Willingness to Communicate in English : A Study of Malaysian Pre-Service English Teachers

Language instruction in Malaysia emphasizes the significance of the English language. This study investigates Malaysians’ willingness to communicate (WTC) in English as a second language (ESL). A hypothesized model that integrates WTC in English, communication apprehension and competence, motivation, and language learning communication strategies was tested using structural equation modelling. The results show that Language learning communication strategies directly affect motivation, self-perceived communication competence, and WTC in English. Further, the results also demonstrate that motivation influences the two components of communication confidence and influences WTC indirectly through the two variables self-perceived communication competence and communication apprehension. The final model correlates well with the data, thereby indicating the potential of using Language learning communication strategies with WTC constructs to account for ESL communication.


Introduction
Communication is an effective tool that can be used to positively or negatively influence people, and interaction is the principle means of communication among individuals.Furthermore, classroom learning is seen as a positive climate that could nurture student involvement and decrease passivity (Sidelinger & Booth-Butterfield, 2010).This highlighted the importance of finding real solutions to encourage oral communication among L2 learners in their classroom.However, the heuristic willingness to communicate (WTC) model, emerged from situations related to L2 use.Generally, there is a need to understand the various relevant linguistic, situational, individual, and communication strategies factors that play a vital role in students' WTC in L2.Previous research has focused much attention on learner characteristics, such as motivation, language anxiety, aptitude, and language learning strategies, and their influence on L2 learning (Gardner, 2009).Using WTC as an important means of enhancing English language capability, the current study aims to investigate the relationships among the variables considered to influence Malaysian learners' WTC in English.WTC in a second language has become a core concept of second language acquisition (SLA) and communication (Peng, 2007).Meanwhile, in the current study, language learning communication strategies can be used as a systematic technique to help learners overcome communication difficulties, which, in turn, could increase their self-confidence in class.Moreover, the current study supported Dörnyei's (1995) theory in which communication strategies play a significant role in helping L2 learners to vanquish their communication difficulties in conversation and make their messages more understandable to the listener.No previous studies have explored the effect of oral communication strategies on students' WTC using an L2, particularly in a classroom setting.However, because of the role of communication strategies in strengthening the target language interaction (Tarone, 1981;Faerch & Kasper, 1983;Dörnyei & Scott, 1997), communication strategies have been proven to have a significant effect on language performance (Rost & Ross, 1991;Dörnyei, 1995).Thus, the current study aims to incorporate these variables into the wider study on WTC, focusing on Malaysian university students in their ESL classrooms.

Language Learning Communication Strategies
Learning strategies are activities that are rationally selected by learners for enhancing their learning (Oxford, 1990).Tarone, Cohen, and Dumas (1976) were the first to perform research on the use of problem-solving behaviour through learners' communication strategies.However, Rubin (1975) defined good L2 learners as those who are good guessers, willing to communicate, express, and analyze situations in an L2 production, leaders of their own speech, and mindful and observant of the meaning of words they use in conversation.Dörnyei (1995) found that researchers present different types of classifications and taxonomies of communication strategies.However, Faerch and Kasper (1983) categorized communication strategies into two types: achievement strategies and reduction strategies.

Achievement Strategies
Enable learners to maintain their conversation goals, particularly in the classroom.There are several such strategies that are classified under achievement strategies, which are explained in the following manner.

Help-Seeking Strategies Are of Two Types
Appeal for help and asking for repetition.Appeal for help is used when interlocutors ask for help due to lack of knowledge of the target language.For example, learners may ask for help from their partners by saying, "I am sorry, I don't understand."Asking for repetition is used when interlocutors have not heard or understood their partners' speech; for example, "Please repeat."

Signals for Negotiation Strategies
Signals for negotiation comprise confirmation checks, comprehension checks, and clarification requests.It is used when interlocutors send a message for negotiation in order to solve communication difficulties.Confirmation checks are the confirmation of other people's preceding utterances by asking questions such as, "Do you mean that you got a high grade?"Comprehension checks are used by interlocutors to assure that their message has been understood and transferred to their partners; for example, "Did you get it?"Clarification requests are used when speakers do not understand the other speakers' preceding utterance by verbalizing questions or statements such as, "I don't understand.What do you mean?"

