
9
11
12
13
10
14
Inte
rvie
win
g o
n s
cre
en
0H
Appendix 4.3
356
15C
aro
l Nic
ho
ls p
refe
rre
d t
o u
se s
ma
ller
con
sulta
nci
es
be
cau
se
Ash
e h
ad
pre
vio
us
exp
erie
nce
of
the
m.
Bth
ey
cou
ld d
eve
lop
alo
ng
sid
e h
er
com
pa
ny.
Csh
e w
ou
ld b
e a
ble
to
ha
ve c
on
tro
l ove
r th
em
.
Dth
ey
wo
uld
imp
rove
he
r d
ep
art
me
nt’s
re
pu
tatio
n.
16P
au
l Ed
en
sa
ys o
ne
ad
van
tag
e o
f sm
alle
r co
nsu
ltan
cie
s is
th
at
Acl
ien
ts b
en
efit
fro
m c
on
tinu
ou
s in
div
idu
al c
on
tact
.
Bth
ey
ha
ve a
cle
are
r u
nd
ers
tan
din
g o
f cl
ien
ts’b
ran
ds.
Ccl
ien
ts f
ee
l th
ey
ge
t a
be
tte
r re
turn
on
th
eir in
vest
me
nt.
Dth
ey
are
ab
le t
o s
ell
the
ir id
ea
s to
clie
nts
mo
re e
ffect
ive
ly.
17P
en
ny
Wh
ite p
oin
ts o
ut
tha
t sm
alle
r co
nsu
ltan
cie
s ca
n
Ab
e f
lexi
ble
ab
ou
t co
-op
era
ting
with
oth
er
firm
s.
Bsp
en
d t
ime
re
sea
rch
ing
a w
ide
ra
ng
e o
f is
sue
s.
Cp
rovi
de
use
ful i
ntr
od
uct
ion
s to
oth
er
firm
s.
Da
dvi
se f
irm
s o
n w
ays
to
re
du
ce o
verh
ea
ds.
18P
en
ny
Wh
ite s
ays
th
at
larg
er
con
sulta
nci
es
do
no
t
Atr
ain
th
eir c
on
sulta
nts
to
wo
rk w
ith s
ma
ller
com
pa
nie
s.
Ba
pp
reci
ate
th
e f
un
ctio
n o
f p
sych
olo
gy
in b
usi
ne
ss.
Cd
eliv
er
the
re
sults
th
at
pro
ject
s a
re s
et
up
to
ach
ieve
.
Dtr
ansf
er
their e
xpert
ise f
ully
to t
heir c
lients
.
19A
cco
rdin
g t
o B
ill D
aw
kin
s, la
rge
r co
nsu
ltan
ts
Aa
re a
ble
to
pa
ss o
n e
con
om
ies
of
sca
le t
o c
lien
ts.
Bh
ave
a d
ee
pe
r u
nd
ers
tan
din
g o
f in
du
stria
l iss
ue
s.
Cre
pre
sen
t a
mo
re s
ecu
re in
vest
me
nt
for
som
e c
lien
ts.
Dd
iffe
ren
tiate
th
eir a
dvi
ce a
cco
rdin
g t
o c
ou
ntr
y.
20T
he
write
r co
ncl
ud
es
by
reco
mm
en
din
g u
sin
g s
ma
ller
con
sulta
nci
es
Ain
situ
atio
ns
requirin
g q
uic
k re
sults
.
Bfo
r m
on
itorin
g p
roje
cts’
pro
gre
ss.
Cin
co
nju
nct
ion
with
ea
ch o
the
r.
Dfo
r th
e f
iner
deta
ils o
f pro
ject
s.
PA
RT
TH
RE
E
Qu
esti
on
s 15
– 2
0
•R
ea
d t
he
fo
llow
ing
art
icle
ab
ou
t d
iffe
ren
t-si
zed
ma
na
ge
me
nt
con
sulta
nci
es
an
d t
he
qu
est
ion
s o
n
the
op
po
site
pa
ge
.
•F
or
ea
ch q
ue
stio
n 1
5 –
20,
mark
one le
tter
(A,
B,
Cor
D)
on
yo
ur
An
swe
r S
he
et
for
the
an
swe
r
you
ch
oo
se.
Afe
w
years
ago,
when
Caro
lN
ichols
arr
ived
as
head
of
hum
an r
eso
urc
es
with
NV
CT,
the
fast
-exp
andin
g
tele
com
s and
softw
are
se
rvic
es
com
pany,
she
knew
th
at
from
day
one
work
ing
with
m
anagem
ent
consu
ltancy
fir
ms
would
be
an
inte
gra
l part
of
her
role
. ‘I
had
already
deci
ded
on
the
kind
of
consu
ltanci
es
I w
ante
d
toem
plo
y,’
she
says
. ‘W
hen
I st
art
ed,
I w
as
pre
tty
much
a
one-w
om
an
depart
ment.
So
itw
as
import
ant
for
me
to
form
part
ners
hip
s to
help
me s
upport
the g
row
th o
f the d
epart
ment a
nd
the c
om
pany.
What I
wante
d w
as
smalle
r co
nsu
ltanci
es
with
whom
I co
uld
est
ablis
h
pers
onal
rela
tionsh
ips
– fi
rms
whic
h w
ould
gro
w
with
us,
and
be
flexi
ble
enough
to
resp
ond
to
our
changin
g n
eeds.
’
Paul E
den,
Managin
g D
irect
or
of
NV
CT,
co
nfir
ms
the
desi
rabili
tyof
smalle
r co
nsu
ltanci
es.
‘Larg
er
firm
s have
a te
ndency
to u
se o
ne
pers
on
to
sell,
and
anoth
er
todeliv
er, w
ith th
e r
esu
lt th
at c
lients
may
not
really
know
who o
r w
hat
they
are
buyi
ng.
With
a s
malle
rfir
m,
you
are
buyi
ng
the
consu
ltant
as
much
as
the
pro
duct
- t
he p
ers
on r
ath
er
than
the b
rand.’
Penny
White
, fin
anci
al
serv
ices
gro
up Inte
rco’s
Head o
f S
trate
gic
Managem
ent,
hig
hlig
hts
oth
er
adva
nta
ges
of
the
smalle
rco
nsul
tanc
y.
‘Asm
alle
rco
nsul
tanc
yre
cognis
es
that
itca
nnot
do
eve
ryth
ing,
and
ism
uch
more
will
ing t
o w
ork
with
oth
er
pre
ferr
ed
consu
ltants
fo
rth
e g
ood o
f th
e c
lient,’s
he s
ays
.‘A
nd
on
fees,
sm
alle
rco
nsu
ltanci
es
can b
e l
ess
rig
idand m
ore
co
st-e
ffect
ive,
sim
ply
beca
use
th
eir
ove
rheads
are
low
er. T
hat is
not to
say
that th
ey
need
to
underc
ut
to
win
busi
ness
, but
part
of
a
small
consu
ltancy
’s
stra
tegy
must
be
to t
horo
ughly
inve
stig
ate
how
to
add v
alu
e t
o e
very
thin
g i
t does.
Larg
er
consu
ltanci
es
are
gain
ing
exp
ert
ise in
busi
ness
psy
cholo
gy
and a
pply
ing it
to r
unnin
g c
hange
pro
gra
mm
es,
but
they
still
tend
to
bring
in
their
ow
n
team
to
imple
ment
pro
ject
s,
whic
hm
eans
that
when t
hey
move
on,
the
know
-how
goes
with
th
em
,le
avi
ng
the
clie
nt
with
a
know
ledge
vacu
um
, not
the
inte
gra
ted
train
ing
that
small
firm
s, in
part
icula
r, r
eally
need.’
But
the l
arg
er
consu
ltanci
es
do
have
th
eir
advo
cate
s.
Bill
Daw
kins,
edito
r of
Con
sulta
ncy
Toda
y:
‘One
are
a
where
th
e
indust
ry g
iants
have
an e
dge i
sw
here
m
ajo
r glo
bal
com
panie
sre
quire
a
standard
ised
serv
ice
acr
oss
a
num
ber
of
diff
ere
nt
countr
ies.
S
uch
cl
ients
are
frequently
sp
endin
g
subst
antia
lsu
ms
of
money
in
consu
lting
engagem
ents
, and,
not
surp
risi
ngly
, th
ey
are
seeki
ng t
he
reass
ura
nce
of
a r
eco
gnis
ed a
nd
resp
ect
ed
bra
nd
whic
h
they
know
they
can t
rust
to d
eliv
er.’
When
it co
mes
to
choosi
ng
whic
h k
ind o
f consu
ltancy
to u
se,
there
is
no r
ight
or
wro
ng i
n a
ny
abso
lute
se
nse
. B
y th
eir
very
natu
re,
smalle
r entr
ants
are
able
to
move
m
ore
sw
iftly
th
an
the
larg
er
firm
s. B
ut
the q
uest
ion i
sw
heth
er
they
have
the n
ece
ssary
subst
ance
and
track
re
cord
behin
d t
hem
to s
ee l
arg
er-
scale
pro
gra
mm
es
thro
ugh.
Choose
asm
alle
r co
nsu
ltancy
fo
r pilo
tim
ple
menta
tions
where
you w
ant
‘look
and s
ee’s
olu
tions
in a
short
space
of
time.
Then
turn
to
a
larg
er
firm
for
full
imple
menta
tion
and t
ransf
orm
atio
n p
rogra
mm
es.
Incr
easi
ngly
, th
e c
hoic
e b
etw
een
big
and
small
is
not
mutu
ally
excl
usiv
e, b
ut c
ompl
emen
tary
. The
two
ofte
n fin
d th
emse
lves
wor
king
togeth
er
on t
he s
am
e p
roje
ct -
creatin
g a
com
bin
atio
n n
eith
er
of
them
can a
chie
ve o
n it
s ow
n.
Appendix 4.3
357
Exa
mp
le: A
avai
labl
eB
conv
enie
ntC
appo
inte
dD
obta
inab
le
21A
deci
sion
sB
cons
ider
atio
nsC
conc
lusi
ons
Dre
solu
tions
22A
rela
ting
Bob
serv
ing
Cre
gard
ing
Dac
coun
ting
23A
appe
alB
wel
com
eC
requ
est
Din
vite
24A
impl
icat
edB
conc
erne
dC
incl
uded
Dre
ferr
ed
25A
At
BTo
CB
yD
With
26A
expr
esse
sB
mar
ksC
exhi
bits
Din
dica
tes
27A
prem
ium
Bbo
nus
Cco
mm
issi
onD
rew
ard
28A
mas
sB
volu
me
Cbu
lkD
capa
city
29A
prac
tice
Bco
urse
Cpr
oced
ure
Dha
bit
30A
indu
cem
ent
Bm
otiv
eC
influ
ence
Dpr
ovoc
atio
n
AB
CD
0
PA
RT
FO
UR
Qu
esti
on
s 21
– 3
0
•R
ea
d t
he
art
icle
be
low
ab
ou
t p
rici
ng
po
licie
s.
•C
ho
ose
th
e c
orr
ect
wo
rd t
o f
ill e
ach
ga
p f
rom
A,
B,
Cor
Do
n t
he
op
po
site
pa
ge
.
•F
or
ea
ch q
ue
stio
n21
– 3
0, m
ark
one le
tter
(A,
B,
Cor
D)
on
yo
ur
An
swe
r S
he
et.
•T
he
re is
an
exa
mp
le a
t th
e b
eg
inn
ing
, (0
).
Whe
neve
r a
prod
uct
or s
ervi
ce i
s m
ade
......
(0)..
....
for
sale
, on
e of
the
mos
t im
port
ant
......
(21)
......
to b
e m
ade
isth
e on
e re
late
d to
the
pri
ce t
o be
cha
rged
. To
hav
e no
cohe
rent
pol
icy
......
(22)
......
pric
e –
mer
ely
to ‘
thin
k of
anu
mbe
r’–
is t
o ...
...(2
3)...
...tr
oubl
e.T
he b
asic
poi
nt a
s fa
r as
pri
cing
is
......
(24)
......
is t
oan
swer
the
que
stio
n, ‘
......
(25)
......
wha
t le
vel
shou
ld w
epi
tch
our
pric
es?’
Are
lati
vely
hig
h pr
ice
(in
com
pari
son
toth
e co
mpe
titi
on)
......
(26)
......
that
th
e pr
oduc
t ha
sso
met
hing
spe
cial
abo
ut it
not
fou
nd in
the
othe
r pr
oduc
ts.
In
othe
r w
ords
, th
e cu
stom
er
is
expe
cted
to
pa
y a
......
(27)
......
for
the
extr
a-sp
ecia
l qu
alit
ies
to b
e fo
und
inth
e pr
oduc
t. T
his
also
app
lies
to
serv
ices
lik
e an
y fo
rm o
fm
aint
enan
ce o
r re
pair
wor
k. U
nfor
tuna
tely
, it
is
a w
ell-
esta
blis
hed
econ
omic
law
tha
t th
e hi
gher
the
pri
ce,
the
low
er t
he .
.....(
28)..
....
sold
. N
onet
hele
ss,
both
end
s of
the
mar
ket
can
be e
qual
ly p
rofi
tabl
e.T
he
ques
tion
of
di
scou
nts
is
impo
rtan
t to
o.
Som
eor
gani
sati
ons
offe
r di
scou
nts
out
of .
.....(
29)..
....
, w
hile
othe
rs
neve
r gi
ve
any
kind
of
di
scou
nt.
A‘q
uant
ity
disc
ount
’ca
n at
trac
t cu
stom
ers:
the
mor
e th
ey b
uy,
the
low
er
the
unit
pr
ice.
‘P
rom
pt-p
aym
ent
disc
ount
s’ar
ean
othe
r...
...(3
0)...
...to
the
cus
tom
er (
usua
lly
reta
iler
s),
whe
reby
if p
aym
ent i
s m
ade
quic
kly
(say
, with
in te
n da
ys),
the
amou
nt p
ayab
le i
s le
ss t
han
it w
ould
nor
mal
ly b
e.
Prici
ng p
olic
ies
Appendix 4.3
358
PA
RT
SIX
Qu
esti
on
s 41
– 5
2
•R
ea
d t
he
te
xt b
elo
w a
bo
ut
the
ho
tel i
nd
ust
ry.
•In
mo
st o
f th
e li
ne
s 41
– 5
2th
ere
is o
ne
ext
ra w
ord
. It
is e
ithe
r g
ram
ma
tica
lly in
corr
ect
or
do
es
no
t fit
in w
ith t
he
se
nse
of
the
te
xt.
So
me
lin
es,
ho
we
ver,
are
co
rre
ct.
