communication strategies: an interplay between proficiency...

INDIAN JOURNAL OF APPLIED LINGUISTICS

VOL. 34, NO. 1-2, JAN-DEC 2008

Communication Strategies: An Interplay

between Proficiency and Gender

HUSSEIN SHOKOUHI

FARZAD ANGAMEH

Shahid Chamran University of Ahwaz, Iran

ABSTRACT

This paper is intended to investigate the interplay between proficiency

and gender in the use of communication strategies. Sixty Iranian

university male and female subjects studying English took part in the

experiment and performed two tasks: word recognition and picture-

story narration. The results indicate that proficiency had a more

perceptible effect on the frequency and types of communication

strategies. Tasks also had a strong effect on the number and type of

strategies chosen. Gender did not yield any significant results except

in the case of low proficiency level of female participants. The reason

was attributed to the subject of study and formal educational system.

INTRODUCTION

Communication strategies (henceforth CSs) are linguistic or non-

linguistic devices that learners apply to solve instantaneous, often

unforeseen, problems that rise in the course of communication. They

are the means that bridge the gap caused by the incongruity between the

speaker’s intentions and his linguistic repertoire (Færch & Kasper

1983a: 36, 1984: 47).

Research on the use of CSs is numerous. Bialystok (1983), for

example, by looking into the effects of different strategies such as L2-

based, L1-based, etc. used by different learners in a picture-story

reconstruction found that the more proficient learners among her

intermediate subjects used more L2-based strategies. Paribakht (1985,

1986) through a word and concept-identification task explored types

and frequency of CSs across three groups at three different proficiency

levels. She observed that there is a direct relation between the speakers’

use of CSs and their level of the target language proficiency.

HUSSEIN SHOKOUHI & FARZAD ANGAMEH

122

Proficiency and task-related factors were also the goal of a study

carried out by Poulisse & Schils (1989). They selected three groups of

learners at three levels of proficiency (low, intermediate, and advanced)

and asked them to participate in three tasks: a) a picture-story

description, b) a story retell task, and c) a 20-minute interview. They

saw that there is an inverse relation between number of CSs used and

the proficiency level. That is, the more proficient the learners are, the

fewer strategies they use. However, they found that task-related factors

had a more perceptible effect on the use of strategies than the

proficiency level.

Kocoglu (1997), in a study on the relation between gender and

proficiency in the utilization of CSs, employed ten subjects paired with

ten native speakers as participants in casual conversations. She found

that gender of the native speaker can pose an influence on the use of

CSs since more strategies were utilized by female than by male

interlocutors. She concluded that the communication success depends

on pairing and on interlocutor’s personalities.

Most of the literature in the field seems to be lacking a general

comprehensive perspective of CSs. Many researchers investigated one

or two dimensions at a time, such as personality, L1 background,

effectiveness of the learner’s CSs, etc., with a small population.

However, as Firth & Wagner (1997) and Poulisse (1997) have reported,

there could be more than one factor determinant in one’s linguistic

performance.

This study targets particularly to compensate for the shortcomings

seen in the previous research; mainly in terms of a) the investigation of

the role of gender, b) inclusion of two tasks, c) inclusion of a bigger

population, and d) extension of the framework of the study.

THE STUDY

This study undertook to investigate the interplay of two factors –

proficiency and gender – in two different tasks: word recognition and

picture-story narration. It was primarily intended to find out the

difference in the frequency of CSs employed by the more or less

proficient learners of each sex. Then, it was to see what types of CSs in

terms of quality will be employed in a given task by the more proficient

and less proficient learners of each sex, and what the effect of the task-

related factors will be in this regard.

COMMUNICATION STRATEGIES: AN INTERPLAY

BETWEEN PROFICIENCY AND GENDER

123

METHODOLOGY

Participants

At the outset, one hundred and thirty university students majoring in

English who had three years of education in the field in Iran took part in

this research. A general proficiency test was administered among the

participants to screen out 60 participants for the final test. Based on the

scores from the test, the 60 subjects were divided into four groups: a)

group 1: 15 High Proficient Male Participants (HPMP), b) group 2: 15

Low Proficient Male Participants (LPMP), c) group 3: 15 High

Proficient Female Participants (HPFP), and d) group 4: 15 Low

Proficient Female Participants (LPFP).

Tasks

Two criteria in the selection and use of CSs are, according to Færch &

Kasper (1984), “Problem-orientedness” and “potential-consciousness”.

However, levels of consciousness differ with respect to various layers

of language. The lexical layer, among other linguistic layers like syntax

and morphology, is admittedly selected more consciously (Blum-Kulka

& Levenston 1983: 120; Færch & Kasper 1984: 54). In order to induce

subjects’ use of CSs when appropriate lexical item is lacking, tasks had

to be designed to meet these criteria: a) involve the learners’ attempt to

convey a meaning, b) neither encourage nor hinder the employment of

certain CSs, c) motivate the learners in a ‘problem-solving’ situation, d)

provide an incentive for the learner to attempt to tell difficult things,

and e) be kept under control so that the CSs for conveying of the items

in the tasks could be examined.

To this end, communicative tasks chosen in this study were

concept-identification tasks. They were word description and picture

story narration. The word description included four words: two concrete

and two abstract words. The choice of four words would provide a

rather homogenous condition and clear comparison. Choice of abstract

words was because, lacking visual clues, such words were expected to

place heavier linguistic and cognitive burdens on the speakers than

concrete nouns (Carter 1998: 192), hence a stronger struggle for more

use of CSs.

The picture story tells the account of an old hat-seller who decides

to take a nap under the shades of a tree, ignorant of the monkeys on the

tree. When he falls asleep, the monkeys take the chance and take away

his hats. When the old man wakes up, he sees that the monkeys have

taken away his hats, and they do nothing but imitate him. He comes up


124

with a plan and throws his remaining hat on the ground. Seeing this, the

monkeys do the same thing and the old man manages to get his hats back.

