indexed_4_ngai_empirical examination of the adoption of webct using tam

8/14/2019 INDEXED_4_Ngai_Empirical Examination of the Adoption of WebCT Using TAM

1/18

Empirical examination of the adoption of WebCT using TAM

E.W.T. Ngai a, * , J.K.L. Poon b , Y.H.C. Chan a

a Department of Management and Marketing, The Hong Kong Polytechnic University, Hung Hom,Kowloon, Hong Kong, PR China

b The Institute of Vocational Education, Hong Kong, PR China

Received 29 March 2004; received in revised form 3 November 2004; accepted 17 November 2004

Abstract

Web Course Tools (WebCT) have enhanced the ability and motivation of institutes of higher educationto support e-learning. In this study, we extended the Technology Acceptance Model to include technicalsupport as a precursor and then investigated the role of the extended model in user acceptance of WebCT.Responses from 836 university students were used to test the proposed structural model. The data showedthat technical support has a signicant direct effect on perceived ease of use and usefulness, while perceivedease of use and usefulness are the dominant factors affecting the attitude of students using WebCT. Theresults indicate the importance of perceived ease of use and perceived usefulness in mediating the relation-ship of technical support with attitude and WebCT usage. 2005 Elsevier Ltd. All rights reserved.

Keywords: WebCT; Technology acceptance model; Web-based learning systems

1. Introduction

Online teaching and learning is becoming an increasingly important part of higher educa-tion. The current delivery modes and instructional designs of higher education meet the chal-lenges from the advancement of technology. The major changes will be mostly from the area

0360-1315/$ - see front matter 2005 Elsevier Ltd. All rights reserved.doi:10.1016/j.compedu.2004.11.007

* Corresponding author. Tel.: +852 2765 6611; fax: +852 2766 7296.E-mail addresses: [email protected]; [email protected] (E.W.T. Ngai).

www.elsevier.com/locate/compedu

Computers & Education 48 (2007) 250267
mailto:[email protected]:[email protected]


2/18

of e-learning. E-Learning, also known as Web-based learning, is dened as an Internet-enabledlearning process ( Gunasekaran, Mcneil, & Shaul, 2002 ). According to the report conducted bythe International Data Corporation (2000) , The number of colleges and universities in US offer-ing e-learning will more than double, from 1500 in 1999 to more than 3300 in 2004. Studentenrollment in these courses will increase 33% annually during this time. They believe the marketfor e-learning in higher education will be worth almost US$750 million by 2004.

In order to support e-learning, various Web-based learning systems have been developed forhigher education. For example, Web Course Tools (WebCT), the Web Course Homepage System(WebCH), Blackboard Learning System and the System for Multimedia Integrated Learning(Smile), are the latest waves of technology-based pedagogical tools. They provide e-learning plat-forms that use the Internet as a delivery mechanism to allow students from all over the world toaccess a number of learning tools such as discussion boards, chat rooms, course content manage-ment, etc. Many institutions of higher education adopt such Web-based learning systems for their

e-learning courses. For instance, WebCT has been used by 2100 institutions of higher educationall over the world, including famous universities such as Stanford, UCLA, and so forth (WebCT,2001). However, there is a lack of empirical examination of the adoption of web-based learningsystems. In Hong Kong, many institutions of higher education offer e-learning as part of their cur-riculum, and have some form of on-line learning using these Web-based learning systems. WebCTis being used nearly in all institutions of higher education in Hong Kong.

In this study, we propose a model, based on the extension of the Technology Acceptance Model(TAM) ( Davis, Bagozzi, & Warshaw, 1989 ), to investigate the factors that affecting the acceptanceof WebCT for supporting e-learning. The aims of this paper are:

(a) to determine the current usage of WebCT teaching and learning in Hong Kong institutions of higher education,

(b) to identify the factors affecting the acceptance of WebCT teaching and learning in HongKong institutions of higher education, and

(c) to develop a model for the acceptance of WebCT in Hong Kong for higher education basedon the TAM.

This paper is organized as follows. First, we present a review of the literature on web-basedlearning systems and the TAM. Then, the research model and hypotheses are proposed. Next,the research method used in this study is described. The results of the collected data and the pro-posed model, which were analyzed using structural equation modeling (SEM), are reported. The

nal section discusses the ndings of the study and concludes the paper.

2. Literature review

2.1. Web-based learning systems

With the widespread use of the World Wide Web (WWW), many institutions of higher educa-tion have identied opportunities for developing courses for Web-based learning. Unlike tradi-tional classroom-based learning, Web-based learning offers various benets. Some studies

E.W.T. Ngai et al. / Computers & Education 48 (2007) 250267 251


3/18

compared the learning outcomes of online and traditional classroom-based courses and foundthat students taking the online course outperformed those taking the traditional classroom-basedcourse ( Kekkonen-moneta & Moneta, 2002; Hofmann, 2002 ). Their study showed that Web-based learning is appropriate for teaching and learning.