•If a
line is
corr
ect
, w
rite
CO
RR
EC
To
n y
ou
r A
nsw
er
Sh
ee
t.
•If
th
ere
is a
n e
xtra
wo
rd in
th
e li
ne
, w
rite
th
e ex
tra
wo
rdin
CA
PIT
AL
LE
TT
ER
S o
n y
ou
r a
nsw
er
she
et.
•T
he
exe
rcis
e b
eg
ins
with
tw
o e
xam
ple
s, (
0) a
nd (
00).
Exa
mp
les:
0W
IL
L
00C
OR
RE
CT
0Is
th
ere
an
yon
e in
th
e h
ote
l in
du
stry
wh
o w
ill,
inst
ea
d o
f ju
st c
om
pla
inin
g t
ha
t th
ey
00ca
n’t
find
re
liab
le q
ua
lifie
d s
taff,
no
t to
me
ntio
n r
eta
in t
he
m,
is a
ctu
ally
pre
pa
red
41to
he
lp s
tud
en
ts w
ith c
on
tinu
ed
pro
fess
ion
al d
eve
lop
me
nt?
At
pre
sen
t tim
e I
’m a
n
42a
sso
cia
te m
em
be
r o
f a
pro
fess
ion
al b
od
y in
th
e h
ote
l ma
na
ge
me
nt.
In
ord
er
to
43u
pg
rad
e t
o f
ull
me
mb
ers
hip
, I
de
cid
ed
ho
w t
o u
nd
ert
ake
th
e P
rofe
ssio
na
l Ce
rtifi
cate
.
44A
s p
art
of
the
co
urs
e,
I h
ad
re
qu
ire
me
nt
to c
om
ple
te a
n a
ssig
nm
en
t o
n f
ron
t-o
ffice
45o
pe
ratio
ns.
Th
is s
ee
me
d s
tra
igh
tfo
rwa
rd,
bu
t I
cou
ldn
’t ye
t fin
d o
ne
est
ab
lish
me
nt
46th
at
wa
s p
rep
are
d t
o a
llow
me
to
vis
it a
nd
ga
the
r th
e in
form
atio
n I
re
qu
ire
d.
So
me
of
47th
e 1
2 h
ote
ls I
co
nta
cte
d,
on
ly t
wo
act
ua
lly h
ad
th
e d
ece
ncy
to
exp
lain
th
at
‘no
rma
lly
48it
wo
uld
be
OK
’, b
ut
at
the
mo
me
nt
the
y co
uld
n’t
spa
re t
he
tim
e o
r st
aff.
Ju
st a
s fo
r
49th
e r
est
, it
wa
s si
mp
ly ‘N
o’.
Wo
uld
so
me
on
e p
lea
se t
ell
to m
e,
an
d a
ll th
e o
the
r
50d
ed
ica
ted
ho
spita
lity
pro
fess
ion
als
ou
t th
ere
wh
o a
re t
ryin
g t
o f
urt
he
r o
n t
he
ir c
are
er
51p
rosp
ect
s vi
a c
on
tinu
ed
pro
fess
ion
al d
eve
lop
me
nt,
exa
ctly
ho
w w
e a
tta
in t
he
52q
ua
lific
atio
ns
tha
t th
e in
du
stry
re
qu
ire
s u
s, w
he
n t
he
ind
ust
ry s
ee
ms
un
will
ing
to
he
lp?
PO
OR
SU
PP
OR
T FO
R H
OTE
L ST
UD
ENTS
PA
RT
FIV
E
Qu
esti
on
s 31
– 4
0
•R
ea
d t
he
art
icle
be
low
ab
ou
t th
e im
po
rta
nce
of
the
offi
ce e
nvi
ron
me
nt.
•F
or
ea
ch q
ue
stio
n 3
1 –
40,
write
on
e w
ord
in C
AP
ITA
LL
ET
TE
RS
on
yo
ur
An
swe
r S
he
et.
•T
he
re is
an
exa
mp
le a
t th
e b
eg
inn
ing
, (0
).
Exa
mp
le:
0A
N
Is y
our
offi
ce ..
....(0
).....
.attra
ctiv
e a
nd c
om
fort
able
pla
ce?
Is
it sp
eci
fically
desi
gned
to
ensu
re
that
whate
ver
stre
sses
you e
nco
unte
r in
the c
ours
e o
f yo
ur
work
, yo
ur
surr
oundin
gs
make
lif
e
just
th
at
little
bit
......
(31)
......
beara
ble
?...
...(3
2)...
...yo
u g
reete
d e
very
morn
ing b
y cu
tflo
wers
, th
e s
mell
of
fresh
ly b
rew
ed c
offe
e a
nd a
colo
ur
schem
e th
at .
.....(
33)..
....e
asy
on th
e e
ye? O
r do y
ou h
ave
to
settle
fo
r a
desk
co
vere
d
with
th
e
pen
mark
s of
num
ero
us
form
er
em
plo
yees
and a
sta
tionery
cupboard
that
can b
e o
pened o
nly
with
a p
icka
xe?
If t
he s
eco
nd s
cenario s
ounds
more
fam
iliar,
you a
re b
y...
...(3
4)...
...m
eans
alo
ne. A
rece
nt
surv
ey
found t
hat
38%
of
em
plo
yees
feel
the i
nte
rior
desi
gn o
f th
e o
ffice
they
work
in p
reve
nts
them
fro
m p
erf
orm
ing .
.....(
35)..
....
the
best
of
their a
bili
ties.
Many
em
plo
yers
re
fuse
to
ente
rtain
th
e
thought
of
impro
ving a
nd u
pdatin
g t
heir o
ffice
s ...
...(3
6)...
...of
the
cost
s in
volv
ed.
In
the
long
run,
how
eve
r,
it m
ight
be
unw
ise
to
be
too
tight-
fiste
d
......
(37)
......
it co
mes
toem
plo
yees’
com
fort
. T
he
work
ing
envi
ronm
ent
......
(38)
......
a d
irect
effe
ct o
n p
roduct
ivity
, and 7
8%
of
boss
es
......
(39)
......
resp
onded to
the s
urv
ey
agre
ed th
at a
ple
asa
nt
offi
ce
is
a
majo
r in
fluence
in
attra
ctin
g
and
reta
inin
g
good-q
ualit
y w
ork
ers
. E
mplo
yee
under-
perf
orm
ance
can n
ot
only
spell
financi
al l
oss
; ...
...(4
0)...
...als
o f
uels
pers
onal
frust
ratio
n w
hen t
he e
mplo
yee f
eels
unfu
lfille
d.
And
it’s
hig
hly
lik
ely
th
at
the
dis
satis
fied
secr
eta
ry
will
lo
ok
to
gre
ener
past
ure
s –
or
cleaner
offi
ces.
Bea
utif
ul i
s b
est
Appendix 4.3
359
TIME 1 hour 10 minutes
INSTRUCTIONS TO CANDIDATES
Do not open this paper until you are told to do so.
Write your name, Centre number and candidate number in the spaces at the top of this page and on each
sheet of answer paper used.
Read the instructions carefully.
This paper requires you to complete two tasks.
Answer the Part 1 task and one task from Part 2.
Write your answers on the separate answer paper provided.
Write clearly in pen, not pencil. You may make alterations but make sure that your work is easy to read.
If you use more than one sheet of paper, fasten the sheets together.
At the end of the examination hand in both this question paper and your answer paper.
INFORMATION FOR CANDIDATES
Part 2 carries twice as many marks as Part 1.
UNIVERSITY OF CAMBRIDGE LOCAL EXAMINATIONS SYNDICATE
Examinations in English as a Foreign Language
BUSINESS ENGLISH CERTIFICATE 0353/2HigherTest of Writing Test 023
Wednesday 5 JUNE 2002 Morning 1 hour 10 minutes
Additional materials:Answer Paper
CandidateCentre Number Number
Candidate Name
This question paper consists of 3 printed pages.
SP (SM/JB) S30782© UCLES 2002
Appendix 4.3
360
PA
RT
TW
O
Answ
er
on
eof
the q
uest
ions
2,3
or
4belo
w.
Qu
esti
on
2
•A
min
or
acc
ide
nt
ha
s re
cen
tly t
ake
n p
lace
in y
ou
r o
rga
nis
atio
n.
Yo
ur
ma
na
ge
r h
as
ask
ed
yo
u t
o
find
ou
t a
bo
ut
the
acc
ide
nt
an
d w
rite
a r
ep
ort
su
mm
arisi
ng
th
e in
form
atio
n y
ou
ha
ve g
ath
ere
d.
•W
rite
your
rep
ort
, in
clu
din
g t
he
fo
llow
ing
info
rma
tion
:
•a
brie
f d
esc
rip
tion
of
the
acc
ide
nt
•w
ha
t yo
u t
hin
k ca
use
d t
he
acc
ide
nt
•w
he
the
r a
ny
me
asu
res
ne
ed
to
be
ta
ken
to
pre
ven
t si
mila
r a
ccid
en
ts h
ap
pe
nin
g.
•W
rite
200
– 2
50w
ord
s o
n t
he
se
pa
rate
an
swe
r p
ap
er
pro
vid
ed
.
Qu
esti
on
3
•Y
ou
r d
ep
art
me
nt
ha
s re
cen
tly t
ake
n o
n a
nu
mb
er
of
ne
w s
taff.
T
he
Hu
ma
n R
eso
urc
es
ma
na
ge
r
of
you
r co
mp
an
y h
as
ag
ree
d t
o a
re
qu
est
fo
r a
se
rie
s o
f tr
ain
ing
se
ssio
ns
tha
t w
ou
ld e
na
ble
th
e
ne
w e
mp
loye
es
to w
ork
mo
re e
ffici
en
tly.
Sh
e h
as
ask
ed
yo
u t
o w
rite
a p
rop
osa
l fo
r th
e t
rain
ing
.
•W
rite
the p
rop
osa
l, o
utli
nin
g t
he
tra
inin
g r
eq
uire
d,
an
d in
clu
de
th
e f
ollo
win
g p
oin
ts:
•a
brie
f d
esc
rip
tion
of
the
ro
les
of
the
ne
w s
taff
•th
e t
ype
of
tra
inin
g t
he
y n
ee
d a
nd
wh
y
•a
ny
follo
w-u
p t
rain
ing
or
ass
ess
me
nt
tha
t m
igh
t b
e r
eq
uire
d.
•W
rite
200
– 2
50w
ord
s o
n t
he
se
pa
rate
an
swe
r p
ap
er
pro
vid
ed
.
Qu
esti
on
4
•Y
ou
are
he
lpin
g t
o a
rra
ng
e a
n e
ven
t fo
r a
ll yo
ur
com
pa
ny’
s cu
sto
me
rs,
at
wh
ich
yo
ur
ne
w r
an
ge
of
pro
du
cts
will
be
de
mo
nst
rate
d.
•W
rite
a l
ette
rto
yo
ur
cust
om
ers
, in
clu
din
g t
he
fo
llow
ing
info
rma
tion
:
•th
e n
atu
re a
nd
pu
rpo
se o
f th
e e
ven
t
•d
eta
ils o
f th
e e
ven
t
•w
hy
the
eve
nt
will
be
wo
rth
att
en
din
g
•h
ow
yo
ur
cust
om
ers
sh
ou
ld r
esp
on
d t
o y
ou
r le
tte
r.
•W
rite
200
– 2
50w
ord
s o
n t
he
se
pa
rate
an
swe
r p
ap
er
pro
vid
ed
.
PA
RT
ON
E
Qu
esti
on
1
•T
he
gra
ph
be
low
co
mp
are
s th
e e
xpo
rt s
ale
s o
f a
co
mp
an
y ca
lled
LT
G w
ith it
s d
om
est
ic s
ale
s
fro
m 1
99
7 t
o 2
00
1.
•U
sin
g t
he
info
rma
tion
fro
m t
he
gra
ph
, w
rite
a s
ho
rt r
epo
rtco
mp
arin
g t
he
tw
o s
ets
of
sale
s
figu
res
du
rin
g t
his
pe
rio
d.
•W
rite
ab
ou
t 12
0 –
140
wo
rds
on
th
e s
ep
ara
te a
nsw
er
pa
pe
r p
rovi
de
d.
£10
£8
£6
£4
£2
Sal
es f
igu
res
(mill
ion
s)
Begin
nin
g o
f 1997
2000
2001
1999
1998
LTG
SA
LE
S
Dom
est
ic s
ale
s
Exp
ort
sale
s
Yea
r
Appendix 4.3
361
TIME Approx. 40 minutes (including 10 minutes’ transfer time)
INSTRUCTIONS TO CANDIDATES
Do not open this paper until you are told to do so.
Write your name, Centre number and candidate number in the spaces at the top of this page. Write these
details in pencil on your Answer Sheet if these are not already printed.
Listen to the instructions for each part carefully.
Try to answer all the questions.
Write your answers on this question paper.
At the end of the test you will have 10 minutes to copy your answers onto your Answer Sheet.
Read the instructions for completing your Answer Sheet carefully.
Write all your answers in pencil.
At the end of the examination hand in both this question paper and your Answer Sheet.
INFORMATION FOR CANDIDATES
Instructions are given on the tape.
You will hear everything twice.
There are thirty questions on this paper.
UNIVERSITY OF CAMBRIDGE LOCAL EXAMINATIONS SYNDICATE
Examinations in English as a Foreign Language
BUSINESS ENGLISH CERTIFICATE 0353/3HigherTest of Listening Test 023
Wednesday 5 JUNE 2002 Morning Approx. 40 minutes(including 10 minutes’ transfer time)
Additional materials:Answer Sheet
CandidateCentre Number Number
Candidate Name
This question paper consists of 5 printed pages.
SP (SM) S30783© UCLES 2002
Appendix 4.3
362
PA
RT
TW
O
Qu
esti
on
s 13
– 2
2
•Y
ou
will
he
ar
five
diff
ere
nt
pe
op
le t
alk
ing
ab
ou
t ru
nn
ing
th
eir o
wn
bu
sin
ess
.
•F
or
ea
ch e
xtra
ct t
he
re a
re t
wo
ta
sks.
Fo
r Ta
sk O
ne
, d
eci
de
th
e r
ea
son
th
ey
de
cid
ed
to
se
t u
p
the
ir o
wn
bu
sin
ess
fro
m t
he
list
A–
H.
Fo
r Ta
sk T
wo
, d
eci
de
wh
at
pro
ble
m t
he
y e
nco
un
tere
d
wh
ile r
un
nin
g t
he
ir b
usi
ne
ss f
rom
th
e li
st A
– H
.
•Y
ou
will
he
ar
the
re
cord
ing
tw
ice
.