Procedure

In the first task, the four words were written on separate 5 by 7 inches

cards along their Persian (Farsi) equivalent. The subjects were admitted

into a room individually to avoid any contact between them during the

experiment, and they were not informed as being high or low proficient

learners.

Each data collection session consisted of three phases: warm-up,

relating the lexical item, and narrating the picture story. In the warm up

stage, the interviewer explained to each subject the nature of the task

and informed him/her clearly what to do afterwards. Following this

phase, the first task started. In the first task, relating the lexical item,

each subject was given a card one at a time and was asked to describe

the word to the interviewer without using the word itself. This was done

for all the four words. Interaction between the subject and the

interviewer continued until either the interviewer was convinced of the

appropriate description of the word or the subject gave up.

In the picture story narration, six sequential drawings of a picture

were given to each subject individually, and they were asked to relate

them to the interlocutor. They were told to take their time and begin

whenever they felt ready. During all these procedures, the interviewer

remained fairly callous and non-responsive. This is based on Færch &

Kasper (1984), whereby although the interaction remains

unidimensional, it still is considered communicative. All the

participants’ responses were video and audio-taped for later coding and

analysis. After a preliminary screening of the videotapes, all the

interviews were transcribed for all the tasks employed from all sessions.

Taxonomy

The first step for data analysis was to develop a taxonomy. Reasons for

such a procedure could be attributed to the diversity of the population,

nature of the tasks and type of the tasks. The insight to develop a

taxonomy was taken from the previous studies conducted by Færch &

Kasper 1983a & b; Bialystok 1983; Paribakht 1985 and Chen 1990,

among others. The taxonomy included five categories: 1) L2-based

strategies), 2) L1-based strategies), 3) general knowledge (World

Knowledge), 4) no information (Avoidance), and 5) gestures

(Paralinguistic parameters). Each category including its specifications is

discussed below.



125

1. L2-based strategies

a) Synonym: producing the same meaning by alternative lexical item

self-reliance for (self-confidence)

HPMP71

b) Circumlocution: describing the characteristic features of the

intended item. This involves substantial restructuring of the

message, often resulting in awkward verbosity

do the job that the man was doing for (imitating)

LPFP7

c) Paraphrase: rendition of lexical item that is more appropriate and

more concise than circumlocution

There was a man whose job was to sell hats for (hat seller)

HPMP11

d) Super-ordinate terms: using a high coverage word for a

subordinate term

It is a kind of animal for (monkey)

LPMP1

e) Generalization: substituting one word with another with the same

meaning while the context for the word is different

made of something like wire for (string)

LPFP12

2 L1-based strategies

a) Code-switching: appealing to another linguistic code, typically a

mother tongue word, to solve the problem

I think this instrument have four sim for (string)

LPMP11

b) Literal translation: the process of translating a lexical item, an

idiom, etc., from L1 literally

Khoshal shod for (he got happy)

LPMP14

3. Knowledge-based strategies

a) Exemplification: providing examples to indicate the meaning of

the item

Uh, when suppose you are in court and you are a judge, you have

to be ... have with the .. with the two men that are in front of you,

you have to behave with them equally for (justice)

LPMP4

b) General knowledge: providing general characteristics of

somebody or something


126

I think it can be played in two ways like Flamingo and classical,

OK for (guitar)

LPMP7

c) Simile: use of comparison to relate meaning

has three things like hair for (string)

LPFP8

4. Avoidance

a) Message abandonment: leaving the message unfinished because

of some language problem

It is the name of a set in music that is very common in Iran and but

I don’t exactly know about for (guitar).

LPFP1

b) Topic reduction: reducing the message by avoiding certain

structures or topics considered problematic.

For example, LPFP12 avoided mentioning the old man’s amazement

and anger when he saw that his hats were taken by the monkeys.

5. Paralinguistic strategy:

Use of gestures to indicate a concept either by accompanying the verbal

output or completely replacing it

a) Accompanying verbal output:

For example, LPFP6 indicated how the guitar is played by

demonstrating the movements of the hand.

b) Replacing verbal output

Then, the man is thinking about and doing like this (scratching his

head)

LPMP3

RESULTS AND DISCUSSION

A total of 735 strategies of all different types outlined above were used

by all the participants in approximately 490 minutes of talk. As far as

the relation between the proficiency and the frequency counts is

concerned, seen in Table 1, the CSs used by the low proficient were 499

(67.90%), which is a little more than twice the number used by the high

proficient ones, 236 (32.10%). Gender, in general, however, did not

produce any significant results in term of frequency (Table 2): the



127

number of CSs used by males is close to that of females (48.70% versus

51.30%, respectively). Table 3 combines proficiency and gender effects

on the number of CSs employed by the participants in each group. For

convenience, Figure 1 below pictures this information graphically. As

viewed, HPMP and HPFP used fewer number of strategies (15.24% and

16.87%) compared with LPMP and LPFP who respectively used

33.47% and 34.42% of the total strategies.

Table 1. Number of CSs in relation to Proficiency

HP LP

Total & Percentage 236 (32.10%) 499 (67.90%)

Table 2. Number of CSs in relation to Gender

Males Females

Total & Percentage 358 (48.70%) 377 (51.30%)

Table 3. Number of CSs in relation to Proficiency & Gender

HPMP LPMP HPFP LPFP Total

112 246 124 253 735

0%

5%

10%

15%

20%

25%

30%

35%

40%

Series1 15.24% 33.47% 16.87% 34.42%

H PM P LPM P HPFP LPFP

Figure 1. A Graph Bar of ‘Table 3’

Interplay between proficiency and gender on type count

Tables 4 and 5 below show breaking down of these figures in relation to

proficiency and gender of the participants respectively. Use of CSs

across both levels of proficiency (inter-group) indicate a great

difference between them. As Table 5 below demonstrates, number of

L2-based strategy in HP is surprisingly higher than LP (60.65% versus

39.35%). However, preponderance of the use of other CSs was clearly

ascertained by LP (76.51% of L1-based strategy was used by LP versus


128

23.49% by HP, 86.40% of world knowledge by LP in comparison to

13.60% for HP, etc.) LP also outnumbered HP in the use of avoidance

with 78.30% versus 21.70%.