Modified Output Strategies
Modified output strategies are used when interlocutors modify their previous phrases in order to amend and continue the interaction with their partners.Moreover, these strategies help learners enhance their language skills.The following is an example of this strategy: Student: The conference started at 8 a.m.
Friend: Sorry, when did it start?Student: I thought the conference started at 10 a.m., not 8 a.m.

Time-Gaining Strategies
Time-gaining strategies are used when learners have difficulty conveying the message to their listeners.The conscious use of fillers, such as, "Umm" and "Oh …" enables them to keep the conversation going.

Response for Maintenance Strategies
Response for maintenance strategies comprises two types: providing active response and shadowing.The first type is characterized by utilizing positive comments or using other conversation gambits such as, "Really, it seems good".The next type, shadowing, includes the same or partial repetition of the interlocutor's foregoing phrases.The following example shows this type of strategy: Interviewer: We have a three-day celebration.
Student: Three days.Ok, I would like to join you.

Self-Repairing Strategies
Self-repairing strategies are used when interlocutors are faced with problems because of their lack of linguistic resources.They try to use appropriate linguistic expressions through paraphrasing, approximating, and restructuring strategies.Paraphrasing, or circumlocution, occurs when interlocutors use periphrasis and redundancy on the target object or action.For example, if learners do not know the word "yashmak," they replace it by saying "the veil that women use to cover their faces."Approximation is the strategy in which learners use an alternative word to estimate a term for the target word, such as, "egg-shaped" for "head."Restructuring is used when learners face difficulty in completing sentences, for example, "Have you eaten…?Do you have any food?"

Reduction Strategies
Reduction strategies are used when interlocutors feel pressurized, particularly in the testing stage.It comprises three types of strategies: 2.1.2.1 Message Abandonment Strategy Message abandonment strategy is employed when interlocutors are unable to finish a message because of the inability to find the appropriate words to communicate in the target language.Such as in the following example: Lecturer: There is no extra exercise paper available.

Interlanguage-Based Reduction Strategies
In this strategy, learners avoid topics and certain L2 language structures because of a shortage of linguistic resources.The following is an example of this strategy: Interviewer: The class will finish at 11:30 a.m.
Student: 11:30 … I heard it leaves at 10 a.m.However, Graham (1997) emphasized two purposes of learning communication strategies: to decrease students' anxiety and to increase students' willingness to participate in conversations.The current study was enhanced using the Oral Communication Strategy Inventory developed by Nakatani (2006).Nakatani conducted a survey of 400 EFL Japanese university students in order to understand strategies used to overcome speaking and listening difficulties.However, the current study focuses only on speaking as well as the strategies that are employed to cope with speaking problems.Language learning communication strategies is considered to be a very effective tool in helping learners initiate a certain degree of interaction (Gallagher-Brett, 2001) and motivating them to learn the language (Bandura, 1986).Therefore, the following hypotheses were derived: H1a: Language learning communication strategies is positively related to motivation to learn English.
H1b: Language learning communication strategies is positively related to students' Self-perceived communication competence.