TAS
K O
NE
– T
HE
RE
AS
ON
•F
or
qu
est
ion
s 13
– 1
7, m
atc
h t
he
ext
ract
s w
ith t
he
re
aso
ns,
list
ed
A–
H.
•F
or
ea
ch e
xtra
ct,
de
cid
e w
hy
the
sp
ea
ker
set
up
his
or
he
r o
wn
bu
sin
ess
.
•W
rite
on
ele
tter
(A–
H)
ne
xt t
o t
he
nu
mb
er
of
the
ext
ract
.
13..
....
....
....
....
....
....
....
....
....
....
....
14..
....
....
....
....
....
....
....
....
....
....
....
15..
....
....
....
....
....
....
....
....
....
....
....
16..
....
....
....
....
....
....
....
....
....
....
....
17..
....
....
....
....
....
....
....
....
....
....
....
TAS
K T
WO
– T
HE
PR
OB
LE
M
•F
or
qu
est
ion
s 18
– 2
2, m
atc
h t
he
ext
ract
s w
ith t
he
pro
ble
ms,
list
ed
A–
H.
•F
or
ea
ch e
xtra
ct,
de
cid
e w
ha
t p
rob
lem
ea
ch s
pe
ake
r e
nco
un
tere
d w
hile
ru
nn
ing
th
eir b
usi
ne
ss.
•W
rite
on
ele
tter
(A–
H)
ne
xt t
o t
he
nu
mb
er
of
the
ext
ract
.
18..
....
....
....
....
....
....
....
....
....
....
....
19..
....
....
....
....
....
....
....
....
....
....
....
20..
....
....
....
....
....
....
....
....
....
....
....
21..
....
....
....
....
....
....
....
....
....
....
....
22..
....
....
....
....
....
....
....
....
....
....
....
AI
wa
nte
d m
ore
fre
e t
ime
.
BI
ha
d b
ee
n m
ad
e r
ed
un
da
nt.
CI
felt
that
I la
cked jo
b s
ecu
rity
.
DI
fou
nd
a p
art
icu
lar
pro
du
ct a
pp
ea
ling
.
EI
wa
nte
d m
ore
re
wa
rds
for
my
wo
rk.
FI
fou
nd
th
at
the
ne
cess
ary
ca
pita
l wa
s a
vaila
ble
to m
e.
GI
wa
nte
d t
o d
eve
lop
a p
art
-tim
e h
ob
by.
HI
wa
nte
d t
o w
ork
fro
m h
om
e.
AI
ne
arly
ran
ou
t o
f fu
nd
ing
.
BI
ha
d p
rob
lem
s w
ith s
up
plie
rs.
CI
had d
iffic
ulty
recr
uiti
ng s
taff.
DI
un
de
rest
ima
ted
th
e s
et-
up
co
sts.
EI
ha
d t
o w
ork
to
o m
an
y h
ou
rs.
FI
sud
de
nly
ha
d t
o c
ha
ng
e lo
catio
n.
GI
ha
d p
rod
uct
ion
pro
ble
ms.
HI
ne
ed
ed
mo
re m
ark
et
rese
arc
h.
PA
RT
ON
E
Qu
esti
on
s 1
– 12
•Y
ou
will
he
ar
the
Ch
airm
an
of
Ma
son
s, a
n in
tern
atio
na
l re
taile
r, r
ep
ort
ing
to
a m
ee
ting
of
sha
reh
old
ers
.
•A
s yo
u li
ste
n,
for
qu
est
ion
s 1
– 12
, co
mp
lete
th
e n
ote
s u
sin
g u
p t
o t
hre
ew
ord
s o
r a
nu
mb
er.
•Y
ou
will
he
ar
the
re
cord
ing
tw
ice
.
MA
SON
S
An
nu
al r
epo
rt t
o s
har
eho
lder
s
Rev
iew
1A
nnua
l pro
fits
rose
by
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
....
2£8
9.4m
pro
fit in
the
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
.
356
0,00
0 sq
uare
met
res
of n
ew....
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
.....
4R
etai
l tur
nove
r up
by
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
......
Dev
elo
pm
ents
5N
ew....
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
.......
issue
d at
end
of t
he y
ear.
6Yo
rk s
tore
has
dev
elop
ed n
ew....
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
......
7In
tens
e co
mpe
titio
n ne
cess
itate
d m
any
........
........
........
........
........
........
........
........
........
........
........
........
........
........
......
.
8....
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
.mak
e up
a q
uart
er o
f foo
d pr
oduc
ts.
9C
ompe
titor
s ha
ve s
low
er....
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
.
10Ea
ch M
ason
s st
ore
has
a di
ffere
nt....
........
........
........
........
........
........
........
........
........
........
........
........
........
........
........
......
.
The
fut
ure
11Pr
ofits
will
be
affe
cted
ove
r th
e ne
xt e
ight
een
mon
ths
by....
........
........
........
........
........
........
........
........
........
...
12O
ur p
rese
nt p
lan
for
........
........
........
........
........
........
........
........
........
........
........
will
gua
rant
ee lo
ng-t
erm
pro
fits.
Appendix 4.3
363
28
Acc
ord
ing
to
Me
lan
ie,
ho
w d
o s
om
e m
an
ag
ers
re
act
wh
en
th
eir s
taff
sho
w in
itia
tive
as
a r
esu
lt o
f
tra
inin
g?
AT
hey
fail
to n
otic
e it
.
BT
he
y a
pp
reci
ate
th
e c
ha
ng
e.
CT
he
y fe
el t
hre
ate
ne
d b
y it.
29
Ho
w d
oe
s M
ela
nie
be
lieve
th
at
ma
na
ge
rs s
ho
uld
co
ntr
ibu
te t
o t
he
tra
inin
g p
roce
ss?
Ab
y te
ach
ing
so
me
se
ssio
ns
of
the
tra
inin
g p
rog
ram
me
s th
em
selv
es
Bb
y e
xam
inin
g t
he
co
nte
nt
of
em
plo
yee
tra
inin
g p
rog
ram
me
s
Cb
y re
po
rtin
g o
n t
he
ove
rall
succ
ess
of
tra
inin
g p
rog
ram
me
s
30
In in
tro
du
cin
g s
taff
tra
inin
g,
Me
lan
ie s
ays
th
at
it is
imp
ort
an
t th
at
em
plo
yee
s
Aa
re c
on
sulte
d a
bo
ut
the
co
nte
nt
of
cou
rse
s.
Ba
re n
ot
ma
de
to
fe
el t
he
y m
ust
pa
rtic
ipa
te.
Cd
o n
ot
mis
un
de
rsta
nd
its
pu
rpo
se.
Th
at
is t
he
en
d o
f th
e L
iste
nin
g T
es
t. Y
ou
no
w h
av
e t
en
min
ute
s t
o t
ran
sfe
r y
ou
r a
ns
we
rs t
oy
ou
r A
ns
we
r S
he
et.
PA
RT
TH
RE
E
Qu
esti
on
s 23
– 3
0
•Y
ou
will
he
ar
a r
ad
io in
terv
iew
with
aco
mp
an
y tr
ain
er.
•F
or
ea
ch q
ue
stio
n 2
3 –
30,
ma
rk o
ne
lett
er
(A,
Bor
C)
for
the
co
rre
ct a
nsw
er.
•Y
ou
will
he
ar
the
re
cord
ing
tw
ice
.
23A
cco
rdin
g t
o M
ela
nie
Ch
am
be
rs,
ma
ny
tra
inin
g p
rog
ram
me
s fa
il to
Ae
nco
ura
ge
wo
rke
rs t
o s
tud
y a
lon
e.
Bd
ea
l with
wh
at
wo
rke
rs la
ck.
Co
ffer
wo
rke
rs t
he
ski
lls t
he
y a
sk f
or.
24T
he
inte
rvie
we
r su
gg
est
s th
at
Hu
ma
n R
eso
urc
e m
an
ag
ers
ha
ve
Ap
rob
lem
s h
irin
g t
he
rig
ht
kin
d o
f st
aff.
Bin
suffi
cie
nt
inte
rest
in c
ert
ain
asp
ect
s o
f th
eir w
ork
.
Cd
iffic
ulty
ta
kin
g a
n in
-de
pth
vie
w o
f p
erf
orm
an
ce.
25M
ela
nie
cite
s th
e e
xam
ple
of
Gly
mo
Ele
ctric
sto
sh
ow
th
at
ma
na
ge
rs m
ay
Ah
ave
se
lfish
re
aso
ns
for
no
t tr
ain
ing
sta
ff.
Bb
e h
eld
re
spo
nsi
ble
fo
r p
oo
r tr
ain
ing
pro
gra
mm
es.
Cb
e ju
stifi
ed
in ig
no
rin
g in
ad
eq
ua
cie
s in
sta
ff p
erf
orm
an
ce.
26M
ela
nie
sa
ys t
ha
t th
e m
ost
imp
ort
an
t th
ing
ab
ou
t b
asi
c sk
ills
tra
inin
g is
th
at
Ait
mu
st f
ocu
s o
n in
div
idu
al s
kills
an
d n
ot
inte
gra
te t
he
m.
Bit
sho
uld
be
sp
eci
fica
lly t
ailo
red
to
em
plo
yee
s’jo
bs.
Cit
sho
uld
be
ta
ug
ht
in t
he
wo
rk e
nvi
ron
me
nt.
27M
ela
nie
an
d t
he
inte
rvie
we
r a
gre
e t
ha
t th
e r
esu
lts o
f a
tra
inin
g p
rog
ram
me
Aca
n t
ake
ma
na
ge
rs b
y su
rprise
.
Bo
fte
n e
xce
ed
em
plo
yee
s’e
xpe
cta
tion
s.
Cm
ay
no
t b
e im
me
dia
tely
ob
vio
us.
Appendix 4.3
364
Appendix 4.3
365
Appendix 4.3
366
Appendix 4.3
367
Fo
r 2
ca
nd
ida
tes
Inte
rlo
cu
tor
Now
, in
this
part
of
the t
est,
I’m
goin
g t
o g
ive e
ach o
f you a
choic
e o
f th
ree d
iffe
rent
topic
s.
I’d lik
e y
ou t
o s
ele
ct
one o
fyour
topic
s a
nd t
alk
about
it f
or
about
ON
E m
inute
. Y
ou’ll
have
aro
und a
min
ute
to p
repare
this
. Y
ou’r
e a
llow
ed t
o m
ake
note
s,
if y
ou w
ant
to.
All
rig
ht?
Here
are
your
topic
s.
You c
an m
ake n
ote
s o
n t
he
spare
paper
while
you a
re p
repari
ng
to t
alk
. P
lease d
on’t w
rite
anyth
ing
on y
our
topic
card
.
[Hand e
ach c
andid
ate
a d
iffe
rent
topic
card
, e.g
. S
ets
1 a
nd 5
,and s
om
e s
pare
paper
and a
pencil
for
note
s.
Allo
w 1
min
ute
’s p
repara
tion t
ime.
Both
candid
ate
s p
repare
their
talk
sat
the s
am
e t
ime,
separa
tely
.]
All
rig
ht.
Now
,*B
, w
ould
you b
eg
in b
y t
elli
ng
us w
hic
h t
opic
you’v
e c
hosen?
When y
ou’v
e f
inis
hed t
alk
ing
,*A
will
ask y
ou a
question a
bout
your
talk
. *A
, you’r
e a
llow
ed t
o t
ake n
ote
s w
hile
*B
is t
alk
ing
.
*B,
would
you lik
e t
o b
eg
in?
[Candid
ate
B s
peaks f
or
1 m
inute
. O
nly
inte
rrupt
if h
e/s
he is
unable
to s
peak w
ithout
pro
mpting.]
Thank y
ou.
*A,
is t
here
anyth
ing
you’d
lik
e t
o a
sk
*B?
[If
candid
ate
B’s
talk
does n
ot
genera
te a
question f
rom
candid
ate
A,
the I
nte
rlocuto
r m
ay f
eel it a
ppro
pri
ate
to a
sk a
question inste
ad,
if t
ime a
llo
ws.]
* U
SE
CA
ND
IDA
TE
S' N
AM
ES
TH
RO
UG
HO
UT
TH
E T
ES
T
PA
RT
2:
Min
i-p
res
en
tati
on
s (
ab
ou
t 6
min
ute
s)
Appendix 4.3
368
Thank y
ou,
*B.
Now
,*A
, it’s
your
turn
. W
hen y
ou’v
e f
inis
hed
talk
ing
,*B
will
ask y
ou a
question a
bout
your
talk
.*B
, you’r
eallo
wed t
o t
ake n
ote
s w
hile
*A
is t
alk
ing
.
All
rig
ht?
*A,
do y
ou n
eed a
few
seconds t
o t
hin
k a
bout
your
topic
ag
ain
?
[Allo
w c
andid
ate
A t
en s
econds if
necessary
.]
Can y
ou t
ell
us w
hic
h t
opic
you’v
e c
hosen t
o t
alk
about,
*A?
Would
you l
ike t
o b
eg
in? [
Use o
nly
if
candid
ate
does n
ot
sta
rtth
e t
alk
im
media
tely
.]
[Candid
ate
A
speaks fo
r 1 m
inute
. O
nly
in
terr
upt
if he/s
he is
unable
to s
peak w
ithout
pro
mpting.]
Thank y
ou.
*B,
is t
here
anyth
ing
you’d
lik
e t
o a
sk
*A?
[If
candid
ate
A
’s
talk
does
not
genera
te
a
question
from
candid
ate
B
, th
e In
terlocuto
r m
ay fe
el
itappro
pri
ate
to
ask a
question inste
ad,
if t
ime a
llo
ws.]
Thank y
ou.
[R
etr
ieve m
ate
rials
.]
* U
SE
CA
ND
IDA
TE
S' N
AM
ES
TH
RO
UG
HO
UT
TH
E T
ES
T
Appendix 4.3
369
Fo
r 2 c
an
did
ate
s
Inte
rlo
cu
tor
[Hold
the c
ard
show
ing t
he t
ask w
hile
giv
ing t
he instr
uctions
belo
w.]
[Pla
ce t
he c
ard
in f
ront
of
the c
andid
ate
s.]
[If
ne
ce
ssa
ry,
giv
e c
larifica
tio
n.
Th
en
allo
w 3
0 s
eco
nd
s f
or
candid
ate
s t
o a
bsorb
the info
rmation a
nd t
o t
hin
k h
ow
to b
egin
.A
fter
about
30 s
econds,
encoura
ge c
andid
ate
s t
o b
egin
the
task,
if t
hey h
ave n
ot
alr
eady d
one s
o.]