Many institutions of higher education have already made the transition to online teaching bycontracting technology and courseware providers to design the infrastructure needed to custom-ize their own online programs. Real improvement began to emerge in the 1990s, when many spe-cially designed Web-based learning systems such as WebCT, WebCH, Smile, etc., wereintroduced. According to the denition of the IEEE Learning Technology Standard Committee ,a Web-based learning system is: A learning technology system that uses Web-browsers as theprimary means of interaction with learners, and the Internet or an intranet as the primary meansof communication among its subsystems and with other systems. These systems work as a plat-form to facilitate teaching and learning. They assist in managing the mundane tasks as well as

stimulating new insights into conducting classes and delivering course content. Also, they can beused to publish an entire online course or merely to make supplemental materials available on-line. Nearly all the universities in Hong Kong are using WebCT to support Web-based teachingand learning. These include the University of Hong Kong, The Chinese University of HongKong, The Hong Kong Polytechnic University, Hong Kong Baptist University, City Universityof Hong Kong, Lingnan University, The Hong Kong University of Science and Technology andall institutions under Vocational Training Council. WebCT is known as one of the best course-management systems available, and provides a number of learning tools. There are a limited of studies on Web-based learning using WebCT. Lu, Yu, and Liu (2003) undertook a study to iden-tify the impact of learning styles, learning patterns and other selected factors on the learning per-formance of students in a WebCT MIS graduate course. Morss (1999) , Wernet, Olliges, andDelicath (2000) and Withnam, Krockover, Ridgway, and Zinsmeister (2002) surveyed the per-spectives of university students on Web-based learning using WebCT. The feedback from thestudents was generally positive. Overall, students were likely to accept using Web-based systemsfor teaching and learning.

2.2. Technology acceptance model

While many institutions of higher education have been using the Web for teaching and learning,little research has been done to identify the factors affecting students acceptance of the WebCT

learning system, nor has any research been conducted on attitudes towards WebCT.TAM, introduced by Davis et al. (1989) , was based on the Theory of Reasoned Action (TRA)(Fishbein & Ajzen, 1975 ) and specically designed for explaining and predicting user acceptanceof specic types of technology. TAM was built on collective ndings suggesting that the desiredtechnology was greatly dependent on user acceptance of technology. It suggested that perceivedusefulness and perceived ease of use were important factors in determining the use of informationsystems. A number of studies have successfully adopted TAM to examine the acceptance of newtechnologies such as personal computers ( Igbaria, Zinatelli, Cragg, & Cavaye, 1997 ), word pro-cessors and spreadsheets ( Chau, 1996 ). Recently, as the WWW has emerged as a new form of information technology, researchers have used TAM to investigate various WWW contexts in

252 E.W.T. Ngai et al. / Computers & Education 48 (2007) 250267


4/18

predicting acceptance of technology. These include Web browsers ( Morris & Dillon, 1997 ), theuse of Websites ( Lederer, Maupin, Sena, & Zhuang, 2002; Lin & Lu, 2002; van der Heijden,2003 ), Web retailing ( Chen, Gillenson, & Sherrell, 2002; O cass & Fenech, 2003 ), online purchaseintentions ( van der Heijden, Verhagen, & Creemers, 2003 ), etc. However, there are not many stud-ies explaining the acceptance of Web-based learning systems using the TAM. Lou, Luo, andStrong (2000) examined the critical mass, as an external variable, of the acceptance of groupware,based on the TAM. Groupware technologies support group communication and collaborationsuch as e-mail and electronic bulletin boards, which are similar to Web-based learning systems.They found that all of the constructs (critical mass, perceived usefulness and perceived ease of use), affect the intention to use groupware. Selim (2003) investigated student use and acceptanceof course websites based on the usefulness of the course website, ease of use and usage. The resultsshowed that there is a signicant relationship between the usefulness of a course website and easeof use in determining the usage of a course website. In fact, other external variables should be

included in TAM for measuring the specic technology, since they may inuence the perceivedease of use and perceived usefulness of that technology ( Davis et al., 1989 ).

3. Research model and hypotheses

The TAM has been widely used to predict the acceptance of a new technology. In this study, aresearch model is based on the TAM to study the adoption of WebCT for higher education. Thecomponent of technical support has been incorporated in the TAM model, and serves as anextension to TAM for measuring the acceptance of Web-based learning systems. The research mod-el explains the system usage of Web-based learning systems for higher education. It consists of tech-nical support, perceived usefulness, perceived ease of use, attitude and intention to use (see Fig. 1 ).

TechnicalSupport

Perceived Easeof Use

PerceivedUsefulness

Attitude Intention to Use System Usage

TS1

TS2

TS3

TS4

TS5

TS6

P EO U1 P EOU 2 P EO U3 PE OU 4 P EO U5

PU1 PU2 PU3 PU4 PU5 PU6

A1 A2

A3 A4

IU1 IU2 SU1 SU2

H2

H1

H5

H4

H7

H10

H8

H12H3

H6

H9

H11

Fig. 1. Proposed structure model for the acceptance of Web-based learning systems.