Willingness to Communicate and the Components of Self-Confidence in an L2
The concept of WTC was originally introduced by McCroskey and Associates (McCroskey & Baer, 1985;McCroskey & Richmond, 1987, 1990) based on Burgoon's (1976) 2001) determined L2 WTC in terms of the four skills of speaking, reading, writing, and listening.This model was suggested to be less accurate in describing situations occurring in a language classroom (Weaver 2005).At that time, utilizing the Rash model, Weaver (2005) developed a model for measuring L2 WTC in the speaking and writing modes.Among 490 university students as respondents, the model proved its validity, reliability, and psychometric benefits (Weaver 2005).However, researchers in second/foreign language learning supported the idea that the rate of language proficiency among L2 learners is associated with their level of anxiety and apprehension (Horwitz, Horwitz, & Cope, 1986).Moreover, it was proved that a significant relationship exists between L2 speaking anxiety and oral achievement (Woodrow, 2006).However, Malaysian students have a high degree of L2 communication apprehension; this is supported by Mustafa and Zain (2009) in their study of Malaysian ESL learners in University Malaysia Sarawak, in which the participants recorded high levels of communication apprehension, with a mean score of 74.49.Anxiety is affected by many factors, such as intergroup tension, unpleasant experiences, and increased fear of comprehension (MacIntyre et al., 1998).According to Horwitz and Young (1990), "Anxiety about speaking a language can affect the quality of oral language production, making individuals appear less fluent than they really are" (p.56), which means that a high level of communication apprehension is associated with weak communication (McCroskey, 1997).However, communicative competence can be defined as, "adequate ability to pass along or give information; the ability to make known by talking or writing" (McCroskey & McCroskey, 1988: 109).Self-perceived communication competence pertaining to the state of reticence was highlighted by the works of Phillips (1968Phillips ( , 1984) ) in which he stated two major reasons for uncommunicative people to have reservations regarding their communication abilities: anxiety and a lack of communication skills.The two constructs, L2 competence and a lack of anxiety, combine to form one's linguistic self-confidence (Clément, 1980(Clément, , 1986)).More importantly, linguistic self-confidence proves to be the most forthright antecedent of L2 WTC (MacIntyre & Charos, 1996;Yashima, 2002;Clément, Baker, & MacIntyre, 2003).However, a negative relationship between these two variables was emphasized (MacIntyre, Noels, & Clément, 1997;Noels, Pon, & Clément, 1996).Moreover, the current study evaluates students' communication apprehension in four communication contexts-group discussions, meetings, interpersonal conversations, and public speaking-for the purpose of minutely investigating students' communication apprehension within their classrooms.Thus, the following hypotheses were highlighted: H2a: Communication apprehension is negatively related to students' willingness to communicate in L2.
H2b: Communication apprehension is negatively related to students' Self-perceived communication competence.H3a: Self-perceived communication competence is positively related to students' willingness to communicate in L2.

Gardner's Approach to Motivation
Because WTC was originally presented as a personal trait affection (MacIntyre et al., 1998), MacIntyre et al.'s (1998) heuristic model is enhanced using Gardner's (1985) socio-educational model.Motivation influences WTC either directly, as shown in studies conducted in Turkey (Cetinkaya, 2005), or indirectly, through the two components of communication self-confidence, as shown in studies conducted in Japan (Yashima 2002, Yashima, Zenuk-Nishide, & Shimizu, 2004) and China (Peng & Woodrow, 2010).Many researchers have discussed the implementation of the socio-educational model in the L2 context (Clément et al., 2003;Hashimoto, 2002;MacIntyre & Doucette, 2010).Gardner and Lambert (1959) developed an approach to motivation that is valuable in L2 learning and acquisition.They created the distinction between integrative and instrumental motivation.Integrative motivation is an individual's willingness to join in the target language community and the inclination toward the target language, while instrumental motivation refers to the pragmatic reasons for learning a language, such as to study abroad or find a better job.However, Malaysian university students proved to be both integratively and instrumentally motivated to learn the English language (Vijchulata & Lee, 1985).Gardner (1985) proposed that the language learner's initial motivation originated from the general attitudes of the learner.In the model, the two attitudinal constructs, integrativeness and attitude toward the learning situation, are proven to contribute to one's level of motivation, which constitutes integrative motivation.In turn, the motivation level influences linguistic outcome, such as proficiency to talk in an L2.Integrativeness involves emotional determination in another cultural group.Students with a high level of integrativeness and stronger L2 learning motivation will be more willing to interact with the L2 language group than their peers (Yashima, 2002).According to Gardner, "Motivation ... refers to the combination of effort plus desire to achieve the goal of learning the language plus favourable attitudes toward learning the language."In the social-educational model, these three elements-effort, desire, and favourable attitude-are associated with distinguishing among individuals in terms of their respective levels of motivation.In order to achieve true motivation, learners should have a combination of effort, desire, and positive expression in order to learn other languages (Gardner, 1985).Gardner's (1985) model has been widely applied to L2 WTC research, with inconsistent results.According to previous studies, such as Hashimoto (2002) and Yashima (2002), a significant correlation exists between L2 WTC and motivation.Therefore, the following hypotheses were derived: H4a: Motivation is negatively related to students' communication apprehension.
H4b: Motivation is positively related to students' Self-perceived communication competence.H4c: Motivation is positively related to students' willingness to communicate in L2.