[Do n
ot
join
in t
his
sta
ge o
f th
e d
iscussio
n u
nle
ss it
isnecessary
to p
rom
pt
the c
andid
ate
s t
o t
alk
.]
[Aft
er
the c
andid
ate
s h
ave f
inis
hed s
peakin
g t
he inte
rlocuto
rasks q
uestions a
nd f
inis
hes t
he s
peakin
g t
est,
as d
irecte
d o
nth
e e
xam
iner’s c
opy o
f th
e t
ask c
ard
.]
PA
RT
3:
Tw
o-w
ay c
oll
ab
ora
tive
ta
sk
an
d d
isc
us
sio
n (
ab
ou
t 7
min
ute
s)
You h
ave a
bout
30 s
econds
to r
ead t
his
task c
are
fully
, and
then a
bout
3 m
inute
s t
o d
iscuss a
nd d
ecid
e a
bout
it t
og
eth
er.
You’r
e e
xpecte
d t
o g
ive r
easons f
or
your
decis
ions a
nd
opin
ions.
You d
on’t n
eed t
o w
rite
anyth
ing
. I
s t
hat
cle
ar?
Are
you r
eady t
o b
eg
in?
I'll ju
st
liste
n a
nd t
hen a
sk y
ou t
osto
p a
fter
about
3 m
inute
s.
Ple
ase s
peak s
o t
hat
we c
an
hear
you.
Now
, th
is p
art
of
the t
est
is a
dis
cussio
n a
ctivity.
Appendix 4.3
370
Ta
sk
22
– E
xa
min
er’
s C
op
y
[Ret
riev
e m
ater
ials
.]
Than
k y
ou. T
hat
is
the
end o
f th
e sp
eakin
g t
est.
Inte
rlo
cuto
r: [
Sel
ect
on
e o
r m
ore
of
the
foll
ow
ing
qu
esti
on
s a
s a
pp
rop
ria
te,
to
redre
ss a
ny
imbala
nce
in P
art
3, or
to b
roaden
the
dis
cuss
ion.]
•W
hic
h i
nce
nti
ves
do y
ou t
hin
k a
re m
ost
eff
ecti
ve
for
enco
ura
gin
g p
eople
to w
ork
har
d? (
Wh
y?)
•W
hat
dis
ad
van
tag
es
could
ther
e be
in i
ncen
tive
schem
es? (
Wh
y?)
•Is
it
esse
nti
al f
or
com
pan
ies
to r
ewar
d e
xtr
a e
ffo
rt?
(W
hy/W
hy n
ot?
)
•H
ow
do y
ou t
hin
k a
com
pan
y c
an i
nsp
ire l
oy
alt
y i
n e
mp
loyees
?
•H
ow
do
yo
u t
hin
k t
he
trad
itio
nal
em
plo
yer/e
mp
loy
ee r
ela
tio
nsh
ip m
igh
t
chan
ge
in t
he
futu
re? (
Wh
y?)
Incen
tive S
ch
em
e f
or S
taff
Your
com
pan
y i
s co
nsi
der
ing
sett
ing
up a
n i
nce
nti
ve
sch
eme
to i
mpro
ve
staf
f per
form
ance
. Y
ou h
ave
been
ask
ed t
o m
ake
reco
mm
endat
ions
for
the
schem
e.
Dis
cuss
, an
d d
ecid
e to
get
her
:
•w
hat
ben
efit
s an
in
centi
ve
sch
eme
would
bri
ng t
o t
he
com
pan
y
•w
hat
typ
es o
f in
centi
ves
could
be
off
ered
Appendix 4.3
371
An
sw
er
Ke
ys
Read
ing
Part
1
1E
2D
3D
4A
5C
6B
7A
8E
Part
2
9C
10
B11
G12
F13
A14
D
Part
315
B16
A17
A18
D19
C20
A
Part
4
21
A22
C23
D24
B25
A26
D27
A28
B29
D30
A
Part
5
31
more
32
are
33
is34
no
35
to36
because
37
when
38
has
39
who/t
hat
40
it
Part
6
41
tim
e42
the
43
ho
w44
requir
em
ent
45
yet
46
som
e4
7C
OR
RE
CT
48
just
49
to50
on
51
CO
RR
EC
T52
us
Lis
ten
ing
Part
1
1 1
1%
2 F
inancia
l S
erv
ices D
ivis
ion
3 s
elli
ng s
pace
4 5
.5%
5 m
ail
ord
er
cata
logu
e6 d
isp
lay
form
at
7 p
rice r
eductions
8 n
ew
(pro
duct)
lines
9 in
no
vation
rate
/ra
te o
f in
no
va
tio
n10 la
yout
11 (
ne
w/m
ajo
r) investm
ents
12 e
xp
ansio
n
Part
2
13
D14
A15
F16
C17
H18
H19
G20
C21
A22
E
Part
3
23
B24
C25
A26
B27
A28
C29
B30
C
Ta
sk
26
– E
xa
min
er’
s C
op
y
Fo
r th
ree c
an
did
ate
s
[Ret
riev
e m
ate
rials
.]
Than
k y
ou. T
hat
is
the
end o
f th
e sp
eakin
g t
est.
Inte
rlo
cuto
r: [
Sel
ect
on
e o
r m
ore
of
the
foll
ow
ing
qu
esti
on
s a
s a
pp
rop
ria
te,
to
redre
ss a
ny
imbala
nce
in P
art
3, or
to b
roaden
the
dis
cuss
ion.]
•W
hic
h i
nce
nti
ves
do y
ou t
hin
k a
re m
ost
eff
ecti
ve f
or
enco
ura
gin
gp
eople
to w
ork
har
d? (
Wh
y?)
•W
hat
dis
ad
van
tag
es
could
ther
e be
in i
ncen
tive
schem
es? (
Wh
y?)
•Is
it
esse
nti
al f
or
com
pan
ies
to r
ewar
d e
xtr
a e
ffo
rt?
(W
hy/W
hy n
ot?
)
•H
ow
do y
ou t
hin
k a
com
pan
y c
an i
nsp
ire l
oy
alt
y i
n e
mp
loyees
?
•H
ow
do
yo
u t
hin
k t
he
trad
itio
nal
em
plo
yer/e
mp
loy
ee r
ela
tio
nsh
ip m
igh
tch
ange
in t
he
futu
re? (
Wh
y/W
hy n
ot?
)
Incen
tive S
ch
em
e f
or S
taff
Your
com
pan
y i
s co
nsi
der
ing
sett
ing
up a
n i
nce
nti
ve
sch
eme
to i
mpro
ve
staf
f per
form
ance
. Y
ou h
ave
been
ask
ed t
o m
ake
reco
mm
endat
ions
for
the
schem
e.
Dis
cuss
, an
d d
ecid
e to
get
her
:
•w
hat
ben
efit
s an
in
centi
ve
sch
eme
would
bri
ng t
o t
he
com
pan
y
•w
hat
typ
es o
f in
centi
ves
could
be
off
ered
•w
hic
h e
mplo
yees
in t
he
com
pan
y s
hould
be
targ
eted
Appendix 4.3
372
£2
mill
ions
to
near
ly £
9 m
illio
ns.
On
the
other
han
d,
we
can
noti
ce t
hat
dom
esti
c sa
les
hav
e bee
n ne
glec
ted,
as
its
amou
nt
firs
t re
mai
ned
cons
tant
in
19
97
, th
en
fluc
tuat
ed
slig
htl
ybet
wee
n 19
98
and
20
00
, to
fin
ally
fal
l st
eadily
to
£6
mill
ions
.
Perh
aps,
th
e st
rate
gy
of
LTG
is
to
im
prov
e cu
rren
tly
its
expo
rtat
ions
wit
hou
t qu
ite
givi
ng a
way
its
dom
esti
c m
arke
t.
Scri
pt
B
All
majo
r conte
nt
poin
ts a
re i
nclu
ded
an
d t
he a
nsw
er
is w
ell
org
anis
ed
with
cle
ar
para
gra
phin
g,
an i
ntr
od
uctio
n a
nd a
conclu
sio
n.
The l
ang
uag
e,
ho
wever,
is a
dequ
ate
ra
ther
than g
ood f
or
this
leve
l.
Ba
nd
3
Scri
pt
C
This
rep
ort
com
pare
s th
e dom
esti
c an
d e
xpo
rt s
ales
mad
e by
LTG
fro
m 1
99
7to
20
01
by
usin
g a
grap
h a
s a
refe
renc
e.
The
firs
t th
ing
that
im
med
iate
ly c
atch
es t
he
eye
when
loo
king
at
the
grap
his
that
while
expo
rts
hav
e don
e ve
ryw
ell
grow
ing
dra
mat
ical
ly b
etw
een
199
7an
d 2
00
1, d
omes
tic
sale
s ar
e dec
reas
ing.
In 1
99
7 d
omes
tic
sale
s st
ood a
t £
8,0
00
,00
0.
Bet
wee
n 19
98
and
19
99
they
dro
pped
to
£7
,50
0,0
00
. T
his
los
s w
as f
ollo
wed
by
a ti
mid
gro
wth
in
20
00
,w
ith s
ales
ris
ing
to £
8,0
00
,00
0 a
gain
. H
owev
er,
this
ten
den
cy s
oon
stop
ped
and in
20
01
sale
s fe
ll to
£6
,00
0,0
00
.
Expo
rts
onth
e co
ntra
ry,
hav
e bee
n co
nsta
ntly
gro
win
g.
The
yra
pidly
ros
efr
om £
5,0
00
,00
0 in
199
8-1
99
9 t
o ne
arly
£10
,00
0,0
00
in
20
01.
Ther
efor
eI
wou
ld r
ecom
men
d L
TG
to
conc
entr
ate
excl
usiv
ely
on f
orei
gncu
stom
ers.
Scri
pt
C
There
is o
ne m
inor
conte
nt
om
issio
n -
no s
tart
ing f
igure
for
export
s.
Ap
art
fro
m t
hat
the a
nsw
er
isw
ell-
org
an
ised
with
cle
ar
para
gra
phin
g,
intr
oduction a
nd c
onclu
sio
n.
There
are
fe
w e
rrors
an
d a
reasona
bly
good r
ang
e o
f la
ngu
age.
Ba
nd
4
Wri
tin
g S
am
ple
Sc
rip
ts
Pa
rt 1
Scri
pt
A
LTG S
ALES
1.T
erm
s of
Ref
eren
ces
As
an s
tuden
t in
Par
k La
ne C
olle
ge,
I hav
e bee
n as
ked t
ow
rite
are
port
com
pari
ng t
he
two
sets
of
sale
s (d
omes
tic
and e
xpo
rt s
ales
)fr
om 1
99
7 t
o 2
00
1.
2.
Proc
eedin
gsI
hav
e us
ed a
grap
h co
mpa
ring
th
e ex
port
sa
les
of
LTG
wit
h
its
dom
esti
c sa
les
from
19
97
to
20
01.
3.
Fin
din
gsD
omes
tic
Sal
es:
Fro
m
199
7
till
199
8
sale
s ke
pt
stea
dly
in
£
8,
dec
reas
ing
in 1
99
9 t
o £
7,
then
inc
reas
e st
eadly
to
£8
in
20
00
and
then
aga
in d
ecre
ase
to £
6 in
20
01.
Expo
rt S
ales
: I
n 19
98
inc
reas
e sh
arpl
y fr
om £
2 t
o £
5,
stea
dly
till
199
9 a
nd t
hen
inc
reas
e dra
mat
ical
ly t
o £
9 in
20
01.
4.
Con
clus
ion
Dom
esti
c S
ales
has
kep
t st
eadly
in
£8
, al
teri
ng i
ts s
ales
£1
up a
nddow
n, a
lthou
gh f
rom
20
00
-20
01
dec
reas
ed t
o £
6.
Expo
rt
Sal
es
has
in
crea
sed
shar
ply
thro
ugh
all
that
ye
ars
(19
97
-2
00
1).
Scri
pt
A
This
task
fails
to r
each a
n a
deq
uate
sta
ndard
on b
oth
task a
chie
vem
ent
and l
angu
age
. I
t la
cks
range a
nd
the
ap
pro
pri
ate
la
ngu
age f
or
a d
escri
ption
of
a g
rap
h.
Ba
nd
2
Scri
pt
B
Gen
eral
ly s
peak
ing,
bet
wee
n 19
97
and
20
01,
the
amou
nt o
f LT
G s
ales
hav
ele
velle
d o
ff, pa
ssin
g fr
om £
10 m
illio
ns t
o ne
arly
£15
mill
ions
. B
ut d
omes
tic
and E
xpo
rt s
ales
hav
e ev
olve
d d
iffe
rent
ly.
As
the
com
pany
hav
e dev
elop
ped i
ts E
xpo
rt s
ales
, th
e am
ount
of
this
sal
ehav
e in
crea
sed d
ram
atic
ally
, ex
cept
in
199
8 w
hen
it
rem
aine
d c
onst
ant,
fro
m
Appendix 4.3
373
Scri
pt
EQ
uesti
on
3
A P
ropo
sal fo
r Tra
ining
New E
mploy
ees
This
pro
posa
l ou
tlin
es r
elev
ant
trai
ning
for
our
rece
ntly
hir
ed n
ew p
roje
ctm
anag
ers.
T
hei
rdut
ies
cons
ist
of t
akin
g th
e ov
eral
l re
spon
sibili
ty o
f th
epr
ojec
ts, m
eeti
ng t
he
tim
etab
les,
org
anis
ing
the
proj
ect
wor
k an
d m
otiv
atin
gth
e pa
rtic
ipan
ts.
Tra
inin
g at
th
e ve
ry
beg
inni
ng
is
need
ed
to
mak
e su
re
that
th
e ne
wem
ploy
ees
can
focu
s th
eir
effo
rts
effi
cien
tly.
T
he
prop
osed
tra
inin
g w
ould
incl
ude
less
ons
in m
anag
emen
t an
dpr
ojec
t w
ork.
La
ngua
ge l
esso
ns w
ould
impr
ove
our
new
pr
ojec
t m
anag
ers’
ab
ility
to
co
mm
unic
ate
wit
h
fore
ign
proj
ect
part
icip
ants
. Le
sson
s in
man
agem
ent
and p
roje
ct w
ork
are
need
ed t
oun
der
stan
d t
he
dyn
amic
s of
bus
ines
s lif
e an
d c
o-op
erat
ion.
Our
cu
stom
ers
follo
w
very
ca
refu
lly
how
w
e pe
rfor
m
thin
gs
we
hav
epr
omis
ed.
As
a la
st
item
in
th
e pr
opos
ed
trai
ning
w
ould
be
less
ons
inte
amw
ork.