5/18

Several previous studies have shown that there are various external factors that indirectly inu-ence the acceptance of technology through perceived usefulness and perceived ease of use ( Daviset al., 1989; Szajna, 1996 ). In this study, we expect technical support to be one such external factoraffecting the acceptance of WebCT for higher education. Ralph (1991) dened technical supportas knowledge people assisting the users of computer hardware and software products, whichcan include help desks, hotlines, online support services, machine-readable support knowledge-bases, faxes, automated telephone voice response systems, remote control software and otherfacilities. Technical support is one of the important factors in the acceptance of technology forteaching ( Hofmann, 2002; Sumner & Hostetler, 1999; Williams, 2002 ) and in user satisfaction(Mirani & King, 1994 ). High levels of organizational support, including management supportand information center support, were thought to promote more favorable attitudes about the sys-tem among users and information specialists, and lead to greater success for personal computingsystems ( Igbaria, 1990 ). Igbaria et al. (1997) argued that internal/external personal computing

support/training was affecting the acceptance of personal computing in small rms. As a result,the following three hypotheses are proposed:

H1. Technical support has a positive effect on the perceived usefulness of WebCT.

H2. Technical support has a positive effect on the perceived ease of use of WebCT.

H3. Technical support has a positive effect on attitudes towards using WebCT.

In the generic TAM, the belief-attitude-intention-behavior relationship has been demonstratedin various studies. However, the TAM-related hypotheses in the context of WebCT have not beenveried. In this study, the perceived ease of use of the WebCT is dened as the degree to which theuser believes that using the WebCT will be free of effort ( Davis, 1989 ). Davis (1989) showed thatease of use had a direct effect on perceived usefulness. Further studies on TAM also demonstratedstrong empirical support for a positive relationship between perceived ease of use and perceivedusefulness ( Adams, Nelson, & Todd, 1992; Szajna, 1996 ). The perceived usefulness of the WebCTis dened as the degree to which the user believes that using the WebCT would enhance his/herlearning performance ( Davis, 1989 ). The TAM posits that perceived usefulness and perceived easeof use has a direct effect on attitudes towards using a new technology. Attitude is the degree towhich the user is interested in specic systems, which has a direct effect on the intention to usethose specic systems in the future ( Davis et al., 1989 ) and the actual usage of the systems ( Bajaj& Nididumolu, 1998 ). The extent to which systems are used over a xed unit of time is inuencedby the intention to use ( Davis et al., 1989 ). In addition, the usage of the system is also affected by

perceived ease of use and perceived usefulness ( Davis et al., 1989; Igbaria et al., 1997;Selim, 2003 ). As a result, the following hypotheses based on the TAM-relationship areproposed:

H4. Perceived ease of use has a positive effect on attitudes toward the use of WebCT.

H5. Perceived ease of use has a positive effect on the perceived usefulness of WebCT.

H6. Perceived ease of use has a positive effect on the use of WebCT.

H7. Perceived usefulness has a positive effect on attitudes towards the use of WebCT.



6/18


7/18

the 1400 questionnaires distributed, 1263 were collected and used for analysis. The overall re-sponse rate was about 90%. We invested much effort in obtaining a high response rate. The highresponse rate in our study was due to the personal approach, but other factors of our study designalso contributed. We marked the survey itself with the Hong Kong Polytechnic University(PolyU) logo and explained to the respondent the purpose of the survey and motivated him/herto reply personally. We mentioned that PolyU was sponsoring the survey. Also the condentialityof the results had been stressed. Other factors that might have increased the direct response ratewere the involvement of student interviewers, face-to-face interviews at the place of commonrooms and or canteens in each university.

5. Analysis and results

5.1. Demographics and descriptive statistics

The returned sample characteristics are illustrated in Table 1b . In this study, a total of 1400questionnaires were distributed, of which 1263 were collected. Table 1b shows that the major-ity of the respondents were undergraduates in their rst to third years of study, and that theycame from a variety of programs. About 88.6% were full-time students in the campus. Of the

respondents, 66.2% reported that they had the experience in using Web-based learning sys-tems, while 714 students out of 836 were using WebCT (See Table 2 ). WebCT seems tobe the dominant Web-based learning system in institutions of higher education in HongKong.

5.2. Data analysis

In this study, the sample was split randomly into two sub-samples, S1 and S2, with 418 in eachsub-sample, from the total of 836 respondents who reported that they had experience in using

Table 1aStratications by University

University Name Total no. of

students

Sample sizes

in Strata

Sampling fractions

in Strata (%)City University of Hong Kong (CityU) 16,142 200 1.24The Chinese University of Hong Kong (CU) 14,161 200 1.41Hong Kong Baptist University (HKBU) 7400 200 2.70The Hong Kong Polytechnic University (PolyU) 17,859 200 1.12University of Hong Kong (HKU) 14,216 200 1.41The Hong Kong University of Science andTechnology (HKUST) 7332 200 2.73Lingnan University (LU) 2000 200 10

Total 79,110 1400 1.77



8/18

Web-based learning systems. The initial analysis was performed on sample S1, to examine thereliability and construct validity using SPSS. Then, AMOS 4 was used with a different sampleS2 to test the proposed structural model.