Study Locale and Participants
The study was conducted at the Universiti Sains Malaysia (USM).Similar to any other university in Malaysia, USM requires students to pass the Malaysian University English Test (MUET) in order to qualify for admission.This requirement emphasizes the importance of using and learning the English language in the university.The participants included 377 undergraduate English majors' students from four academic levels; they are pre-service English teachers who will guide students and make them aware of the importance of WTC as a key factor in learning, as well as to encourage them to communicate using the English language in the classroom.A stratified sampling technique was used to collect data from them.However, from among the 377 students, only 313 participants returned complete questionnaires (male, N= 49; female, N= 264) (year 1, N= 53; year 2, N= 86; year 3, N= 85; year 4, N= 89).

Procedure
The current study involved a two-stage process.Stage I validated the instrumentation using confirmatory factor analyses (CFAs), Stage II testing the hypothesized structural relationships among the constructs.The hypothesized relationships were tested using the Analysis of Moment Structures (AMOS) Version 17 (Arbuckle, 2006) statistics program and were estimated using the maximum likelihood estimation.

Instruments
A questionnaire written in English was designed to comprise two parts.The first part contained self-reporting demographic information, such as gender, age, race, and academic year.The second part contained questions for measuring indicator variables, such as students' L2 learning motivation, Language learning communication strategies, and communication tendencies.The following are the measuring indicator variables of the questionnaire:

Oral Communication Strategy Inventory
This inventory was used to assess the traits of students' use of OCS.Adapted from Nakatani (2006Nakatani ( , 2010)), OCSI examines strategies for coping with speaking problems related to strategic behaviour in L2 communicative tasks.It comprises 32 items, including eight factors: social-affective, fluency-oriented, negotiation-for-meaning, accuracy-oriented, message-reduction and message-alteration, nonverbal message-for-speaking, message-abandonment, and attempt-to-think English strategies.These items were developed as Likert-type psychometric scale questionnaires, ranging from one (never, or almost never, true) to five (always, or almost always, true).This instrument showed highly acceptable internal consistency (Cronbach's alpha = 0.88).

Willingness to Communicate in English
Adapted from Weaver (2005), this variable assessed students' WTC in speaking.A total of fifteen items were rated on a five-point Likert scale (Cronbach's alpha = 0.92).

Perceived Communication Competence in English
Fifteen items were rated using the five-point Likert scale.The same items in WTC were used as templates for measuring competence, as this method is commonly applicable in L2 WTC research (MacIntyre & Charos, 1996;Yashima, 2002).The effectiveness of this scale has been proved in measuring perceived communication competence in the study by Peng and Woodrow (2010) (Cronbach's alpha = 0.88).

Desire to Learn English
Six items from Gardner and Lambert (1972) were used in a seven-point scale format, instead of the original format of multiple-choice, to assess students' desire to learn English.The scale ranges from one (strongly disagree) to seven (strongly agree) (Cronbach's alpha = 0.71).

Motivational Intensity
This measure of motivation comprises six items pertaining to motivational intensity derived from Gardner and Lambert (1972).The original format was formulated in a seven-point scale, ranging from one (strongly disagree) to seven (strongly agree) (Cronbach's alpha = 0.72).