It i
s im
port
ant
to e
mph
asiz
e th
e m
eani
ng o
f go
od t
eam
wor
k fo
rth
e ne
w e
mpl
oyee
s.
A g
ood c
o-op
erat
ion
wit
h c
olle
ague
s an
d s
har
ing
of w
ork
and inf
orm
atio
n bri
ngs
usua
lly t
he
bes
t re
sult
s, t
oo.
Aft
erth
ebas
ic t
rain
ing
prop
osed
in
this
pap
er,
it w
ould
be
good
if
we
coul
dfo
llow
, ho
wth
e ne
wem
ploy
ees
are
able
to
us
e in
th
eir
wor
k th
e ne
wkn
owle
dge
ga
ined
in
th
e tr
aini
ng.
F
or
this
pu
rpos
e w
e co
uld
arra
nge
afe
edbac
k m
eeti
ng a
bou
t tw
o m
onth
s af
ter
the
bas
ic t
rain
ing.
Sum
mary
– T
raining
needed
-m
anag
emen
t &
pro
ject
wor
k-
lang
uage
s-
team
wor
king
ski
lls.
Scri
pt
E
There
is a
reaso
nab
le a
chie
vem
ent
of
the t
ask,
assum
ing t
ha
t H
R k
no
ws
wh
ich d
epart
ment
has
‘recently
hir
ed n
ew
pro
ject m
anagers
’. L
ang
uag
e a
nd v
ocab
ula
ryare
adeq
uate
.
Ba
nd
3
Pa
rt 2
Scri
pt
DQ
uesti
on
2
Intro
duc
tion
The
aim
of t
his
repo
rt i
s to
des
crib
e th
e ac
ciden
t th
at h
as r
ecen
tly
take
npl
ace
in o
uror
gani
sato
n an
d t
ose
e if
any
mea
sure
s ca
n be
take
n to
pre
vent
sim
ilar
acci
den
ts h
appe
ning
.
Findings
The
key
find
ings
are
out
lined
bel
ow.
It w
as f
ound
that
rec
ent
dat
es r
elat
ed t
o ap
poin
tmen
ts w
ith c
lient
s, f
inal
cont
ract
s th
at h
ad t
o be
sign
ed a
nd s
ever
al inv
oice
sth
at h
ad t
o be
sent
, had
bee
n m
isse
d.
This
was
due
to
the
fact
that
our
com
pute
r ag
enda
stop
ped
wor
king
for
sev
eral
day
s la
stw
eek.
This
was
pro
bab
ly a
s a
resu
lt o
f ei
ther
a t
echni
cal
prob
lem
of
the
com
pute
rsy
stem
its
elf
orju
st a
n el
ectr
icit
y bla
ck o
ut.
Ther
efor
e, t
he
dep
artm
ent
has
alr
eady
calle
d s
ome
tech
nici
ans
to h
ave
ever
ythin
g ch
ecke
d.
Con
clus
ion
It is
clea
r th
at o
uror
gani
sati
on c
an n
ot m
iss
impo
rtan
t dat
es j
ust
bec
ause
it
keep
s al
l th
e in
form
atio
n in
a c
ompu
ter
syst
em.
It i
s ob
viou
s th
at s
ome
mea
sure
s ne
ed t
o be
take
n to
avo
id t
hat
sim
ilar
acci
den
ts w
ill h
appe
n ag
ain
in t
he
futu
re.
Reco
mmend
ation
It is
ther
efor
e su
gges
ted t
hat
our
em
ploy
ee r
egis
ter
all th
e in
form
atio
n an
dth
e dat
es n
ot o
nly
in t
he
com
pute
r ag
enda
but
in
a pa
pery
one
too
.
In t
his
way
, if
som
ethin
g hap
pens
to
the
com
pute
r sy
stem
, ou
ror
gani
sati
onw
ill b
e ab
le a
nyw
ay t
o ga
ther
the
info
rmat
ion
it n
eeds
and n
ot t
o lo
ose
the
dat
es!
Scri
pt
D
This
answ
er
does
not
ade
qua
tely
dea
l w
ith t
he t
ask s
et, d
escri
bin
g a
n i
ncid
ent
rath
er
than a
naccid
ent.
Ba
nd
2
Appendix 4.3
374
Scri
pt
F
This
answ
er
sho
ws a
goo
d r
ea
lisa
tio
n o
f th
e t
ask s
et
with
all
majo
rconte
nt
poin
tsin
clu
de
d.
There
are
few
err
ors
, but
the r
ange
of
langu
age c
ould
be i
mpro
ved
up
on,
part
icu
larl
y th
e o
ver-
relia
nce
on ‘w
ill’.
Ba
nd
4
Scri
pt
FQ
uesti
on
4
Mr.
Bla
ckS
t. J
ohn’
s H
ospi
tal
‘Addre
ss’
4 J
une
02
Dea
r M
r. B
lack
,
Osl
o Ph
arm
aceu
tica
l T
rade
Fai
r
I am
wri
ting
to
invi
te y
ou t
o a
trad
e fa
ir t
akin
g pl
ace
in O
slo
in J
uly
this
year
.
The
Osl
o Ph
arm
aceu
tica
l T
rade
Fai
r is
arr
ange
d e
very
fif
th y
ear
and i
s on
eof
th
e gr
eate
st
trad
e fa
irs
run
in
Sca
ndin
avia
. T
he
mos
t im
port
ant
phar
mac
euti
cal
com
pani
es
in
Nor
ther
n E
urop
e w
ill
atte
nd
this
fa
ir,
and
Am
ersh
am
Hea
lth
will
hav
e a
stan
d
ther
e.
T
he
dif
fere
nt
com
pani
esat
tend
ing
will
pre
sent
thei
r pr
oduc
ts.
The
fair
will
be
arra
nged
fro
m 5
-7 J
uly
20
02
, st
arti
ng a
t 10
am
. T
he
even
tw
ill t
ake
plac
e at
‘Sj
lyst
con
fere
nce
cent
re’ j
ust
outs
ide
Osl
o.
Am
ersh
am H
ealt
h A
S w
ill p
rese
nt a
nd d
emon
stra
te t
he n
ew p
har
mac
euti
cal
prod
ucts
of t
he
com
pany
. W
ew
ill f
ocus
on
our
ultr
asou
nd c
ontr
ast
med
ia.
In a
ddit
ion,
lec
ture
rsre
late
d t
ore
sear
ch p
roduc
ts w
ill b
e hel
d.
We
will
aim
to g
ive
det
aile
d i
nfor
mat
ion
abou
t ou
r pr
oduc
ts.
It
shou
ld b
e of
par
ticu
lar
inte
rest
for
radio
logi
sts
and m
edic
al d
octo
rs t
ose
e how
ou
r ul
tras
ound
cont
rast
med
ia,
Opt
ison
®,
can
faci
litat
e in
vest
igat
ions
of
hea
rt p
erfu
sion
.T
his
alo
ne s
hou
ld m
ake
the
fair
wor
th v
isit
ing.
We
wou
ld a
ppre
ciat
e if
you
or
any
of y
our
colle
ague
s w
ould
lik
e to
att
end
this
eve
nt.
We
wou
ld b
e gr
atef
ul if
you
coul
d r
espo
nd b
y fa
x o
r e-
mai
l by
25
Jun
e 0
2.
Plea
se s
ee t
he
lett
er h
eadin
g fo
r det
ails
.
I lo
ok v
ery
muc
h f
orw
ard t
o hea
ring
fro
m y
ou.
You
rs s
ince
rely
Mar
iann
e W
.Wul
ffA
mer
sham
Hea
lth A
S.
Appendix 4.3
375
Superv
isor:
HI
GH
ER
BE
C H
igh
er
Read
ing
An
sw
er
Sh
eet
00
00 1 2 3 4 5 6 7 8 9
1 2 3 4 5 6 7 8 9
1 2 3 4 5 6 7 8 9
1 2 3 4 5 6 7 8 9
Ca
nd
ida
te N
am
eIf
no
t a
lre
ad
y p
rin
ted
, w
rite
na
me
in C
AP
ITA
LS
an
d c
om
ple
te t
he
Can
did
ate
No
. g
rid
(in
pen
cil).
Can
did
ate
’s S
ign
atu
re
Exam
inati
on
Tit
le
Cen
tre
If t
he
ca
nd
ida
te is A
BS
EN
To
r h
asW
ITH
DR
AW
Nsh
ad
e h
ere
Can
did
ate
No
.
Cen
tre N
o.
Exam
inati
on
Deta
ils
9Part
2
�
Tu
rn o
ver
for
Part
s 5
an
d 6
Ins
tru
cti
on
sU
se
a P
EN
CIL
(B
or
HB
).
Ru
b o
ut
an
y a
nsw
er
yo
u w
ish
to
ch
an
ge
with
an
era
se
r.
Fo
r P
art
s 1
to
4:
Ma
rk o
ne
bo
x f
or
ea
ch
an
sw
er.
Fo
r e
xa
mp
le:
If y
ou
th
ink C
is t
he
rig
ht
an
sw
er
to t
he
qu
estio
n,
ma
rk y
ou
r a
nsw
er
sh
ee
t like
th
is:
Fo
r P
art
s 5
an
d 6
:
Write
yo
ur
an
sw
er
cle
arly in
CA
PIT
AL
LE
TT
ER
S.
Write
on
e le
tte
r in
ea
ch
bo
x.
Fo
r e
xa
mp
le:
AB
C0
0
10
11
12
13
14
15P
art
3
16
17
18
19
20
21P
art
4
22
23
24
25
26
27
28
29
30
1Part
1
2 3 4 5 6 7 8
BE
C H
- R
DP
46
2/3
62
AB
CD
EF
GH
AB
CD
EF
GH
AB
CD
EF
GH
AB
CD
EF
GH
AB
CD
EF
GH
AB
CD
EF
GH
AB
CD
AB
CD
AB
CD
AB
CD
AB
CD
AB
CD
AB
CD
AB
CD
AB
CD
AB
CD
AB
CD
AB
CD
AB
CD
AB
CD
AB
CD
AB
CD
AB
CD
E
AB
CD
E
AB
CD
E
AB
CD
E
AB
CD
E
AB
CD
E
AB
CD
E
AB
CD
E
Appendix 4.3
376
Superv
isor:
HI
GH
ER
BE
C H
igh
er
Lis
ten
ing
An
sw
er
Sh
eet
00
00 1 2 3 4 5 6 7 8 9
1 2 3 4 5 6 7 8 9
1 2 3 4 5 6 7 8 9
1 2 3 4 5 6 7 8 9
Can
did
ate
Nam
eIf
no
t a
lre
ad
y p
rin
ted
, w
rite
na
me
in C
AP
ITA
LS
an
d c
om
ple
te t
he
Can
did
ate
No
. g
rid
(in
pen
cil).
Can
did
ate
’s S
ign
atu
re
Exam
inati
on
Tit
le
Cen
tre
If t
he
ca
nd
ida
te is A
BS
EN
To
r h
asW
ITH
DR
AW
Nsh
ad
e h
ere
Can
did
ate
No
.
Cen
tre N
o.
Exam
inati
on
Deta
ils
Instr
ucti
on
sU
se
a P
EN
CIL
(B
or
HB
).
Ru
b o
ut
an
y a
nsw
er
yo
u w
ish
to
ch
an
ge
with
an
era
se
r.
Continue o
n the o
ther
sid
e o
f th
is s
heet
�
Fo
r P
art
s 2
an
d 3
:
Ma
rk o
ne
bo
x f
or
ea
ch
an
sw
er.
Fo
r e
xa
mp
le:
If y
ou
th
ink C
is t
he
rig
ht
an
sw
er
to t
he
qu
estio
n,
ma
rk y
ou
r a
nsw
er
sh
ee
t lik
e t
his
:
AB
C0
Fo
r P
art
1:
Write
yo
ur
an
sw
er
cle
arly in
CA
PIT
AL
LE
TT
ER
S.
Write
on
e le
tte
r o
r n
um
be
r in
ea
ch
bo
x.
If t
he
an
sw
er
ha
s m
ore
th
an
on
e w
ord
, le
ave
on
e b
ox e
mp
ty b
etw
ee
n w
ord
s.
Fo
r e
xa
mp
le:
Part
1
1
01
1
2
01
2
3
01
3
4
01
4
0
BE
C H
- L
DP
46
4/3
64
Part
5
32
31
01
01
34
01
01
35
01
31
32
33
34
35
36
01
37
01
36
37
38
01
39
01
38
39
40
01
40
33 P
art
6
42
41
01
01
44
01
01
45
01
41
42
43
44
45
46
01
47
01
46
47
48
01
49
01
48
49
50
01
50
43
51
01
51
52
01
52
Appendix 4.3
377
Fe
ed
ba
ck
Fo
rm
BE
C H
igh
er
Exam
inati
on
Rep
ort
Ju
ne 2
002
We a
re inte
reste
d in h
eari
ng y
our
vie
ws o
n h
ow
usefu
l th
is r
eport
has b
een.
We w
ou
ld b
e m
ost gra
tefu
l if
you c
ould
bri
efly
answ
er
the f
ollo
win
g q
uestions
and r
etu
rn a
ph
oto
co
py
of
this
pa
ge t
o th
e f
ollo
win
g a
ddre
ss:
BE
C C
o-o
rdin
ato
rC
am
bri
dge E
SO
L1 H
ills R
oad
Cam
bri
dge
CB
1 2
EU
UK
Fax: 44 1
223
46
027
8
1.
Ple
ase d
escri
be y
our
situa
tio
n: (e
.g.
EF
L t
eacher,
Dir
ecto
rof
Stu
die
s,
Exam
inatio
ns O
ffic
er,
Local
Secre
tary
, etc
)
2.
Have y
ou p
repare
d c
an
did
ate
s f
or
BE
C H
igher?
3.
Do y
ou p
lan to
pre
pare
ca
nd
idate
s f
or
BE
C H
igher
in the
futu
re?
4.
Ho
w h
ave
you u
sed t
his
rep
ort
? (
e.g
. to
pro
vid
e f
eed
back to o
ther
teach
ers
, fo
r exam
ination
pra
ctice, e
tc.)
5.
Whic
h p
art
s o
f th
is r
eport
did
you f
ind m
ost usefu
l?
6.
Whic
h p
art
s w
ere
no
t so u
sefu
l?
7.
What extr
a info
rmation w
ould
yo
u lik
e to s
ee inclu
de
d in
th
is r
ep
ort
?
8.
Your
nam
e: (O
ption
al)
Centr
e/S
cho
ol:
Thank y
ou.