Apart from these HP/ LP differences, the other large difference was

seen in paralinguistic strategy where 20% were used by males and 80%

by females. Type of specific CSs across both genders remained almost

the same. As Table 5 below shows, L2-based remained constant

between males and females (49.53% versus 49.77% respectively), L1-

based for males was 54.37% and for females 45.63%, world knowledge

for males was 48.80% versus 51.20% for females, avoidance 55.75%

for females and 44.25% for males, and paralinguistics were used in

equal number by both genders (50%).

Table 6 below shows breaking down of CSs according to each

group). As the analysis of CS types used by each group indicates (Table

6), L2-based strategy was used by HPFP more than any other group

(60). L1-based and world knowledge were favored by LPMP (65 and 59

respectively). Avoidance was used more by LPFP (114) while

paralinguistic strategy was used more by LPMP.

Table 4. Number and Percentage Type of CSs Types in relation to

Proficiency2

HP LP Total

L2 131 (60.65) 85 (39.35) 216 (29.39)

L1 35 (23.49) 114 (76.51) 149 (20.27)

WK 17 (13.60) 108(86.40) 125 (17.00)

A 51 (21.70) 184 (78.30) 235 (31.97)

P 2 (20.00) 8 (80.00) 10 (1.37)

Total 236 (32.10) 499 67.90) 735 (100.00)

Table 5. Number and Percentage of Types of CSs in relation to Gender

Males Females Total

L2 107 (49.53) 109 (49.77) 216 (29.39)

L1 81 (54.37) 68 (45.63) 149 (20.27)

WK 61 (48.80) 64 (51.20) 125 (17.00)

A 104 (44.25) 131 (55.75) 235 (31.97)

P 5 (50.00) 5 (50.00) 10 (1.37)

Total 358 (47.70) 377 (51.30) 735 (100.00)

Table 6. Number of CSs used in each Category by all Four Groups

HPMP LPMP HPFP LPFP Total

L2 60 47 71 38 216

L1 16 65 19 49 149



129

WK 2 59 15 49 125

A 34 70 17 114 235

P - 5 2 3 10

Total 112 246 124 253 735

Interplay between proficiency, gender, and task

In word recognition (WR), learners applied far more strategies than

picture story narration (PSN); that is 422 versus 313, respectively. Table

7 indicates CSs used by different proficiency groups. LP participants

applied more CSs in both tasks (298 in WR and 201 in PSN). However,

gender remained somewhat neutral to task-related factors (Table 8).

Males and females used 205 and 217 respectively in WR. In PSN males

used 153 and females used 160 CSs. The important finding was that

qualitatively each task brought forth certain CSs in higher frequencies

than others. In PSN, as seen in Table 9, avoidance most distinctly (162)

outnumbers other strategies while world knowledge, L2 and L1-based

strategies remained approximately at the same level in WR. Figure 2,

which is a graph bar for Table 9 for the reader’s convenience,

demonstrates CSs in terms of participants and task respectively. It can be

seen that avoidance with 51.76% was the most used CSs in PSN while

world knowledge came first in WR (29.15%).

Table 7. Number and Percentage of CSs in each Task with regard to

Proficiency

HP LP Total

Word Recognition 124 (29.38) 298 (70.62) 422 (57.41)

Picture Story Narration 112 (35.78) 201(64.22) 313 (42.59)

Total 236 (32.10%) 499 (67.90%) 735

Table 8. Number and Percentage of CSs in each Task concerning Grammar

Males Females Total

Word Recognition 205 (48.58) 217 (51.42) 422 (57.41)

Picture Story Narration 153 (48.89) 160 (51.11) 313 (42.59)

Total 358 (48.70) 377 (51.30) 735

Table 9. Number and Percentage of CSs in each Task

L2 L1 WK A P Total

Word Recognition 114

(27.01)

106

(25.12)

123

(29.15)

73

(17.30)

6

(1.42)

422

(57.41)

Picture Story Narration 102

(32.59)

43

(13.74)

2

(0.64)

162

(51.76)

4

(1.27)

313

(42.59)


130

0 %

1 0 %

2 0 %

3 0 %

4 0 %

5 0 %

6 0 %

L 2 2 7 .0 1 % 3 2 .5 9 %

L 1 2 5 .1 2 % 1 3 .7 4 %

W K 2 9 .1 5 % 0 .6 4 %

A 1 7 .3 0 % 5 1 .7 6 %

P 1 .4 2 % 1 .2 7 %

W o rd R e c o g n it io n P ic tu re S to ry N a rra t io n

Figure 2. CS Types in each Task

Analyses of the participants’ responses indicate a significant difference

caused by the proficiency level of the participants. Gender, however,

did not generally yield great differences among the subjects’ responses,

except for one strategy. In the same vein task-related factors were also

found to play a crucial role in the selection of specific strategies.

Proficiency and frequency

Based on the results, we can assert that there is a difference in the

number of strategies used by the high and low proficient participants, as

shown in Table 1 above. A t-test on the mean number of CSs by the

four groups, as shown in Table 10 below, reveals that there is a

statistically significant difference between the two groups at the level of

0.05 of mean differences.

Table 10. T - Test results of Proficiency Groups in the Number of CSs.

*p< 0.05

Group No. of

Subjects Mean σ t df Sig. Outcome

HP 30 7.8667 1.4559

LP 30 16.6333 2.0592 -19.040 8 .000* HP<LP

Further examination of the results indicates that this double use in the

number of CSs for LP remained sustained as figure 4.1 indicates –

HPMP 15.24%; LPMP 33.47%; HPFP 16.87%; and LPFP 34.42%.