Attitudes towards Learning English
This variable is defined with five items adapted from Gardner (1985) and formulated in a seven-point scale ranging from one (strongly disagree) to seven (strongly agree) (Cronbach's alpha = 0.77).

Test of the Measurement Model
CFA conducted in order to determine the quality of the measurement model.Absolute fit indices are used to measure how well the proposed model duplicates the observed data, considering the X 2 statistic, root mean square error of approximation (RMSEA), comparative fit index (CFI), and Tucker-Lewis index (TLI).A good model fit has some rule-of-thumb criteria for goodness-of-fit indices.The chi-square normalized by degrees of freedom (CMIN/df) should not exceed 2 (Hu & Bentler, 1998).The TLI and CFI should both be above 0.95.The RMSEA should be less than 0.05 to be considered a good model fit (Hu & Bentler, 1998).In order to improve the model and its fit indices, standardized residuals and modification indices should be examined (Hair, Anderson, Tatham, & Black, 1998).In the current study, the modification indices suggest modifying the model by deleting the two-factor structures in language learning communication strategies, namely, fluency-oriented strategies and accuracy-oriented strategies, and one-factor structure from motivation, namely, attitude toward learning English.However, after the modifications, the measurement model showed that the factor loadings of each variable on the constructs and all the parameter estimates were significant at the p = .05level.Moreover, there was a good model fit for the measurement model (X 2 = 53.85,X 2 /df = 1.42,TLI = 0.98, CFI = 0.987, RMSEA = 0.037).The adequacy of the measurement model reveals that the variables are reliable indicators of the hypothesized constructs, consequently allowing the test for the structural relationships to commence.Figure 1 present the resulting path coefficients of the research model; all the paths were significant, except that from motivation to students' WTC in English.A new path from the use of Language learning communication strategies and students' WTC in English was determined.model adequacy indicated that the model is statistically fit to the data: RMSEA value was 0.019, as recommended by Byrne (2001), comparative fit index (CFI) was 0.996, Tucker-Lewis index (TLI) was 0.995, and the overall good fit index (GFI) was 0.973.Moreover, the chisquare statistics of X 2 = 51.32,df = 46, p = 0.273 and relative Chi-Square (CMIN/df) = 1.12, which was smaller than 3, as recommended by Hu & Bentler (1998).The combination of these results suggests that the structural model exhibits a good level of model fit.However, in the current study, Cohen's f 2 was used to estimate the effect size (ES) of the multiple-squared correlations (R 2 ) for each variable, as f 2 = R 2 /1 -R 2 .According to Cohen's (1992) interpretations, f 2 = 0.02 is a small effect, f 2 = 0.15 is a medium effect, and f 2 = 0.35 is a large effect (p. 157).Note.F1 = communication competence; F2 = motivation; F3 = communication apprehension; F4 = WTC in English; F5 = language learning communication strategies; PC1 = communication competence in form focus activities; PC2 = communication competence in meaning focus activities.

Test of the Structural Model
Table 1 presents the standardized parameter estimates for each variable in the structural model.As observed from the table, all the variables have a large effect size of the variance, except CA, which has a medium effect size of the variance.