Part
1 c
on
tin
ue
d
16
15
175
01
5
AB
CD
EF
GH
AB
CD
EF
GH
AB
CD
EF
GH
Part
2 -
Task O
ne
23
AB
C
14
13
AB
CD
EF
GH
AB
CD
EF
GH
Part
3
25
24
26
AB
C
AB
C
AB
C
27
28
AB
C
AB
C
29
AB
C
6
01
6
7
01
7
8
01
8
9
01
9
10
01
10
11
01
11
12
01
12
Part
2 -
Task T
wo
21
20
22
AB
CD
EF
GH
AB
CD
EF
GH
AB
CD
EF
GH
19
18
AB
CD
EF
GH
AB
CD
EF
GH
30
AB
C
Part
1 -
Co
nvers
ati
on
Tw
o
Part
1 -
Co
nvers
ati
on
Th
ree
16
15
175
01
5
AB
CD
EF
GH
AB
CD
EF
GH
AB
CD
EF
GH
Part
2 -
Secti
on
On
e
23
AB
C
14
13
AB
CD
EF
GH
AB
CD
EF
GH
Part
3
25
24
26
AB
C
AB
C
AB
C
27
28
AB
C
AB
C
29
AB
C
6
01
6
7
01
7
8
01
8
9
01
9
10
01
10
11
01
11
12
01
12
Part
2 -
Secti
on
Tw
o
21
20
22
AB
CD
EF
GH
AB
CD
EF
GH
AB
CD
EF
GH
19
18
AB
CD
EF
GH
AB
CD
EF
GH
30
AB
C
Fe
ed
ba
ck
Fo
rm
BE
C V
an
tag
e E
xam
inati
on
Rep
ort
Ju
ne 2
002
We a
re inte
reste
d in h
eari
ng y
our
vie
ws o
n h
ow
usefu
l th
is r
eport
has b
een.
We w
ou
ld b
e m
ost gra
tefu
l if
yo
u c
ould
bri
efly a
nsw
er
the f
ollo
win
g q
uestions
and r
etu
rn a
ph
oto
co
py
of
this
pa
ge t
o th
e f
ollo
win
g a
ddre
ss:
BE
CR
ep
ort
s C
o-o
rdin
ato
rC
am
bri
dge E
SO
L1 H
ills R
oad
Cam
bri
dge
CB
1 2
EU
UK
Fax: 44 1
223
46
027
8
1.
Ple
ase d
escri
be y
our
situa
tio
n: (e
.g.
EF
L t
eacher,
Dir
ecto
rof
Stu
die
s,
Exam
inatio
ns O
ffic
er,
Local
Secre
tary
, etc
)
2.
Have y
ou p
repare
d c
an
did
ate
s f
or
BE
CV
an
tag
e?
3.
Do y
ou p
lan to
pre
pare
cand
idate
s f
or
BE
CV
an
tag
e in th
e f
utu
re?
4.
Ho
wh
ave
yo
uused t
his
rep
ort
? (
e.g
. to
pro
vid
e f
eed
back
to o
ther
teach
ers
, fo
rexam
ination
pra
ctice, e
tc.)
5.
Whic
h p
art
s o
fth
is r
eport
did
you f
ind m
ost usefu
l?
6.
Whic
h p
art
s w
ere
not so u
sefu
l?
7.
What extr
a info
rmation w
ould
yo
u lik
e to s
ee inclu
de
d in
th
is r
ep
ort
?
8.
Your
nam
e: (O
ption
al)
Centr
e/S
cho
ol:
Tha
nk y
ou
.
378
Additional information on testsof language for businesspurposes
Tests of European language for business purposes: ALTE members
* BULATS is currently available in English, French, German and Spanish
CEFLevels
Italian English French German Spanish
C2 DSEC(DiplômeSupérieurd’EtudesCommerciales)
DEN (Diplomade Espanol de losNegocios)
C1 CIC(Certificato diItalianoCommerciale)advanced
BEC(BusinessEnglishCertificate)Advanced
PWD(PrüfungWirtschaftsdeutschInternational)
B2 BEC(BusinessEnglishCertificate)Intermediate
ZDf B(ZertifikatDeutsch fürden Beruf)
CEN(Certificado deEspanol de losNegocios)
B1 CIC(Certificato diItalianoCommerciale)Intermediate
BEC(BusinessEnglishCertificate)Preliminary
A2
A1
Contact information for business language test developers
Additional information on tests of language for business purposes
379
Organisation Address Website
Pitmanqualifications
City and Guilds PitmanQualifications1 Giltspur StreetLondon EC1A 9DD United Kingdom
Fax: +44 (0) 20 7294 3502
www.pitmanqualifications.com
LCCIEB LCCIEB Corporate Headquarters112 Station RoadSidcup KentDA15 7BJUnited Kingdom
Phone: +44 (0) 20 8309 3000Fax: +44 (0) 20 8302 4169
www.lccieb.com/Lcci/Home/Index.asp
TOEIC TOEIC Service International TOEIC Testing Program Educational Testing Service Rosedale Road Princeton, NJ 08541, USA
Phone: +1 (609) 734-1540 Fax: +1 (609) 734 1560
www.toeic.com/index.htm
Cambridge ESOL University of CambridgeESOL Examinations1 Hills RoadCambridgeCB1 2EUUnited Kingdom
Phone: +44 (0) 1223 553355Fax: +44 (0) 1223 460278
www.cambridge-efl.org/index.html
JETRO 2-5, Toranomon 2-chome, Minato-ku,Tokyo 105-8466
Tel: 03-3582-5511Fax: 03-3587-0219
www.jetro.go.jp/it/e/bj/index.html
AllianceFrançaiseCentreinternationald’étudespédagogiques
Alliance Française101, Bd Raspail75270 Paris Cedex 06France
Tel.: +33-1-45 44 38 28Fax: +33-1-45 49 15 82E-mail:[email protected]
www.alliancefr.org/
380
Additional information on tests of language for business purposes
Organisation Address Website
The Goethe-Institut
Goethe-Institut Inter NationesHelene-Weber-Allee 1München80637Germany
Tel +49 89 15921 382Fax +49 89 15921 608E-mail [email protected]
www.goethe.de/dll/prf/pba/pwd/deindex.htm
Università perStranieri, Perugia
University for Foreigners –Perugia Palazzo Gallenga Piazza Fortebraccio, 4 06122 Perugia – Italy
Tel +39/075/5746467 Fax +39/075/5746456 E-mail: [email protected]
www.unistrapg.it
InstitutoCervantes andUniversidad deSalamanca
Instituto CervantesColegio del ReyC/ Libreros, 23Madrid 28801 Alcalá de HenaresSpain
Tel +34-91-745 3334Fax +34-91-745 0058E-mail: [email protected]
www.cervantes.es
381
References
Abdul-Raof, A H (2002) The production of a performance rating scale: An alternativemethodology, Unpublished PhD dissertation, University of Reading.
APA (1999) The Standards for Educational and Psychological Testing, WashingtonDC: American Psychological Association.
Alderson, J C (1998) DIALANG, paper presented at the Language Testing Forum,Swansea, November 1998.
Alderson, J C and Urquhart, A H (1984) ESP Tests: The problem of student backgrounddiscipline, in Culhane, T, Klein-Braley, C and Stevenson, D K (Eds) Practice andProblems in Language Testing, Essex: University of Essex, 1–13.
Alderson, J C and Urquhart, A H (1985) The effect of students’ academic discipline ontheir performance in ESP reading tests, Language Testing 2 (2), 192–204.
Alderson, J C and Urquhart, A H (1988) This test is unfair: I’m not an economist, inCarrell, P, Devine, J and Eskey, D (Eds) Interactive Approaches to SecondLanguage Reading, Cambridge: Cambridge University Press, 168–182.
Alwright, J and Alwright, R (1977) An approach to the teaching of medical English, inHolden, S (Ed.) English for Specific Purposes, Modern English Publications, 58–62.
Anastasi, A (1988) Psychological Testing, New York: Macmillan.
Ashton, M (2003) The change process at the paper level. Paper 1, Reading in Weir, C Jand Milanovic, M (2002) (Eds) Continuity and Innovation: Revising the CambridgeProficiency in English Examination 1913–2002, Cambridge: University ofCambridge Local Examinations Syndicate and Cambridge University Press,121–174.
Bachman, L (1990) Fundamental Considerations in Language Testing, Oxford: OxfordUniversity Press.
Bachman, L and Palmer, A (1996) Language testing in practice, New York: OxfordUniversity Press.
Bachman, L F, Lynch, B and Mason, M (1995) Investigating Variability in Tasks andRater Judgements in a Performance Test of Foreign Language Speaking, LanguageTesting 12 (2), 238–257.
Bachman, L F, Davidson, F, Ryan, K and Choi, I–C (1995) An investigation into thecomparability of two tests of English as a Foreign Language: The Cambridge–TOEFL comparability study, Cambridge: University of Cambridge LocalExaminations Syndicate and Cambridge University Press.
Ball, F (2002) Developing wordlists for BEC, Research Notes 8, 10–13.
Barber, C L (1962) Some Measurable Characteristics of Modern Scientific Prose,reprinted in Swales, J (Ed.) 1988 Episodes in ESP, New York: Prentice Hall, 1–16.
References
382
Barratt, N (2003) The change process at the paper level. Paper 3, Use of English, inWeir, C J and Milanovic, M (2002) (Eds) Continuity and Innovation: Revising theCambridge Proficiency in English Examination 1913–2002, Cambridge: Universityof Cambridge Local Examinations Syndicate and Cambridge University Press,237–314.
Beeching, K (1997) French for specific purposes: the case for spoken corpora, AppliedLinguistics 18 (3), 374–394.
Berry, V (1996) Ethical considerations when assessing oral proficiency in pairs, paperpresented at the Language Testing Research Colloquium, Tampere, Finland.
Berry, V (1997) Gender and personality as factors of interlocutor variability in oralperformance tests, paper presented at the Language Testing Research Colloquium,Orlando, Florida, USA.
Biber, D, Conrad, S and Reppen, R (1998) Corpus linguistics: investigating languagestructure and use, Cambridge: Cambridge University Press.
Bordeaux, D (2002) Communicating for Results, Motor Age, 121 (1), 40–42.
Boroughs, R (2003) The change process at the paper level. Paper 4, Listening, in Weir,C J and Milanovic, M (2002) (Eds) Continuity and Innovation: Revising theCambridge Proficiency in English Examination 1913–2002, Cambridge: Universityof Cambridge Local Examinations Syndicate and Cambridge University Press,315–364.
Brindley, G (1984) The role of needs analysis in adult ESL programme Design, inJohnson, R K (Ed.) The Second Language Curriculum, Cambridge: CambridgeUniversity Press, 63–79.
Brooks, L (2003) Converting an Observation Checklist for use with the IELTS speakingpaper, Research Notes 11, 20–21, Cambridge ESOL.
Brown, A (1995) The Effect of Rater Variables in the Development of an OccupationSpecific Language Performance Test, Language Testing, 12 (1), 1–15.
Brown, A (1998) Interviewer style and candidate performance in the IELTS oralinterview, paper presented at the Language Testing Research Colloquium, MontereyCA.
Brown, A and Lumley, T (1997) Interviewer Variability in Specific-Purpose LanguagePerformance Tests, in Huhta, A Kohonen, V, Kurki-Suonio, L and Luoma, S (Eds)Current Developments and Alternatives in Language Assessment, Jyväskylä:University of Jyväskylä and University of Tampere, 137–150.
Brown, J D (1996) Testing in Language Programs, Upper Saddle River, NJ: PrenticeHall Regents.
Brown, J D (1997) Computers In Language Testing: Present Research And SomeFuture Directions, Language Learning & Technology, 1 (1), 44–59.
Brown, W M, Palmeta, B and Moore, B (2003) Are there nonverbal cues tocommitment? An exploratory study using the zero-acquaintance video presentationparadigm, Evolutionary Psychology 1, 42–69.
Buck, G (2001) Assessing Listening, Cambridge: Cambridge University Press.
BULATS (Undated/a) BULATS Standard Test, EN00, Cambridge: Cambridge ESOL.
BULATS (Undated/b) BULATS Speaking Test, EN40, Cambridge: Cambridge ESOL.
References
383
BULATS (Undated/c) BULATS Writing Test, EN60, Cambridge: Cambridge ESOL.
Bygate, M (1987) Speaking, Oxford: Oxford University Press.
Cambridge ESOL (2002a) Business English Certificates: BEC PreliminaryExamination Report and Past Examination Papers, Cambridge: Cambridge ESOLExaminations.
Cambridge ESOL (2002b) Business English Certificates: BEC Vantage ExaminationReport and Past Examination Papers, Cambridge: Cambridge ESOL Examinations.
Cambridge ESOL (2002c) Business English Certificates: BEC Higher ExaminationReport and Past Examination Papers, Cambridge: Cambridge ESOL Examinations.
Canale, M and Swain, M (1980) Theoretical Bases of Communicative Approaches toSecond Language Teaching and Testing, Applied Linguistics 1, 1–47.
Carver, D (1983) Some propositions about ESP, ESP Journal 2, 131–137.
Center for Applied Linguistics (2001) Self-instructional Rater Training Kits forSimulated Oral Proficiency Interviews, Washington DC: Center for AppliedLinguistics.
Chalhoub-Deville, M (Ed.) (1999) Issues in computer adaptive testing of readingproficiency, New York: Cambridge University Press.
Chalhoub-Deville, M (2001) Language Testing and Technology: Past And Future,Language Learning and Technology 5 (2), 95–98.
Chalhoub-Deville, M (2003) Second language interaction: current perspectives andfuture trends, Language Testing 20 (4), 369–383.
Chapelle, C (1998) Construct definition and validity enquiry in SLA Research, inBachman, L and Cohen, A (Eds) Interfaces between second language acquisitionand language testing research, New York: Cambridge University Press, 32–70.
Chia, R C, Allred, L T, Grossnickle, W F and Lee, G W (1998) Effects of attractivenessand gender on the perception of achievement-related variables, Journal of SocialPsychology 138 (4), 471–477.
CIC (2003a) Certificazione della conoscenza dell’italiano commerciale: Handbook,Perugia: Universiti per Stranreiri.
CIC (2003b) Certificazione della conoscenza dell’italiano commerciale:http://www.cvcl.it/canale.asp?id=34 (website accessed September 2005).
Clapham, C M (1996) The Development of IELTS: A Study of the Effect of BackgroundKnowledge on Reading Comprehension, Cambridge: Cambridge University Press.
Congdon, P J and McQueen, J (2000) The stability of rater severity in large-scaleassessment programs, Journal of Educational Measurement 37, 163–178.
Coniam, D (2001) The use of audio or video comprehension as an assessmentinstrument in the certification of English language teachers: A case study, System 29,1–14.