Tukey HSD test, shown in Table 11 below, was conducted to explore

the statistical difference between the four groups.



131

Table 11. Tukey HSD Test results of the Frequency of CSs by the Four

Groups. *p< 0.05

Group Groups Mean Std. Error Sig. Outcome

LPMP -8.9333* .6510 .000 HPMP<LPMP

HPFP -.8000 .6510 .611 n.s HPMP

LPFP -9.4000* .6510 .000 HPMP<LPFP

HPMP 8.9333* .6510 .000 LPFP>HPMP

HPFP 8.1333* .6510 .000 LPMP>HPFP LPMP

LPFP -.4667 .6510 .890 n.s

HPMP .8000 .6510 .611 n.s

LPMP -8.1333* .6510 .000 HPFP<LPMP HPFP

LPFP -8.6000* .6510 .000 HPFP<LPFP

HPMP 9.4000* .6510 .000 LPFP>HPMP

LPMP .4667 .6510 .890 n.s LPFP

HPFP 8.6000* .6510 .000 LPFP>HPFP

As can be seen, high proficient of both genders (HPMP and HPFP) had

a statistically significant mean difference at the level of 5 percent with

the low proficient of both genders (LPMP and LPFP).

This irrefutable finding suggests that there is a reverse or negative

relationship between number of strategies and the proficiency level of

the students. That is, as the students’ level of proficiency rises from low

to high, number of CSs will decrease. Findings of this study are in line

with the previous studies by Bialystok (1983); Færch & Kasper (1983a,

1983b); Ellis (1984); Paribakht (1985 1986); Poulisse & Schils (1989);

Chen (1990); Liskin-Gasparro (1996); and Vandergrift (1997). With an

increase in the level of proficiency, the frequency intends to opt for a

descending order for less use of these strategies. These findings can be

explicated in the following terms.

Firstly, there is the degree of L2 knowledge. CSs are problem-

solving mechanisms which serve to compensate for the imperfections

one has in the target language. This imperfect and immature knowledge

in L2 is more severely and evidently seen in the low proficient learners.

This is because they have to carry out their task with the limited assets

in their possession. They ought to seek other alternatives to compensate

for these shortcomings. They have to take more time and effort to

rephrase and modify their original message. This in itself will increase

the number of CSs. As a proof to this claim, if the number of CSs used

by each group is taken as a rough index of number of problems subjects

face, it can be seen that the LP faced more communicative problems;

hence, more use of CSs. HP learners have richer L2 knowledge and

they can easily use this asset to overcome communicative problems


132

they encounter. Mangubhai (1991: 270) asserts: “the greater the

learner’s proficiency in the SL, the more resources there are available to

the learners in the process of meaning construction, and fewer problems

encountered.”

Secondly, command of the L2 knowledge is another factor that

determines the frequency occurrence. Command in this sense pertains

to the degree of awareness which a person has about his/her abilities in

the target language. One indicator of this command is the degree of

automatization of one’s language. Automatization refers to the rate of

articulation; the ease and speed of activating a procedure (Færch &

Kasper 1985: 127). Færch & Kasper (1983b: 219) believe that learners

normally have a lower degree of automatization of their interlanguage

(IL) than native speakers do. What Færch & Kasper deduce from their

study of learners’ automatization is that rate of articulation is likely to

vary with the learners’ proficiency level. For HP learners, their speech

enjoys more transitional smoothness (Færch & Kasper 1983b: 235).

This is possible because HP subjects, due to their better command, are

better planners for the type of message they intend to say. The lack of

automatization frequently makes the learner of an unrelated language

unable to cope with the time pressure demanded in oral communication.

This command of the L2 gives HP subjects a better evaluation of the

limitation of their resources, and more accurate prediction of the

problems that they might encounter.

This automaticity is not the case for the low proficient learners.

They have less capability to plan in advance because of their limited

and tentative knowledge as a result of which more troubles might come

on their way, hence more CSs. Evidence for this claim comes from

another line of research. In the study of the phenomenon of monitoring

and self-repairs, Kormos (1999: 331) quotes O’Connor (1988) who

analyzed the speech of three beginning and three advanced American

speakers studying French. O’Connor had found that less proficient

speakers resorted to more corrective repair, while advanced students’

self-corrections are likely to be more anticipatory in nature “that is, they

would be used to avoid possible breakdowns or communication

difficulties.” Thus, command of L2 endows learners with more

anticipatory power as their proficiency increases.

Proficiency and types of CSs

Patterns of types of CSs differed with regard to each group. Proficiency

groups differed in their choice of different CSs. The reason is that

proficiency interfered to cause great changes in the pattern of CSs



133

among different groups. To obtain more reliable results of the

differences between HP and LP groups and make the results less

speculative, a t-test was performed on the mean proportional scores of

the two groups. Table 12 below shows the results of the t-test:

Table 12. T-test results of Proficiency Groups in terms of the Number

of CSs. *p< 0.05

Types Group No. of

Subjects Mean σ T f Sig. Outcome

HP 30 4.3667 1.4499 L2

LP 30 2.8333 1.3412 -4.252 8 .000* HP >LP

HP 30 1.667 .8743 L1

LP 30 3.8000 1.5403 -8.144 8 .000* HP<LP

HP 30 .5667 .9714 WK

LP 30 3.6000 1.3025 -10.225 8 .000* HP<LP

HP 30 1.7000 1.5347 A

LP 30 6.1333 2.2854 -8.821 8 .000* HP<LP

HP 30 6.667E-02 .2537 P

LP 30 .2667 .5208 -1.891 8 .064 n.s.