Table 1bProle of the respondents

Frequency Percentage

GenderMale 583 46.2Female 680 53.8

Total 1263 100.0

Academic level Undergraduate 1086 86.0Master 102 8.1Doctor 10 0.8Other (e.g., associate degree) 65 5.1

Total 1263 100.0

Year of studyYear 1 553 43.8Year 2 334 26.4Year 3 353 27.9Year 4 13 1.0>Year 4 10 0.8

Total 1263 100.0

Mode of studyFull-time 1119 88.6Part-time 144 11.4

Total 1263 100.0

Courses currently studying Business Administration 262 20.7Management 102 8.1Computer Science 62 4.9Chemistry 80 6.3Physics 48 3.8Biology 51 4.0Social Work 23 1.8Medicine 38 3.0Mechanical Engineering 21 1.7Civil Engineering 17 1.3Others Engineering 96 7.6Language 122 9.7Arts 29 2.3Others (e.g., Nursing, Journalism) 312 24.7

Total 1263 100.0



9/18

5.3. Analysis of validity and reliability

In this study, construct validity and reliability were examined. Construct validity refers to the

degree to which a construct differs from other constructs. It can be assessed through principalcomponent analysis (PCA). Using varimax rotation, PCA was used in an attempt to reconstructsix composite factors. When interpreting the rotated factor pattern, an item was said to load on afactor if the factor loading is 0.4 or greater ( Nunnally, 1978 ). Using this criterion, the rotated pat-tern matrix was examined for items that did not load on a factor with the other items from thesame scale. Items that cross-loaded on multiple factors were also examined and would be deleted.Besides, eigenvalues were examined to decide the number of factors to extract. An eigenvaluegreater than 1 was used as a criterion to determine the number of factors.

The initial PCA showed that ve factors were extracted. Technical support, perceived use-fulness, perceived ease of use and system usage was loaded into four distinct factors. How-ever, attitude and intention to use were collapsed into one factor. We decided to remove theconstruct of intention to use because the survey showed that the requirement of lecturers to usethe system has a large impact on the use of Web-based learning systems. Even if the attitude of students towards Web-based learning systems is positive, they may have no intention of usingthem if their lecturers do not require it. Therefore, a second PCA was conducted by removingthe item of intention to use. The results of the PCA for each of the constructs are given in Table3. These results show that the loadings of individual items on the construct exceed 0.4 ( Nunnally,1978 ), which veries that the items measured exhibit sufficient validity. All factor loadings werelarger than 0.5 and no cross-loading on multiple factors exceed 0.5. The correlation matrix forthe data set is given in Table 4 . Correlations greater than 0.3 for a sample size of 418 used inthe analysis were statistically signicant at the 0.01 level. An inspection of the correlation matrix

revealed that the majority of the inter-item correlations were signicant (greater than 0.3) at the0.01 level. The items associated with a measure correlated more highly with each other than withitems associated with other measures in the model. The intercorrelations among the items associ-ated with the measures are stronger than their correlations with items representing other mea-sures. As a result, the convergent and discriminant validity of the measurement can bedemonstrated.

Reliability refers to the extent to which the constructs are free from error and yield consistentresults. Cronbach s a was used to measure the internal consistency of the multi-item scales used inthis study. The Cronbach s a (see Table 3 ) for each of the ve factors was much higher than thealpha threshold level of 0.7 ( Nunnally, 1978 ). Table 3 shows that the Cronbach s a coefficients

Table 2Experience in using Web-based learning systems

Frequency Percentage

Experience in using WebCT Yes 836 66.2No 427 33.8

Web-based system currently using WebCT 714 85.4



10/18

Table 3Results of principal component analysis

Component

Factor 1 perceivedusefulness

Factor 2 perceivedease of use

Factor 3 attitude Factor 4 tecsupport

PU1 0.81PU2 0.81PU3 0.84PU4 0.83PU5 0.74PU6 0.83PEOU1 0.83PEOU2 0.80PEOU3 0.80PEOU4 0.75

PEOU5 0.85A1 0.61A2 0.54A3 0.64A4 0.57TS1 0.73TS2 0.82TS3 0.80TS4 0.68TS5 0.54TS6 0.57SU1

SU2 Cronbach s a 0.93 0.90 0.91 0.83 Eigenvalue 8.99 2.38 2.26 1.50 Cum. variance explained (%) 23.65 41.83 55.30 63.53