Cumming, A (2001) ESL/EFL instructors’ practices for writing assessment: specificpurposes or general purposes? Language Testing 18 (2), 207–224.
Davies, A (2001) The logic of testing languages for specific purposes, LanguageTesting 18 (2), 133–147.
Dickerson, L (1975) The learner’s interlanguage as a system of variable Rules, TESOLQuarterly 9 (4), 401–407.
References
384
Douglas, D (2000) Assessing languages for specific purposes, Cambridge: CambridgeUniversity Press.
Douglas, D (2001) Language for specific purposes assessment criteria: where do theycome from? Language Testing 18 (2), 171–185.
Dudley-Evans, R and St John, M J (1996) Report on Business English: A review ofresearch and published teaching materials, TOEIC Research Report #2, PrincetonNJ: Educational Testing Services.
Educational Testing Services (1986) TOEIC Users Guide, Princeton NJ: EducationalTesting Services.
Educational Testing Services (1998) TOEIC Technical Manual, Princeton NJ:Educational Testing Services.
Educational Testing Services (1999) TOEIC Users Guide, Princeton NJ: EducationalTesting Services.
Educational Testing Services (2002) TOEIC: Examinees Handbook, Princeton: TheChauncey Group International.
Educational Testing Services (2003a) Test de français international: TOEIC:www.toeic.-europe.com/pages/eng/tfi.htm
Educational Testing Services (2003b) Test de français international: TOEIC–Europe:www.toeic-europe.com/pages/eng/tfi.htm#description
Educational Testing Services (2001) TSE & SPEAK Score Users Guide: 2001–2002Edition, Princeton NJ: Educational Testing Services.
Elder, C (2001) Assessing the language proficiency of teachers: are there any bordercontrols? Language Testing 18 (2), 149–170.
Ellis, R (1989) Sources of Intra-Learner Variability in Language, in Gass, S, Madden,C, Preston, D and Selinker, L (Eds) Variation in Second Language Acquisition, vol.2: Psycholinguistic Issues, Clevedon PA: Multilingual Matters, 22–45.
Engelhard, G Jr. (1994) Examining rater error in the assessment of written compositionwith a many-faceted Rasch model, Journal of Educational Measurement 31, 93–112.
Feldt, L S and Brennan, R L (1989) Reliability, in Linn, R L (Ed.) EducationalMeasurement, New York: American Council on Education and Macmillan, 105–146.
ffrench, A (2003) The change process at the paper level, Paper 5, Speaking, in Weir, C Jand Milanovic, M (Eds) Continuity and Innovation: Revising the CambridgeProficiency in English Examination 1913–2002, Cambridge: University ofCambridge Local Examinations Syndicate and Cambridge University Press,367–472.
Field, J (2000) CPE Revision Paper 4: Listening, Unpublished internal UCLES report.
Fisher, A G (1994) Development of a functional assessment that adjusts abilitymeasures for task simplicity and rater leniency, in Wilson, M (Ed.) Objectivemeasurement: Theory into practice, Vol II. Norwood, New Jersey: Ablex PublishingCorporation, 145–175.
Foot, M C (1999) Relaxing in Pairs, ELT Journal 53 (1), 36–41.
Fulcher, G (1996) Testing tasks: issues in task design and the group oral, LanguageTesting 13 (1), 23–51.
References
385
Gledhill, C (2000) The Discourse Function of Collocation in Research ArticleIntroductions, English for Specific Purposes 19, 115–135.
Gutteridge, M (2003) Assistive technologies for students with special needs, ResearchNotes 12, 15.
Hawkey, R (1978) English for Special Purposes, London: British Council EnglishTeaching Centre.
Hawkey, R (1982) An investigation of inter-relationships between cognitive/affectiveand social factors and language learning, A longitudinal study of 27 overseasstudents using English in connection with their training in the United Kingdom,Unpublished PhD dissertation, Department of English for Speakers of OtherLanguages, Institute of Education, London.
Hawkey, R (2001) Towards a common scale to describe L2 writing performance,Research Notes 5, 9–13.
Hawkey, R (2004) The development of CELS: A modular approach to testing EnglishLanguage Skills, Cambridge: University of Cambridge Local ExaminationsSyndicate and Cambridge University Press.
Hawkey, R and Barker, F (2004) Developing a common scale for the assessment ofwriting, Assessing Writing 9 (2), 121–159.
Holden, S (Ed.) (1977) English for Specific Purposes, Modern English Publications.
Hughes, A (1989) Language Testing for Teachers, Cambridge: Cambridge UniversityPress.
Hüllen, W (1981a) Movements on Earth and in the Air: a study of certain verbsoccurring in the language of international pilots, ESP Journal 1 (2), 141–153.
Hüllen, W (1981b) The teaching of English for special purposes: a linguistic view, inFreudenstein, R et al (Eds) Language Incorporated: Teaching foreign languages inindustry, Pergamon and Max Hueber Verlag, 57–71.
Hutchinson, T and Walters, A (1987) English for Specific Purposes: A learning-centredapproach, Cambridge: Cambridge University Press.
Hymes, D H (1972) On Communicative Competence, in Pride, J B and Holmes, J (Eds)Sociolinguistics: selected readings, Harmondsworth, Middlesex: Penguin, 269–293.
Jacoby, S and McNamara, T F (1999) Locating competence, English for SpecificPurposes 18 (3), 213–241.
JETRO (2003a) The JETRO JRLT, Introduction, Tokyo: JETRO:www.jetro.go.jp/it/e/bj/guide/index.html
JETRO (2003b) Sample Questions from the Listening and Reading ComprehensionTest, Tokyo: JETRO: www.jetro.go.jp/it/e/bj/guide/test-sample.html
JETRO (2003c) The JETRO JOCT, Introduction, Tokyo: JETRO:www.jetro.go.jp/it/e/bj/guide/joct.html
Johns, A (1980) Cohesion in written business discourse: some contrasts, ESP Journal 1(1), 35–44.
Jones, N (2000) Background to the revised ALTE ‘Can Do’ project and the revisedCommon European Framework, Research Notes 2, 11–14.
Jones, N (2001a) The Can Do Project, paper presented at the IATEFL Conference,Brighton, April 2001.
References
386
Jones, N (2001b) The ALTE ‘Can Do’ Project and the role of measurement inconstructing a proficiency framework, Research Notes 5, 5–8.
Just, M A, Carpenter, P A, Keller, T A, Emery, L, Zajac, H and Thulborn K R (2001)Interdependence of Nonoverlapping Cortical Systems in Dual Cognitive Tasks,NeuroImage 14 (2), 417–426.
Kennedy, C and Bolitho, R (1984) English for Specific Purposes, London: MacmillanPress Ltd.
Kenyon, D (1997) Further research on the efficacy of rater self-training, in Huhta, A,Kohonen, V, Kurki-Suonio, L and Luoma, S (Eds) Current developments andalternatives in language assessment: Proceedings of LTRC 96, Jyvaskyla, Finland:University of Jyvaskyla, 257–273.
Labov, W (1963) The social motivation of sound change, Word 19, 273–307.
Lackstrom, J, Selinker, L and Trimble, L (1973) Technical rhetorical principles andgrammatical choice, TESOL Quarterly 7, 127–136.
Lado, R (1961) Language Testing, New York: McGraw-Hill.
Lamprianou, I and Pillas, K (2003) The Effect of Markers and Optional Questions onthe Results of High-Stakes Exams, paper presented at the AERA Conference,Chicago, April 2003.
Lazaraton, A (1992) The structural organisation of a language interview: aconversational analytic perspective, System 20 (3), 373–386.
Lazaraton, A (1996) Interlocutor support in Oral Proficiency Interviews: the case ofCASE, Language Testing 13 (2), 151–172.
Loevinger, J (1954) The attenuation paradox in test theory, Psychological Bulletin 51,493–504.
LCCIEB (1972) The non-specialist use of foreign languages in industry and commerce,Sidcup: London Chamber of Commerce and Industry Examinations Board.
LCCIEB (2001a) English for Business, Level 3: Extended Syllabus, London: LondonChamber of Commerce and Industry Examinations Board.
LCCIEB (2001b) German for Business, Level 3: Extended Syllabus, London: LondonChamber of Commerce and Industry Examinations Board.
LCCIEB (2001c) Spanish for Business, Level 3: Extended Syllabus, London: LondonChamber of Commerce and Industry Examinations Board.
LCCIEB (2001d) French for Business, Level 3: Extended Syllabus, London: LondonChamber of Commerce and Industry Examinations Board.
LCCIEB (Undated) English for Business, Preliminary Level: Sample Paper, London:London Chamber of Commerce and Industry Examinations Board, available atwww.lccieb.org.uk/lcciebweb/downloads/EFBPrelimSamplePaper.pdf
Longford, N T (1994) Reliability of essay rating and score adjustment, Journal ofEducational and Behavioral Statistics 19, 171–200.
Lumley, T (1998) Perceptions of language-trained raters and occupational experts in atest of occupational English language proficiency, English for Specific Purposes 17(4), 347–367.
Lumley, T (2000) The process of the assessment of writing performance: the rater’sperspective, Unpublished PhD dissertation, The University of Melbourne, Australia.
References
387
Lumley, T and McNamara, T F (1995) Rater characteristics and rater bias: implicationsfor training, Language Testing 12 (1), 54–71.
Lumley, T and O’Sullivan, B (2001) The Effect of Test-Taker Sex, Audience and Topicon Task Performance in Tape-Mediated Assessment of Speaking, Melbourne Papersin Language Testing 9 (1), 34–55.
Lumley, T, Lynch, B K and McNamara, T F (1994) A new approach to standard-settingin language assessment, Melbourne Papers in Language Testing 3, 19–40.
Lunz, M E and Stahl, J A (1990) Judge consistency and severity across grading periods,Evaluation and the Health Professions 13, 425–444.
Lunz, M E, Wright, B D and Linacre, J M (1990) Measuring the impact of judgeseverity on examination scores, Applied Measurement in Education 3, 331–345.
Luoma, S (2004) Assessing Speaking, Cambridge: Cambridge University Press.
McNamara T F (1997) ‘Interaction’ in second language performance assessment:whose performance, Applied linguistics 18, 446–466.
McNamara, T F (1996) Measuring Second Language Performance, London: Longman.
McNamara, T F and Lumley, T (1997) The effect of interlocutor and assessment modevariables in offshore assessments of speaking skills in occupational settings,Language Testing 14 (2), 140–156.
McNeil, L and Valenzuela, A (2000) The Harmful Impact of the TAAS System ofTesting in Texas: Beneath the Accountability Rhetoric, in Orfield, G and Kornhaber,M L (Eds) Raising Standards or Raising Barriers? New York: The Century PressFoundation.
Milanovic, M and Saville, S (Eds) (1996) Performance Testing, Cognition andAssessment: selected papers from the 15th Language Testing Research Colloquium,Cambridge and Arnhem, Cambridge: University of Cambridge Local ExaminationsSyndicate and Cambridge University Press.
Munby, J (1978) Communicative Syllabus Design, Cambridge: Cambridge UniversityPress.
Myford, C M and Wolfe, E W (2000) Strengthening the Ties that Bind: Improving theLinking Network in Sparsely Connected Rating Designs, TOEFL Technical Report15, Princeton NJ: Educational Testing Services.
Myford, C M and Wolfe, E W (2002) When raters disagree, then what: examining athird-rating discrepancy resolution procedure and its utility for identifying unusualpatterns of ratings, Journal of Applied Measurement 3, 300–324.
Norris, J, Brown, J D, Hudson, T and Yoshioka, J (1998) Designing Second LanguagePerformance Assessments, Technical Report #18, Hawai’i: University of Hawai’iPress.
North, B and Schneider, G (1998) Scaling Descriptors for Language Proficiency Scales,Language Testing 15, 217–262.
North, B (1996) The development of a common framework scale of descriptors oflanguage proficiency based on a theory of measurement, Unpublished PhDdissertation, Thames Valley University.
O’Sullivan, B (1995) Oral Language Testing: Does the Age of the Interlocutor make aDifference? Unpublished MA Dissertation. UK: University of Reading.
References
388
O’Sullivan, B (1999) Revision of the Test of English for Educational Purposes, projectfunded by the Centre for Applied Language Studies, UK: University of Reading.
O’Sullivan, B (2000a) Towards a Model of Performance in Oral Language Testing,Unpublished PhD Thesis, UK: University of Reading.
O’Sullivan, B (2000b) Exploring Gender and Oral Proficiency Interview Performance,System, 28 (3) 2000:373–386.
O’Sullivan, B (2002a) Learner Acquaintanceship and Oral Proficiency Test Pair-TaskPerformance, Language Testing, 19 (3):277–295.
O’Sullivan, B (2002b) Defining the Test Taker: Theoretical and Practical Implications,plenary address at the CTELT 2002, Dubai.
O’Sullivan, B (2003) Computers in Language Testing: A Negative Perspective, Pre-conference Event – Debate: Computers in Language Testing, IATEFL Conference,Brighton, April 2003.
O’Sullivan, B and Lu, Y (2003) An empirical study on examiner deviation from the setinterlocutor frame in the IELTS speaking paper, Research project funded byIELTS/British Council.
O’Sullivan, B and Rignall, M (2001) A longitudinal analysis of the effect of feedback onrater performance on the IELTS General Training writing module, Research projectfunded by IELTS/British Council.
O’Sullivan, B and Rignall, M (2002) Assessing the value of multi-faceted Rasch biasanalysis based feedback to raters for the IELTS writing module, Research projectfunded by IELTS/British Council.
O’Sullivan, B and Weir, C (2002) Research Issues in Speaking Assessment, Cambridge:UCLES internal report.
O’Sullivan, B and Weir, C (2000) Issues in Task Difficulty, Paper presented at theLanguage Testing Forum, University of Bristol, November 2000.
O’Sullivan, B and Weir, C (2003) Does the computer make a difference? Reactions ofcandidates to a computer delivered versus traditional hand written academic formsof the IELTS writing component, Research project funded by IELTS/British Council.
O’Sullivan, B, Weir, C and Saville, N (2002) Using observation checklists to validatespeaking-test tasks, Language Testing, 19 (1), 33–56
Oller, J (1979) Language Tests at School, New York: Longman.
Pitman Qualifications (2000) Examinations Report, London: City & Guilds.
Pitman Qualifications (2002) Examinations Report, London: City & Guilds.
Pitman Qualifications (2003) Centre Handbook, London: City & Guilds.
Pitman Qualifications (Undated) English for Business Communications, Level 1: PastPaper, London: City & Guilds.