As seen in Table 12, there is a meaningful mean between CSs used in

each separate group. It can be seen that HP used L2 more than other

CSs while LP used other CSs in greater numbers. P yielded no

significant results for any of the group. Therefore, HP used more L2-

based and they yield the following pattern:

L2-based → avoidance → L1-based → world knowledge → paralinguistic

LP provided the following pattern:

Avoidance → L1-based → world knowledge → L2-based → paralinguistic

Strategy use does not seem to be a casual phenomenon. Paribakht

(1985: 141) argues that strategy use is transitional and dynamic. She

adds that “learners use of CSs has specific characteristics at different

developmental stages of their interlanguages.” Chesterfield &

Chesterfield (1985: 56) affirm that increasing proficiency in the second

language suggests ability to use strategies in more demanding ways. A

certain level of what we may call ‘continuum effect’ that is required in

the selection of CSs types.

Learners drew upon similar resources to compensate for their

incomplete knowledge; however, the proposition of each strategy

differed with the proficiency level of the students. HP used more L2-


134

based CSs. One explanation could be that HP learners obviously have

better grasp of the formal language. They are better planners and can

evaluate their resources beforehand. In this case continuum effect is at

the highest level. This greater command over the L2 and the ensuing

estimation of their abilities enables them to choose the most

economical, direct, relevant, and meaningful types of CSs. A few

examples of HP behavior below could shed some light on this

difference of strategy use.

a kind of musical instrument (L2-based: Guitar)

HPFP3

uhm to make a[n] equal attention to both sides (L2-based: Justice)

HPMP3

Nevertheless, choice of strategy differed in case of the low proficient

learners. They had little command over the target language and were more

uncertain about their resources. They had to depend on other ways such as

avoidance, their L1, knowledge of the world or paralinguistic CSs.

Avoidance is the most used CSs (235) among all. In the early

stages, the propensity for the occurrence of avoidance is highest. As

learners move towards the higher stages, L2-based strategies, which are

typical of upper stages, make themselves more evident and instead

avoidance strategies opt for lower percentages. By the use of avoidance,

LP preferred to give up relating the concept than putting up some effort

into explaining it. Blum-Kulka & Levenston (1983: 198) believe that

avoidance is due to lack of knowledge and assert that other strategies

must be found to fill in the semantic gap. This strategy is the least

demanding of CSs since what you are supposed to do is simply

circumvent the problem.

L1-based strategies provided LP with more trustworthy

mechanisms to draw upon than avoidance. In the case of L1-based,

continuum effect is one step up in comparison with avoidance. In this

strategy the learner tries to relate something. It is different from

avoidance where learners did not say anything. However, willingness to

relate something was limited to applying L1 knowledge. A few

examples help better understand this phenomenon:

Bardasht his hat (L1-based: picked up)

LPFP9

has three things like a hair (L1-based: strings)

LPFP7



135

In L1-based strategy, continuum effect increased but in the absence of

reliable L2 resources learners had to resort to the next dependable

source: knowledge of the first language. Olsen (1999: 201) who studied

grammar and vocabulary in texts which were written by Norwegian

learners of English believes that L1 serves as a reference and assistance

in cases students encounter a linguistic problem.

World knowledge was the least used strategy among the three. In

this case attempt to use one’s world knowledge will involve using one’s

basic knowledge. For instance,

A tool of pop music (world knowledge: guitar)

LPMP12

For example, when we want to take an exam, we must study a lot

and have it for do our exam very well.

(world knowledge: self-confidence)

LPMP2

Here effort had to be put into place to not only use correct, relevant

knowledge, but also linguistic means to make their expression possible.

This would in itself increase the cost for the LP who have to process

two operations: choosing relevant knowledge and culling proper

linguistic means. Continuum effect would proceed to induce the use of

108 world knowledge CSs.

From what has been said it can be claimed that continuum effect

has great influence over CSs selection and preference and that it is

determined by the level of proficiency of the learners.

What has emerged from the data suggests a hierarchy of occurrence

for the arch strategies which can be shown graphically below:

Avoidance L1-based World knowledge L2-based

Low proficiency High proficiency

Proficiency and task

An independent question was about the task-related factors. At this

level it was seen that task-based factors did exert an influence in

number and selection of specific strategies. As Table 9 and figure 2

demonstrate, qualitatively there is a big difference between number and

type of CSs used in each task: avoidance occurred in greater frequency

in PSN than other strategies.


136

In the first task (WR), subjects used more CSs than the second

(PSN). This behavior remained almost constant between HP and LP.

As already explained in methodology, WR task involved relating 4

lexical items: two concrete and two abstract words. In PSN, 6 pictures

were given to the subjects and they were asked to relate them. There

were lexical items in each picture which posed a challenge to the

subjects. In other words, while the first one required relating the words

directly, the second one placed lexical items in the texture of the story.

The difference WR bore was that subjects had to communicate them or

avoid them. They had to come face to face with the problem that each

word presented. The second task, however, offered subjects the

opportunity to come to terms with the problem by ignoring it or choose

any other strategy based on their proficiency and continue narrating the

story. (This willingness to communicate a concept has a direct

relationship with types of CSs, as will be explained below).

Telling words about words brought its own complications and

raised number of CSs. When each task is considered separately, a large

number of CSs was evident among LPs. This can be explained in terms

of the task demands mentioned above. Clearly, task demands did not

put any strains on the behavior of the HP, but influenced the other

group quantitatively.

Qualitatively, selection of CSs varied with regard to each task. As

figure 2 indicates, world knowledge was the most used CSs with

29.15% followed by L2 (27.01%), L1 (25.12%) and lastly Avoidance

(17.30%). As it can be seen, the gap between the use of each CSs is

very small. This phenomenon can be depicted for the simple reason of

the nature of the task. Subjects in WR had to use other lexical items to

relate the target items. The items themselves were everyday vocabulary

words in students’ speech. However, their approach in dealing with

each item and their agility in tackling the items were their biggest

challenge the task sought to examine.