11/18

Table 4Correlations matrix between the measurement items

Mean S.D. TS1 TS2 TS3 TS4 TS5 TS6 PU1 PU2 PU3 PU4 PU5 PU6 PEOU1 P EOU2 P EOU3 P EOU4 P EOU5 A1 A2

TS1 3.85 1.31 1.000TS2 3.41 1.40 0.551 * 1.000TS3 3.03 1.32 0.456 * 0.678 * 1.000TS4 4.16 1.39 0.528 * 0.414 * 0.408 * 1.000TS5 4.63 1.45 0.435 * 0. 25 4 0 .26 3 0 .6 50 * 1.000TS6 3.65 1.35 0.385 * 0.388 * 0.394 * 0.384 * 0.368 * 1.000PU1 4.31 1.27 0.284 0.223 0.176 0.300 * 0.276 0 .335 * 1.000PU2 3.86 1.21 0.294 0.265 0.246 0.269 0.234 0.326 * 0.728 * 1.000PU3 4.17 1.29 0.262 0.228 0.201 0.263 0.235 0.270 0.720 * 0.722 * 1.000PU4 4.25 1.34 0.290 0.189 0.185 0.273 0.285 0.268 0.690 * 0.685 * 0.735 * 1.000PU5 4.15 1.33 0.244 0.146 0.137 0.266 0.230 0.264 0.590 * 0.617 * 0.628 * 0.664 * 1.000PU6 4.31 1.30 0.341 * 0.212 0.126 0.293 0.267 0.292 0.734 * 0.674 * 0.741 * 0.745 * 0.707 * 1.000PEOU1 4.68 1.19 0.229 0.142 0.171 0.230 0.207 0.167 0.302 * 0.198 0.268 0.298 0.243 0.318 * 1.000PEOU2 4.44 1.16 0.208 0.139 0.137 0.314 * 0.267 0 .301 * 0.371 * 0.315 * 0.358 * 0.388 * 0.318 * 0.394 * 0.649 * 1.000PEOU3 4.55 1.10 0.271 0.158 0.149 0.280 0.257 0.235 0.341 * 0.283 0 .338 * 0.370 * 0.328 * 0.397 * 0.592 * 0.615 * 1.000PEOU4 4.39 1.16 0.284 0.199 0.149 0.303 * 0.261 0 .311 * 0.399 * 0.358 * 0.383 * 0.427 * 0.395 * 0.414 * 0.561 * 0.617 * 0.60s7 * 1.000PEOU5 4.61 1.18 0.267 0.125 0.171 0.316 * 0.307 * 0.287 0 .377 * 0.308 * 0.345 * 0.384 * 0.357 * 0.412 * 0.651 * 0.632 * 0.689 * 0.679 * 1.000A1 3.97 1.22 0.282 0.246 0.110 0.237 0.215 0.305 * 0.507 * 0.468 * 0.452 * 0.473 * 0.427 * 0.540 * 0.256 0.375 * 0.329 * 0.425 * 0.336 * 1.00A2 4.50 1.25 0.304 * 0.180 0.229 0.275 0.247 0.207 0.543 * 0.454 * 0.52s8 * 0.545 * 0.430 * 0.580 * 0.368 * 0.362 * 0.398 * 0.419 * 0.443 * 0.58A3 4.18 1.27 0.265 0.273 0.186 0.205 0.209 0.251 0.510 * 0.480 * 0.505 * 0.519 * 0.432 * 0.542 * 0.249 0.389 * 0.322 * 0.431 * 0.380 * 0.72A4 4.23 1.23 0.280 0.210 0.151 0.225 0.218 0.258 0.561 * 0.512 * 0.563 * 0.586 * 0.498 * 0.631 * 0.374 * 0.431 * 0.428 * 0.465 * 0.472 * 0.66SU1 3.81 1.49 .0.158 0.076 0.079 0.119 0.138 0.182 0.178 0.188 0.204 0.204 0.200 0.226 0.095 0.230 0.155 0.180 0.146 0.113 SU2 3.82 1.28 0.108 0.105 0.077 0.119 0.136 0.177 0.102 0.181 0.152 0.146 0.192 0.145 0.011 0.110 0.045 0.110 0.044 0.119

* Indicates correlations signicant at the 0.01 level.


12/18

range from 0.71 to 0.93, which indicates that the instrument can be considered reliable and inter-nally consistent.

5.4. Analysis of the structural model

After removing the construct intention to use, the proposed structural model was revised andexamined using SEM (see Fig. 2 ). SEM is a comprehensive statistical approach to testing hypoth-eses about relations among observed and latent variables ( Hoyle, 1995 ). A major advantage of SEM is the ability to estimate a complete model incorporating both measurement and structuralconsiderations. To analyze and conrm the tness of the revised structural model, a different sam-ple S2 was used.

A model is said to t the observed data to the extent that the covariance matrix it implies isequivalent to the observed covariance matrix ( Hoyle, 1995 ). However, there are no recommended

measures of model t. Six common indices of t that were recommended in the literature ( Hair,Anderson, Tatham, & Black, 1995; Hu, Chau, Sheng, & Tam, 1999 ) were employed in this study.The commonly used measures of model t, based on results from an analysis of the structuralmodel, are summarized in Table 5 . In practice, a Chi-square/degree of freedom of less than 3,GFI, NFI, CFI greater than 0.9, an AGFI greater than 0.8 and a RMSR of less than 0.1 are con-sidered indicators of good t. As seen in Table 5 , all goodness-of-t statistics are in the acceptableranges, with the exceptions of GFI, which is close enough (0.87) to the recommended value of 0.9.