Pitman Qualifications (Undated) English for Business Communications, Level 2: PastPaper, London: City & Guilds.
Pitman Qualifications (Undated) English for Business Communications, Level 3: PastPaper, London: City & Guilds.
References
389
Pollitt, A and Ahmed, A (1999) A New Model of the Question Answering Process, Paperpresented to the International Association for Educational Assessment in Bled,Slovenia, May 1999. Available at: www.ucles-red.cam.ac.uk/conferencepapers/IAEA1999APAA.pdf
Porter, D (1991a) Affective Factors in Language Testing, in Alderson, J C and North, B(Eds) Language Testing in the 1990s, London: Macmillan (Modern EnglishPublications in association with The British Council), 32–40.
Porter, D (1991b) Affective Factors in the Assessment of Oral Interaction: Gender andStatus, in Sarinee Arnivan (Ed.) Current Developments in Language Testing,Singapore: SEAMEO Regional Language Centre, Anthology Series 25, 92–102.
Porter, D and O’Sullivan, B (1999) The Effect of Audience Age on Measured WrittenPerformance, System 27 (1), 65–77.
Porter, D and Shen Shu Hung (1991) Gender, Status and Style in the Interview, TheDolphin 21, Aarhus University Press, 117–128.
Rethinasamey, S (in progress) The effects on rating performance of different traininginterventions, PhD research project, Roehampton, UK: Roehampton University.
Robinson, P C (1980) English for Specific Purposes, Oxford: Pergamon.
Robinson, P C (1985) Needs Analysis: From Product to Process, paper presented at the5th Annual ESP Symposium, Leuven, Belgium, August 1985.
Royal Society of Arts (1987) Report of the pilot scheme for the Certificate in EFL forSecretaries, London: RSA.
Rubinstein, J S, Meyer, D E and Evans, J E (2001) Executive Control of CognitiveProcesses in Task Switching, Journal of Experimental Psychology: HumanPerception and Performance 27 (4), 763–797.
Saville, N (2003) The process of test development and revision within UCLES EFL, inWeir, C J and Milanovic, M (2002) (Eds) Continuity and Innovation: Revising theCambridge Proficiency in English Examination 1913–2002, Cambridge: Universityof Cambridge Local Examinations Syndicate and Cambridge University Press,57–120.
Saville, N and Hargreaves P (1999) Assessing speaking in the revised FCE, ELTJournal 53 (1), 42–51.
Schmidt, M (1980) Coordinate structures and language universals in Interlanguage,Language Learning 30, 397.
Schröder, K (1981) Methods of exploring language needs in industry, in Freudenstein,R et al (Eds) Language Incorporated: Teaching foreign languages in industry,Pergamon & Max Hueber Verlag, 43–54.
Selinker, L and Douglas, D (1985) Wrestling with ‘Context’ in Interlanguage Theory,Applied Linguistics 6 (2), 190–204.
Skehan, P (1998) A Cognitive Approach to Language Learning, Oxford: OxfordUniversity Press.
Smith, J (1989) Topic and Variation in ITA Oral Proficiency, English for SpecificPurposes 8, 155–168.
Spolsky, B (1995) Measured Words, Oxford: Oxford University Press.
References
390
Sprouse, J L and Webb, J E (1994) The Pygmalion Effect and Its Influence on theGrading and Gender Assignment on Spelling and Essay Assessments, UnpublishedMaster’s Thesis, University of Virginia. ERIC_NO: ED374096.
Steffensen, M S and Joag-Dev, C (1984) Cultural knowledge and reading in Alderson, JC and Urquhart, A H (Eds) Reading in a Foreign Language, London: Longman.
Straus, G S, Miles, A J, Levesque, L L (2001) The effects of videoconference,telephone, and face-to-face media on interviewer and judgments in employmentinterviews, Journal of Management 27 (3), 363–381.
Swales, J (1971) Writing Scientific English, London: Thomas Nelson.
Swales, J (1984) ESP comes of age? – 21 years after ‘Some Measurable Characteristicsof Modern Scientific Prose’, UNESCO Alsed – LSP Newsletter 7, 2 (19), 9–20.
Sweedler-Brown, C O (1992) The Effect of Training on the Appearance Bias of HolisticEssay Graders, Journal of Research and Development in Education 26 (1), 24–29.
Tarone, E (1985) Variability in Interlanguage use: A study of style shifting inmorphology and syntax, Language Learning 35 (3), 373–404.
Tarone, E (1988) Variation in Interlanguage, London: Edward Arnold.
Taylor, L (2000) Investigating the paired speaking format, Research Notes 2, 14–15.
Taylor, L and Gutteridge, M (2003) Responding to diversity: providing tests forlanguage learners with disabilities, Research Notes 11, 2–4.
Teasdale, A (1994) Authenticity, validity, and task design for tests of well defined LSPdomains, in Khoo, R (Ed.) LSP Problems & Prospects, Singapore: SEAMEO RELC,230–242.
Thurstun, J and Candlin, C (1998) Concordancing and the teaching of the vocabulary ofAcademic English, English for Specific Purposes 17 (3), 267–280.
University of Cambridge Local Examinations Syndicate (1994) CEIBT Review:March–May 1994, Unpublished internal document.
University of Cambridge Local Examinations Syndicate (1999) Minimum ProfessionalRequirements, Unpublished internal document.
University of Cambridge Local Examinations Syndicate (2000a) EFL Research atUCLES, Research Notes 1, 2–4.
University of Cambridge Local Examinations Syndicate (2000b) BEC Handbook,Cambridge: University of Cambridge Local Examinations Syndicate.
University of Oxford Delegacy of Local Examinations (1990) Oxford InternationalBusiness English Certificate, Oxford: UODLE.
Vargo, S (1994) Attention to Nonverbal Cues: A View Through the Perceptual Lens ofGender Schema, Unpublished PhD, Indiana University.
Weighill, B and Shaw, S (2003) The change process at the paper level, Paper 2, Writing,in Weir, C J and Milanovic, M (2002) (Eds) Continuity and Innovation: Revising theCambridge Proficiency in English Examination 1913–2002, Cambridge: Universityof Cambridge Local Examinations Syndicate and Cambridge University Press,175–236.
Weigle, S C (1994) Effects of training on raters of ESL compositions, Language Testing11 (2), 197–223.
Weigle, S C (1998) Using FACETS to model rater training effects, Language Testing15 (2), 264–288.
Weir, C J (1983) Identifying the language needs of overseas students in tertiaryeducation in the United Kingdom, Unpublished PhD dissertation, UK: University ofLondon.
Weir, C J (1993) Understanding and Developing Language Tests, London: Longman.
Weir, C J (2002) The History of the CPE, 1913–2013, paper presented at the IATEFLConference, York April 2002.
Weir, C J (2003a) A Survey of the history of the Certificate of English (CPE) in thetwentieth century, in Weir and Milanovic (Eds) Continuity and Innovation: Revisingthe Cambridge Proficiency in English Examination 1913–2002, Cambridge:University of Cambridge Local Examinations Syndicate and Cambridge UniversityPress, 1–56.
Weir, C J (2003b) Conclusions and Recommendations, in Weir and Milanovic (Eds)Continuity and Innovation: Revising the Cambridge Proficiency in EnglishExamination 1913–2002, Cambridge: University of Cambridge Local ExaminationsSyndicate and Cambridge University Press, 473–478.
Weir, C J (2004) Language Testing and Validity Evidence, Oxford: Palgrave.
Weir, C J and Milanovic, M (Eds) (2003) Continuity and Innovation: Revising theCambridge Proficiency in English Examination 1913–2002, Cambridge: Universityof Cambridge Local Examinations Syndicate and Cambridge University Press.
West, R (1994) Needs analysis in language teaching, Language Teaching 27 (1), 1–16.
Wilds, C P (1975) The oral interview test, in Jones, R L and Spolsky, B (Eds) TestingLanguage Proficiency, Arlington: Center for Applied Linguistics, 29–44.
Wilson, K M (1989) Enhancing the interpretation of a norm-referenced second-language test through criterion referencing: A research assessment of experience inthe TOEIC testing context, TOEIC Research Reports #1, Princeton NJ: EducationalTesting Services.
Woodford, P E (1982) An introduction to TOEIC: The initial validity study, TOEICResearch Summary – Introduction, Princeton NJ: Educational Testing Services.
References
391
393
AAccommodations 91Administration 20–21, 23, 35, 52, 53, 57–58,
73, 78, 81, 85, 98–112, 115, 127, 146ALTE Can Do 26, 43, 84, 86-91, 187ALTE Levels 45, 48, 63, 78, 84, 97,
107–108, 126–127, 129, 141, 150, 188ALTE, members 47, 104 (Table 2.5) Assessment 43, 48, 83–85, 106, 109, 115,
117, 127, 176Assessment, educational 22Assessment, performance 4, 73, 75Assessment, scales 4, 12–14, 15, 52, 56Attenuation paradox 97Authenticity 1, 3–5, 7–8, 16, 21–23, 28–29,
32–35, 38–39, 45–46, 50–51, 53–54,56–57, 58, 77, 80-81, 96, 159, 168, 173(Table 4.25), 174, 175, 176, 190, 193
Authenticity interactional 7, 15, 22, 29, 35,39, 42, 45, 51, 53, 57, 60, 62, 74, 79, 80–81,98, 183, 187, 188
Authenticity situational 7, 14–15, 21, 29, 35,42, 45, 50, 53, 57, 60, 61–62, 74, 76, 79, 80,93, 98, 177, 182–183
B Background, academic 4Background, cultural 6Background, in business 52, Background knowledge 4-7, 13, 17, 23, 24,
29, 45, 54, 61, 161, 173 (Table 4.25) Background, language 37, 83 Background, variables 92BEC 182, 184, 188–193BEC, original 2, 8, 60, 77–79BEC, revision process 82–118 BEC, major changes 119–129 BEC, wordlists 176–177BEC, detailed changes 130–173Bias 19, 23, 74, 91, 97, 101BULATS 2–3, 37, 63–77, 187
C CEF 11, 25, 43, 45, 48, 54, 63, 87, 89 (Figure
2.4), 90, 93, 107, 126–127, 129, 181 CEFLS 2, 55–57, 61, 62CEIBT 2, 57–60, 61, 63, 78
CIC 2, 42–47Computer 67, 70–71, 108, 186–187Consequential validity 103Consistency, internal 8, 20–21, 59, 94, 97,
111, 131, 184–185, 189Consistency, grading 107–111Construct, definition 10, 13, 15, 17–19, 26,
29, 31, 36–38, 41, 43, 46, 48–49, 51, 52–53,80, 90–94, 131, 132, 139, 170, 173 (Table4.25)
Construct, validity 95–97, 108 Consultation 113, 118Context 3, 71, 78, 79 Context, validity 7, 34–35, 92–93, 175,
179–180, 182 Context, focused 32, 85, 175,189Context, specific 4, 7, 9, 13-15, 21, 29, 42,
45, 50, 53, 69, 76, 79, 82, 125, 138, 145,157, 159, 161
Context, oriented 32, 85, 175, 189Corpus/corpora 3, 82, 93, 176–177
D DIALANG 37Difficulty, level 25, 59, 108, 110–111Difficulty, item 108, 111Difficulty, task 56, 144, 191Dictation 32
E ESP 1–2Examiner 4Examiner effect 32, 53, 73, 186, 188–189Examiner frame (see also Interlocutor frame)
125 (Figure 3.2)Examiner role 31, 52, 53, 58, 100, 109, 125,
150, 159, 170, 192Examiner training 4, 11Examiner, RICTME 106–107
F Facets 100, 185Fairness 85, 117, Function, language 9, 37, 39, 53, 72, 83,
157–159, 161, 167, 168, 191–192Function, rhetorical 6
Index
Index
394
G Generalisability 14, 41, 63, 80, 174, 176–177Grading 106, 109–111, 112Graphical Profiling 119-120
I Impact, of non linguistic factors 4, 15, 23,
29–30, 35, 39–40, 42, 46, 51, 57, 77, 79, 81,173 (Table 4.25), 174, 175
Impact, of test 83, 95, 103–104, 115Indigenous scale 10, 13, 15, 189Instructions 28, 30, 70–71, 74, 108, 180 Item, bank 45, 58, 105, 108–109interlocutor 51, 100, 123, 159–160, 161, 192Interlocutor frame (see also Examiner frame)
168–170, 193Internal consistency 20–21, 97–99, 102, 111,
131, 184–185, 189IRT
J JETRO 2, 52JOCT 52–54
L LCCIEB 1, 2, 5, 6, 24, 35–40, 63, 85, 98, 191LSP 5–8, 10, 13, 17–18, 57, 80, 174, 176,
188, 193
M Marking, double 77, 109–110, MCQ 16, 19, 22, 38, 41, 45, 98, 133
O Observation checklists 84, 85, 157–159OIBEC 60, 61-63
P Paired Format 61, 167, 170Performance conditions 3, 93, 102 Performance test 2, 11, 13, 73, 75, 79, 101,
106Pitman 2, 24–35, 98PLAB 5, 11Practicality 104–107, 192Pronunciation 9, 71, 94 (Figure 2.8), 170Proof-reading 31, 32Psychometrics 19, 75 Psychometric–structuralism 2, 19, 41, 187
R Rasch, multi-faceted 100–102, 108, 187Rater agreement 185Rater reliability 100, 184Rater consistency 189Rater training 4, 11, 186, 189, 190Reliability 97–103, 184–186Royal Society of Arts 2, 55–56,
S Scoring validity 90, 92, 94, 175, 185Senior team leader 106–107, 115, 119Socio-cognitive 92, 178Specifications 26, 37, 59, 64, 92–93, 103,
113, 114, 144Specificity 5, 6, 7, 14–15, 18, 21–23, 28–29,
32–35, 38–39, 42, 45–46, 50–51, 53–54,60, 63, 76, 79, 80, 81, 125, 157, 159, 168,173 (Table 4.25), 174–175, 176–182, 193
Stakeholder 61–62, 78, 84, 113, 115, 126,127, 157, 176
Standardisation 109, 186, 189, 190
T Task format 57–58TEEP 11–12Test development 3, 8–9, 77–78, 82–90, 103,
114, 179, 187, 193Test revision 58, 60, 79, 82–118, 188, 190,
193TFI 2–3, 40–42Theory-based validity 4, 22, 92, 179,
182–183, 187 TOEIC 2, 16–23, 40–42, 52, 55, 60, 86, 191
U UCLES 2, 57, 77, 108 Usefulness 15, 90, 95, 104, 106, 115, 172
V VRIP 95–106, 117
W Washback 121, 123, 140Weighting 46, 117, 120–124, 127–128, 141,
144