As Table 9 shows, PSN witnessed the increase of Avoidance

(51.76%). Following avoidance were L2 (32.59%), L1 (13.74%), and

WK (0.64%). Students’ tendency to use avoidance can be explicated as

such: avoidance granted subjects with an opportunity to go around the

problem whenever they reach a dead-end in communicating the item. In

WR, the option to avoid them was giving up the description task and

coming to full halt. Avoidance constituted only 17.30% in WR while

51.75% in PSN. For example, subjects avoided relating such lexical

items as surprise, scratch, imitate, throw, fist, etc. Each task,

henceforth, carried its unique requirements and dictated its demands to



137

the experiment: WR communicating about the items, PSN

communicating the items themselves. The decisive factors are the

students’ abilities and styles in tackling the obstacles and the

requirements that each task presented

The next widely used CSs were L2-based communication strategy.

L2 occurrence next to avoidance is strange indeed, as they constituted

the two ends of the continuum of CSs occurrence. This can be

understood if again the nature of the task is taken into account. The

embedded items in PSN provided an interesting challenge to HP and

LP equivalently. LP approach was pointed out above in resorting to

Avoidance. HP’s approach, however, differed. They did not avoid the

problem, but related it, for example, in using repeated what the man

was doing (HPMP 5) to communicate the word imitation. They

confronted the problems most efficiently. Their plan intended to

respond to these items by applying their most effective devices, i.e. L2

CSs. While incorporation of avoidance reduced the effectiveness of the

story narrated by the LP, L2-based strategy employed by HP yielded

effective narration of the story. Other researchers have also proposed

stating L2 as the most effective CSs, too.

World knowledge shifted its position drastically. While world

knowledge was used 123 times in WR, in PSN it was used only twice,

which made it the least used communication strategy. The embedded

items provided the least appeal to world knowledge in PSN. Reason for

such an occurrence can be ascribed to the story-telling phenomenon. It

was concrete and tangible enough for the narrator and listener so as not

to impose too much burden. Unlike WR, whereby lexical items were

proposed in isolation, participants in PSN had to create a context for

the words and all types of pragmatic, linguistic and non-linguistic

devices to create a background for the words and facilitate the task of

the listener. Unlike WR, PSN made little cognitive load on both the

listener and the narrator. LPFP4, for instance, used her world

knowledge to relate self-confidence by saying: it’s a sense it is about

psychology. We need some power in our spirit to do well our works.

This reduced the number of world knowledge in PSN, since some of it

could be provided by the elements of the story.

Gender

One of the particular interests of this study was examining the role that

gender could have played in the selection of CSs. Our findings indicate

that gender’s role was characterized with near neutrality. As Table 2 has

revealed, males used 358 (48.70%) and females 377 (51.30%). To see if


138

the same result could be sustained statistically, a t-test was performed.

As seen in Table 13 below, there is no difference between males and

females at the level of 0.05.

Table 13. T- test results of Proficiency Groups in terms of the Number

of CSs. p < 0.05

Group No. of

Subjects Mean σ t df sig. Outcome

Males 30 11.9333 4.8419

Females 30 12.5667 4.7393 .611* 58 .611 n.s.

Qualitatively, as Table 6 above demonstrated, not much difference was

seen in the types of CSs between the two groups either. A t-test was

conducted to explore a deeper statistics of the results, as seen in Table

14 below. It can be seen that there is no statistically significant

difference between males and females in the selection of different types

of CSs at the level of 0.05.

However, further analysis of the results exhibits a minor difference

in terms of types among LPMP and LPFP in the use of avoidance. As

Table 6 indicates, LPMP applied 70 while LPFP resorted to 114 CSs. In

the least it may argued that gender exerted some effect in the choice of

one strategy; that is, avoidance. To further validate the results, a Tukey

HSD test was conducted (Table 15). As seen in the table, unlike the

other participant groups, the difference between LPMP and LPFP in the

selection of avoidance is statistically significant at the level of 0.05.

Table 14. Results of T-test in the use of CSs by Males and Females. p <

0.05

Types Gender No. of

Subjects Mean σ T df Sig. Outcome

Male 30 2.7000 2.0026 L2

Female 30 2.2667 1.6174 -.161 58 .872 n.s

Male 30 2.0333 2.1891 L1

Female 30 2.1333 1.6132 .922 58 .360 n.s

Male 30 3.4667 2.0800

WK Female 30 4.3667 3.6054 -.201 58 .841 n.s

Male 30 .1667 .4611 A

Female 30 .1667 .3790

-

1.184 58 .241 n.s

Male 30 3.5667 1.6121 P

Female 30 3.6333 1.5862 .000 58 1.000 n.s



139

This difference can be attributed to the context of situation. Context of

situation in this case was a formal one. Eggins (1994: 64) mentions the

dimensions of this kind of role relationship. The interviews were

between male interviewer and his female interviewees. The power

relationship is unequal; the contact is occasional, and affective

involvement is low. The effect of this kind of tenor relation is

explained in terms of “face work” and ‘politeness’ below.

Face is defined by Goffman (1999: 306) as “a positive social value

a person effectively claims for himself by the line others assume he has

taken during particular contact.” Goffman (1999: 307) then mentions

avoidance as the surest method of preventing threats to face. One kind

of avoidance is keeping off topics and activities which head towards

expression of information inharmonious with the line of information

one is maintaining. This same strategy was used with the female

participants more than male participants. Females preferred to give up

any attempt to convey the information than resorting to other types of

strategies such as L1-based or world knowledge. These strategies would

reveal their incompetence and threat their face.