TechnicalSupport

Perceived Easeof Use

R 2=0.25

PerceivedUsefulness

R 2=0.27

AttitudeR 2=0.60

System UsageR 2=0.12

TS1

TS2

TS3

TS4

TS5

TS6

PEOU 1 PEOU 2 PEO U3 PEO U4 PEOU 5

PU1 PU2 PU3 PU4 PU5 PU6

A1 A2

A3 A4

SU1 SU2

0.73

0.82

0.80

0.68

0.54

0.57

0.55

0.37

0.23

0.01

0.75

0.370.64 0.57

0.61 0.540.85 0.88

0.06

0.11

0.18

0.81 0.830.740.830.840.81

0.850.750.800.800.83

Fig. 2. Results of revised structure model.



13/18

The results of testing the structural model are presented in Table 6 and a graphical presentationof the results is shown in Fig. 2 . Table 6 shows the values of the coefficient of determination R 2 foreach endogenous variable. Table 6 depicts the R 2 for each endogenous variable. For perceivedease of use, R 2 = 0.25. For perceived usefulness, R 2 = 0.27; while for attitude, it is 25%, 27%and 60%, respectively. On the other hand, only 12% of the variation in the Web-based learningsystem can be explained by the proposed model. The proposed structure model explained 12%of the variance in the use of Web-based learning systems.

Fig. 2 and Table 6 further illustrate the signicant structural relationships among the studyvariables. Hypotheses 1, 2 and 3 postulate that technical support has a positive inuence on per-ceived usefulness (H1), perceived ease of use (H2), and attitude (H3). The results show that tech-

nical support has a strong direct effect on perceived usefulness and perceived ease of use ( b = 0.37and 0.55, respectively; p < 0.05). Although the direct effect of technical support on attitude isinsignicant ( b = 0.01, p > 0.05), it should be noted that the indirect effect on attitude through per-ceived usefulness and perceived ease of use is very strong ( b = 0.57, p < 0.05). As a result, H1, H2are supported while H3 is rejected.

Hypotheses 4, 5 and 6 investigate the relationship of perceived ease of use to attitude (H4), per-ceived usefulness (H5) and WebCT usage (H6). Perceived ease of use has a positive direct effect onattitude ( b = 0.37, p < 0.05), perceived usefulness ( b = 0.23, p < 0.05), and WebCT usage(b = 0.11, p < 0.05). In addition, it has a signicant indirect effect on attitude through perceivedusefulness ( b = 0.18, p < 0.05). Therefore H4, H5 and H6 are supported.

Table 6Effects of variables on the acceptance of Web-based learning systems

Perceived ease of

use

Perceived

usefulness

Attitude System usage

Directeffects

Indirecteffects

Directeffects

Indirecteffects

Directeffects

Indirecteffects

Directeffects

Indirecteffects

Technical support 0.55 * 0.37 * 0.13 0.01 0.57 * 0.19Perceived ease of use 0.23 * 0.37 * 0.18 * 0.11 * 0.08Perceived usefulness 0.75 * 0.18 * 0.05Attitude 0.06R2 0.25 0.27 0.60 0.12

* p 6 0.05.

Table 5Summary statistics of model t

Model goodness-t indexes Recommended value Results in this study

Chi-square/degree of freedom 6 3.0 3.0Goodness-of-t index (GFI) P 0.90 0.87Adjusted goodness-of-t index (AGFI) P 0.80 0.84Normalized t index (NFI) P 0.90 0.90Comparative t index (CFI) P 0.90 0.93Root mean square residual (RMSR) 6 0.10 0.07



14/18

Hypotheses 7 and 9 focus on the impact of perceived usefulness on attitude (H7) and WebCTusage (H9). Similar to the other studies on TAM ( Davis, 1989; Davis, 1993 ), perceived usefulnesshas a positive direct effect on attitude ( b = 0.75, p < 0.05) and WebCT usage ( b = 0.18, p < 0.05).Therefore, H7 and H9 are supported.

Hypotheses 11 postulated a relationship between attitude and the actual use of WebCT (H11).However, attitude has a weak and insignicant direct effect ( b = 0.06, p > 0.05) on WebCT usage.Thus, H11 is rejected.

The analysis of the structural model shows that perceived ease of use and usefulness are thedominant factors affecting the attitude of students. The ndings also show that technical supporthas a signicant effect on perceived ease of use and usefulness. The results demonstrate the impor-tance of perceived ease of use and perceived usefulness in mediating the relationships betweentechnical support and attitude and system usage. Hypotheses 8, 10 and 12 are rejected in thisstudy, since the construct of intention to use is being removed.

6. Discussion and conclusions

This study has investigated the underlying relationship between technical support, perceivedusefulness, perceived ease of use, attitude and the acceptance of the WebCT for higher education.The empirical examination of the adoption of WebCT using a structural model based on theextension of TAM has been tested and validated. Most of the causal relationships between theconstructs postulated by the structural model are well supported. The study provides further evi-dence of the appropriateness of applying TAM to measure the acceptance of WebCT in highereducation.

From the results, technical support was found to have a direct effect on the perceived ease of useand perceived usefulness. It also has a strong indirect effect on attitude. This underscores theimportance of user support and training in inuencing the perceptions of users and, eventually,their use of the system. As a result, it is essential for universities to provide effective user supportand to encourage users to use the system.