The other factor is politeness. Politeness is defined by Holmes (1995:

5) as a behavior which actively expresses positive concern for others in

addition to “non-imposing distance” behavior. She then recognizes two

types of politeness: positive politeness and negative politeness. LPFP

learners of the study were avoiding the latter. As illustrated above, the

context of situation is formal. Therefore, females were induced to appeal

to negative politeness. For females in the study being negatively polite

means avoiding disagreeable topics, and not expressing items which

would jeopardize their validity. They had to do this with their limited

resources without any attempt to reveal their incompetence. As argued in

face work, in the absence of dependable L2 resources, world knowledge

and L1 knowledge could be good candidates. Nevertheless, use of these

strategies equals use of positive politeness in which case they had to

provide necessary information and keep the channels of talk open. As

positive politeness cannot be used, use of the other choice remains in

place. This would include use of every means to avoid tackling problems

for which they could not manage.

Table 15. Results of Tukey HSD Test. p> 0.05

Group Groups Mean Std. Error Sig. Outcome

LPMP .8667* .5033 .322 HPMP>LPMP

HPFP -1.7333 .5033 .470 n.s

L2

HPMP

LPFP 1.4667* .5033 .026 HPMP>LPFP


140

HPMP -.8667* .5033 .322 LPMP<HPMP

HPFP -1.6000* .5033 .013 LPMP<HPFP LPMP

LPFP .6000 .5033 .634 n.s

HPMP .7333 .5033 .470 n.s

LPMP 1.6000* .5033 .013 HPFP>LPMP HPFP

LPFP 2.2000* .5033 .000 HPFP>LPFP

HPMP -1.4667* .5033 .026 LPFP<HPMP

LPMP 1.6000 .5033 .634 n.s LPFP

HPFP -2.2000* .5033 .000 LPFP<HPFP

LPMP -3.2667* .4422 .000 HPMP<LPMP

HPFP -.2000 .4422 .969 n.s HPMP

LPFP -2.2000* .4422 .000 HPMP<LPFP

HPMP 3.2667* .4422 .000 LPMP>HPMP


LPFP 1.0667 .4422 .086 n.s

HPMP .2000 .4422 .969 n.s

LPMP -3.0667* .4422 .000 HPFP<HPFP HPFP

LPFP -2.0000* .4422 .000 HPFP<LPFP

HPMP 2.2000* .4422 .000 LPFP>HPMP

LPMP -1.0667 .4422 .086 n.s

L1

LPFP

HPFP 2.0000* .4422 .000 LPFP>HPFP

LPMP -3.8000* .4012 .000 HPMP<LPMP

HPFP -.8667 .4012 .147 n.s HPMP

LPFP -.8667* .4012 .000 HPMP<LPFP

HPMP -3.1333* .4012 .000 LPMP>HPMP


LPFP 2.9333 .4012 .353 n.s

HPMP .8667 .4012 .147 n.s


LPFP -2.2667* .4012 .000 HPFP<LPFP

HPMP 3.1333* .4012 .000 LPFP>HPMP

LPMP -.6667 .4012 .353 n.s

WK

LPFP

HPFP 2.2667* .4012 .000 LPFP>HPFP

LPMP -2.4000* .5888 .001 HPMP<LPMP

HPFP 1.1333 .5888 .000 n.s HPMP

LPFP -5.333* .5888 .000 HPMP<LPFP

HPMP 2.4000* .5888 .229 LPMP>HPMP


LPFP -2.9333* .5888 .000 LPMP<LPFP

HPMP -1.1333 .5888 .000 n.s


LPFP -6.4667* .5888 .000 HPFP<LPFP

HPMP 5.3333* .5888 .130 LPFP>HPMP

LPMP 2.9333* .5888 .811 LPFP>LPMP

A

LPFP

HPFP 6.4667* .5888 .547 LPFP>HPFP



141

LPMP -.3333 .1501 .130 n.s

HPFP -.1333 .1501 .547 n.s HPMP

LPFP -.2000 .1501 .811 n.s

HPMP .3333 .1501 .811 n.s

HPFP .2000 .1501 .547 n.s LPMP

LPFP .1333 .1501 .811 n.s

HPMP .1333 .1501 .811 n.s

LPMP -.2000 .1501 .547 n.s HPFP

LPFP -6.6667E-02 .1501 .970 n.s

HPMP .2000 .1501 .547 n.s

LPMP -.1333 .1501 .811 n.s

P

LPFP

HPFP 6.667E-02 .1501 .970 n.s

Besides this difference between LPMP and LPFP in selection of

avoidance, the results remained closely related. This homogeneity in

performance can be explained in terms of the subject of the study itself

and the formal environment of the subjects.

It is noteworthy to mention that CSs study per se constitutes one

dimension of foreign language communication. If the definition of CSs

provided by Færch & Kasper above is recalled, we understand that

these phenomena are concerned with problems that arise in the course

of communication and ways to deal with these obstacles. The evidence

in itself indicates a narrow scope with which the subject is concerned

while language use in general and foreign language use in particular are

much more diverse in application and wider in scope. Therefore,

expecting gender differences to occur might not have been a very

realistic view. To sum up, gender is a much-limited space for the CSs to

occur. On the other hand, subjects under study contributed to this

homogeneity of performance and negated any chance for diverse

outcome. Students in Iran including the subjects in the study seem to

undergo similar experiences in the learning of English. Learners of both

genders learn English through formal educational system. Likewise

teachers undergo similar practices in their training. Both the students

and the teachers are brought up in an L1 milieu. They do not experience

any real language contact with the speakers of the target language, nor

can they afford to have that luxury in the target language environment

itself. They are brought up in a space which lacks the social dimension

of language. Holliday (1994: 13) investigated the social dimension of

language. He distinguishes between macro and micro aspects of social

context: the macro context, he believes, comprises the wider societal

and institutional influences on what happens in the classroom. Micro

view involves a discourse or interactive content. Our classes are void of


142

almost both of these dimensions. What remains for the students to

maneuver is in the area of language proficiency or more linguistically

put ‘linguistic competence’. In this framework students might ascend to

higher levels of proficiency based on their personal effort and endeavor.

However, this effort will remain limited in the sense that it will not gain

operationality in a wider sociolinguistic perspective.

CONCLUSION This paper investigated interplay of two factors: proficiency and gender

in two tasks of word recognition (WR) and picture story narration

(PSN). The results of the analysis revealed that proficiency exercised a

crucial role in the selection of CSs. Gender, however, did not generally

yield much significant difference.

Types of CSs underwent some changes when proficiency was at

play. It was seen that with the increase in the level of proficiency,

choice of CSs opted for L2-based ones. Types in the lower levels of

proficiency exhibited a tendency towards the Avoidance strategy. Task-

related factors imposed quantitative and qualitative variations: in PSN,

subjects used less CSs, but in WR they used more. This was explained

in terms of the nature of the task and its demands. The task demand left

a bold imprint on the types of CSs. More avoidance was seen in PSN

while more balanced use of L2, L1, and world knowledge was observed

in WR.

It is worth mentioning that proficiency in both number and types of

CSs brought significant outcomes, but gender continued to be

uninfluential. However, LPFP showed more use of avoidance, which

was attributed to the context of situation. Context of situation reveals its

effect in “face work” and “politeness” phenomena. Reason for this

passive role was attributed to the narrow subject under study and the

formal environment. Focus on linguistic competence is most frequently

so intense that it leaves other aspects of language use unattended.

Depravity of contact with real native-like language use intensified this

situation.

NOTES

1. The letters are abbreviations for each proficiency group, and the number

after each abbreviation indicates the number each participant is labled. For

instance, HPMP7 indicates high proficient male participant number seven

and LPFP12 means low proficient female participant number twelve. For

the abbreviation of all groups, the reader is referred to section 3.1 below.

The italicized words, phrases, and clauses indicate CSs use on the part of



143

the learners. Words and phrases enclosed in brackets followed by the word

‘for’ indicate the translation of the italicized section.

2. The abbreviations stand for the following terms: L2=L2-based, L1=L1-

based; WK=World Knowledge; A =Avoidance; P =Paralinguistic

REFERENCES

Bialystok, E. 1983. Some factors in the selection and implementation of

communication strategies. In C. Færch & G. Kasper (Eds.), Strategies in

Interlanguage Communication (pp. 100-118). London: Longman.

Blum-Kulka, S. & Levenston, E. A. 1983. Universals of lexical simplification.

In C. Færch & G. Kasper (Eds.), Strategies in Interlanguage

Communication (pp. 119-139). London: Longman.

Carter, R. 1998. Vocabulary: Applied Linguistic Perspectives. 2nd ed. London:

Routledge.

Chen, S. Q. 1990. A study of communication strategies in interlanguage

production by Chinese EFL learners. Language Learning, 40, 155-187.

Chesterfield, R. & Chesterfield, K. B. 1985. Natural order in children’s use of

second language learning strategies. Applied Linguistics, 6, 45-59.

Eggins, S. 1994. An Introduction to Systemic Functional Linguistics. London:

Pinter.

Ellis, R. 1984. The communication strategies and the evaluation of

communicative performance. ELT Journal, 38, 39-44.

Færch, C. & Kasper, G. 1985. Perspectives on language transfer. Applied

Linguistics, 8, 111-136.

——. & Kasper, G. 1984. Two ways of defining communication strategies.

Language Learning, 34, 45-63.

——. & Kasper, G. 1983a. Plans and strategies in foreign language

communication. In C. Færch & G. Kasper (Eds.), Strategies in


——. & Kasper, G. 1983b. On identifying communication strategies in

interlanguage production. In C. Færch & G. Kasper (Eds.), Strategies in


——. & G. Kasper (eds.) 1983c. Strategies in Interlanguage Communication.

London: Longman.

Firth, A. & Wagner, J. 1997. On discourse, communication, and (some)

fundamental concepts in SLA research. The Modern Language Journal,

81, 285-300.

Goffman, E. 1999. On face-work: an analysis of ritual elements in social

interaction. In A. Jaworski & N. Coupland (Eds.), The Discourse Reader

(pp. 306-320). London: Routledge.

Holliday, A. 1994. Appropriate Methodology and Social Context. Cambridge:

Cambridge University Press.

Holmes, J. 1995. Women, Men and Politeness. London: Longman.

Kocoglu, Z. 1997. Abstract. The role of gender in communication strategies.

ERIC. Ed 409725.


144

Kormos, J. 1999. Monitoring and self-repair in L2. Language learning, 49, 303-

342.

Liskin-Gasparro, J. 1996. Circumlocution, communication strategies, and the

ACTFL proficiency guidelines: An analysis of student discourse. Foreign

Language Annals, 29, 317-330.

Mangubhai, F. 1991. The processing behaviors of adult second language

learners and their relationship to second language proficiency. Applied


Olsen, S. 1999. Errors and compensatory strategies: a study of grammar and

vocabulary in texts written by Norwegian learners of English. System, 27,

191-205.

Paribakht, T. 1986. On the pedagogical relevance of strategic competence.

TESL Canada Journal, 3, 53-66.

——. 1985. Strategic competence and language proficiency. Applied


Poulisse, N. 1997. Compensatory strategies and the principles of clarity and

economy. In G. Kasper & E. Kellerman (Eds.), Communication

Strategies: Psycholinguistic and Sociolinguistic Perspectives (pp. 49-64).

London: Longman.

——. & Schils, E. 1989. The influence of task and proficiency-related factors

on the use of compensatory strategies: A quantitative analysis. Language

Learning, 39, 15-48.

Vandergrift, L. 1997. The Cinderella of communication strategies: reception

strategies in interactive listening. The Modern Language Journal, 81,

494-505.

HUSSEIN SHOKOUHI

FARZAD ANGAMEH SHAHID CHAMRAN UNIVERSITY OF AHWAZ, IRAN

E-MAIL: <[email protected]>

communication strategies: an interplay between proficiency...

Documents