One interesting observation is that, in testing the structural model, attitude only demon-strated a weak direct effect on system usage. On the other hand, perceived usefulness and per-ceived ease of use both demonstrated a signicant direct effect on system usage, and the effectsthey showed were even stronger than that of attitude. This is different from previous studiesshowing that attitude mediates the effect of perceived ease of use and perceived usefulness on

the acceptance of a system. One of the possible explanations of the weak effect of attitudeon system usage is that, in universities, students are told to use WebCT by their lecturers asa specic subject requirement. As a result, a positive attitude among students towards WebCTmay not generate an increase in the actual use of the system if lecturers do not require them touse the system.

This study provides additional insights for the educator in implementing Web-based learningsystems in their online courses. The ndings indicate that perceived ease of use is a key interveningvariable linking the exogenous variable technical support with perceived usefulness, attitudeand system usage. The importance of perceived ease of use is further illustrated by its direct effecton attitude and even on system usage. This suggests that, in implementing a system, the focus



15/18

should be placed on fostering the self-condence of individuals and their perceptions concerningthe system. If users are struggling, they may actually believe that the system is too hard to use andthat the benets of using the system in terms of performance are outweighed by the effort of usingit. Eventually, they may become reluctant to use the technology, thus defeating the purpose of introducing it. Future study can extend the results of this study by investigating the area of self-efficacy which addresses one s belief in one s abilities to be able to accomplish a specic task,such as successfully using Web-based learning system. As a future direction of research, we willinclude self-efficacy as a factor for further investigation.

As with all empirical research, this study has a few limitations. First, the majority of the respon-dents in this study were full-time undergraduate students. The generalization of the study s nd-ings should be done with care. The lifestyles, educational backgrounds and experiences of thesestudents may differ from those of part-time students. More research is needed to investigate otherswho are not in engaged in full-time undergraduate studies. Second, there is the relatively low R 2

value in the proposed model, so that the perceived ease of use, perceived usefulness, and attitudeexplained 12% of the variance in using Web-based learning systems. This implies that other fac-tors inuence the students to use Web-based learning systems. Future research should exploreother variables that may have an effect on user beliefs and usage, such as peer support, peer pres-sure, an e-learning culture in the institution, infrastructure pressure from lecturers, availabilityand accessibility.

Acknowledgement

This research was supported in part by The Hong Kong Polytechnic University under the pro- ject grant number G-T954. The author is grateful for the constructive comments of the referee onan earlier version of this paper.

Appendix A. Questions used in the study

Technical supportTS1 A help desk is available when there is technical problemTS2 A hotline is available when there is technical problem

TS3 Fax enquiries can be made when there is technical problemTS4 Web-based enquiries can be made when there is technical problemTS5 E-mail enquiries can be made when there is technical problemTS6 The training on the operation of the Web-based learning system is sufficient

Perceived usefulnessPU1 Having Web-based learning in the curriculum enables me to learn more

efficientlyPU2 Web-based learning improves my academic performancePU3 Web-based learning enhances the effectiveness of my learning



16/18

Appendix A (continued )

PU4 Web-based learning makes it easier to learn in university

PU5 Web-based learning gives me greater control over learningPU6 Overall, I nd Web-based learning to be advantageous to my learning

Perceived ease of usePEOU1 Learning to operate a Web-based learning system is easy for mePEOU2 I believe that it is easy to get the Web-based learning system to do what I

want it to doPEOU3 The process of using a Web-based learning system is clear and

understandablePEOU4 It is easy for me to become skillful in using the Web-based learning systemPEOU5 Overall, I believe that the Web-based learning system is easy to use

AttitudeA1 Web-based learning is funA2 Using Web-based learning is a good ideaA3 A Web-based learning system provides an attractive learning environmentA4 Overall, I like using Web-based learning

Intention to useIU1 To the extent possible, I would use Web-based learning system to do

different things, from downloading lecture notes and participating in chatrooms to learning on the Web

IU2 I intend to increase my use of the Web-based learning system in the future

System usageSU1 On average, how often do you use the Web-based learning system?SU2 On average, how many hours per week do you spend using the Web-based

learning system?

References

Adams, D. A., Nelson, R. R., & Todd, P. A. (1992). Perceived usefulness, ease of use, and usage of informationtechnology: a replication. MIS Quarterly, 16 (2), 227247.

Ajzen, I., & Fishbein, M. (1980). Understanding attitudes and predicting social behavior . Engelwood Cliffs, N.J.:Prentice-Hall.

Bajaj, A., & Nididumolu, S. R. (1998). A feedback model to understand information system usage. Information &Management, 33 , 213224.

Chau, P. Y. K. (1996). An empirical assessment of a modied technology acceptance model. Journal of ManagementInformation Systems, 13 (2), 185204.

Chen, L. D., Gillenson, M. L., & Sherrell, D. L. (2002). Enticing online consumers: an extended technology acceptanceperspective. Information & Management, 39 , 705719.



17/18

Davis, F. D. (1989). Perceived usefulness, perceived ease of use and user acceptance of information technology. MIS Quarterly, 13 (3), 319340.

Davis, F. D. (1993). User acceptance of information technology: system characteristics, user perceptions and behavioralimpacts. International Journal of ManMachine Studies, 38 (3), 475487.

Davis, F. D., Bagozzi, R. P., & Warshaw, P. R. (1989). User acceptance of computer technology: a comparison of twotheoretical models. Management Science, 35 (8), 9821003.

Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention and behavior: An introduction to theory and research . Reading,MA: Addison-Wesley.

Gunasekaran, A., Mcneil, R. D., & Shaul, D. (2002). E-learning: research and applications. Industrial and Commerical Training, 34 (2), 4453.

Hair, J. F., Anderson, R. E., Tatham, R. L., & Black, W. C. (1995). Multivariate data analysis with readings . EnglewoodCliffs: Prentice-Hall.

Hofmann, D. W. (2002). Internet-based distance learning in higher education. Tech Directions, 62 (1), 2832.Hoyle, R. H. (1995). The structural equation modeling approach: basic concepts and fundamental issues. In R. H.

Hoyle (Ed.), Structural equation modeling: Concepts, issues, and applications (pp. 114). Thousand Oaks, CA: SagePublications.

Hu, P. J., Chau, P. Y. K., Sheng, O. R. L., & Tam, K. Y. (1999). Examining the technology acceptance model usingphysician acceptance of telemedicine technology. Journal of Management Information Systems, 16 (2), 91112.

IEEE Learning Technology Standards Committee IEEE LTSC Glossary. Available from http://ltsc.ieee.org .Igbaria, M. (1990). End-user computing effectiveness: a structural equation model. OMEGA, 18 (6), 637652.Igbaria, M., Zinatelli, P., Cragg, N., & Cavaye, L. M. (1997). Personal computing acceptance factors in small rms: a

structure equation model. MIS Quarterly, 21 (2), 279305.International Data Corporation (IDC#23539). (2000). Distance learning in higher education: Market forecast and

analysis. pp. 19992004.Kekkonen-moneta, S., & Moneta, G. B. (2002). E-learning in Hong Kong: comparing learning outcomes in online

multimedia and lecture versions of an introductory computing course. British Journal of Educational Technology,33(4), 423433.

Kumar, V. (1999). Essentials of marketing research . New York: John Wiley & Sons.

Lederer, A. L., Maupin, D. J., Sena, M. P., & Zhuang, Y. (2002). The technology acceptance model and the WorldWide Web. Decision Support Systems, 29 , 269282.Lin, J. C. C., & Lu, H. (2002). Towards an understanding of behavioural intention to use a web site. International

Journal of Information Management, 20 , 197208.Lou, H., Luo, W., & Strong, D. (2000). Perceived critical mass effect on groupware acceptance. European Journal of

Information Systems, 9 , 91102.Lu, J., Yu, C. S., & Liu, C. (2003). Learning style, learning patterns, and learning performance in a WebCT-based MIS.

Information & Management, 40 (6), 497507.Mirani, R., & King, W. R. (1994). Impacts of end-user and information center characteristics on end-user computing

support. Journal of Management Information Systems, 11 (1), 141166.Morris, M. G., & Dillon, A. (1997). How user perceptions inuence software use. IEEE Software, 14 (4), 5865.Morss, D. A. (1999). A study of student perspectives on Web-based learning: WebCT in the classroom. Internet

Research: Electronic Network Applications and Policy, 9 (5), 393408.Nunnally, Y. J. (1978). Psychometric theory . New York: McGraw Hill.O cass, A., & Fenech, T. (2003). Web retailing adoption: exploring the nature of internet users Web retailing behaviour.

Journal of Retailing and Consumer Services, 10 , 8194.Ralph, W. (1991). Help! The art of computer technical support . California: Peachpit Press.Selim, H. M. (2003). An empirical investigation of student acceptance of course websites. Computers & Education, 40 ,

343360.Sumner, M., & Hostetler, D. (1999). Factors inuence the adoption of technology in teaching. Journal of Computer

Information Systems, 40 (1), 8187.Szajna, B. (1996). Empirical evaluation of the revised technology acceptance model. Management Science, 42 (1), 8592.

http://ltsc.ieee.org/http://ltsc.ieee.org/


18/18

van der Heijden, H. (2003). Factors inuencing the usage of websites: the case of a generic portal in The Netherlands.Information & Management, 40 , 541549.

van der Heijden, H., Verhagen, T., & Creemers, M. (2003). Understanding online purchase intentions: contributionsfrom technology and trust perspectives. European Journal of Information Systems, 12 , 4148.

Wernet, S. P., Olliges, R. H., & Delicath, T. A. (2000). Postcourse evaluation of WebCT (Web course tools) classes bysocial work students. Research on Social Work Practice, 10 (4), 487504.

Williams, P. (2002). The learning Web: the development, implementation and evaluation of Internet-basedundergraduate materials for the teaching of key skills. Active Learning in Higher Education, 3 (1), 4053.

Withnam, S. A., Krockover, G. H., Ridgway, K. D., & Zinsmeister, W. J. (2002). Lessons online. Journal of CollegeScience Teaching, 32 (4), 264269.


indexed_4_ngai_empirical examination of the adoption of webct using tam

Documents