Discussion
As observed, the final model conformed well to the data.The regression coefficient from the use of Language learning communication strategies and WTC was significant (R 2 = 0.053; f 2 = 0.056, small ES).Moreover, it was found that through motivation and self-perceived communication competence (i.e., 0.71 × 0.38 × 0.22 + 0.20 × 0.30; R 2 = 0.014; f 2 = 0.0142, small ES), there was a significant indirect relationship between Language learning communication strategies and WTC.The important role this variable has in helping students initiate a conversation is noteworthy.This finding is consistent with that of Gallagher-Brett (2001), in which the use of language learning communication strategies is an effective tool for helping students become involved in conversation, particularly in a classroom setting.Consequently, students' WTC in an L2 will increase.The anticipated path from the use of Language learning communication strategies to motivation was significant with a high effect size (R 2 = 0.504; f 2 = 1.016, large ES), thereby justifying the premise that the more students apply Language learning communication strategies, the more they are motivated to learn English.This supports the correlation theory of Bandura (1986) that if an individual has the ability to speak an L2 effectively, he/she would, consequently, be more motivated to learn and practice speaking the language.Moreover, Bandura (1997) and Margolis (2001) found that compensation strategies are important skills in improving and increasing one's motivation, self-confidence, self-efficacy, and communication ability in the target language.
Language learning communication strategies directly influenced students' self-perceived communication competence (R 2 = 0.04; f 2 = 0.042, small ES) and indirectly, through motivation (i.e., 0.71 × 0.35; R 2 = 0.062; f 2 = 0.066, small ES).An emphasis is placed on the important role of the teachers in helping students learn these strategies; the teachers also enhance the students' communication competence by assisting them in meaningful interaction through language tasks.The results support the findings of Van den Branden (2000) and Morell (2004).In the current study, it may be inferred from the data that the major reasons why students employ certain language communication strategies are highlighted in their 1) motivation to learn the language, 2) their inclination to increase their communication competence in order to avoid difficulties in oral communication, and 3) their WTC in an L2.
Among the four remaining composite indicators, negotiation of meaning strategies (f 2 = 1.494, large ES) proved to be higher in Language learning communication strategies, with a large effect size.This finding reveals that among Malaysian university students, as well as in the classroom, negotiation of meaning is a significant factor in improving students' WTC in English.Through negotiation, they can overcome the limits of their language capacity, thereby increasing their active involvement in conversation and enriching their L2 acquisition.Previous research, such as Pica (1994), Bitchener (2004), andNakatani (2010), supports the role of negotiation of meaning through face-to-face interaction and its effect on students' communication.Moreover, a recent study in Malaysia conducted by Al-Mahrooqi and Tuzlukova (2011), proved the significant role of negotiation of meaning strategies in developing students' strategic competence, achieving mutual comprehension, decreasing students' communication anxiety, and providing them with a positive learning environment.This strategy was followed by message-abandonment strategies (f 2 = 0.866, large ES), message-reduction strategies (f 2 = 0.698, large ES), and socio-affective strategies (f 2 = 0.490, large ES); students utilized message-abandonment strategies and message-reduction strategies due to the lack of linguistic knowledge in an L2.However, social-affective strategies proved to have a significant impact on learning (Habte-Gabr, 2006), particularly in oral proficiency (Nakatani, 2010).In the current study, it was found that the more the students relax and enjoy their conversation, the better they are in their communication competence, thereby increasing their WTC in an L2.
Motivation had a direct influence on the two components of communication confidence and influenced WTC indirectly through the two variables, that is, self-perceived communication competence and Communication apprehension (i.e., 0.35 × 0.30 + 0.38 × 0.22; R 2 = 0.0356; f 2 = 0.037, small ES); this result was consistent with those of Yashima (2002), Yashima, Zenuk-Nishide, & Shimizu ( 2004) and Peng & Woodrow (2010).Moreover, a significant path from Communication apprehension to self-perceived communication competence was revealed (R 2 = 0.194; f 2 = 0.241, medium ES), which was consistent with the findings of Hashimoto (2002).In the Malaysian classroom context, students motivated to learn English are not required to communicate in English.
For most Malaysian students, English is studied merely to pass an examination; thus, it is far from achieving communication goals.However, the more motivation students could gain in classrooms, the less apprehension they would have and the more competent they would become in communicating in an L2.This finding is supported by Yashima (2002), who found that the more time students have for studying L2, the more communication confidence they will gain.
Meanwhile, self-perceived communication competence (R 2 = 0.09; f 2 = 0.099, small ES) and Communication apprehension (R 2 = 0.048; f 2 = 0.051, small ES) were found to exert a direct influence on WTC.These findings are in line with the L2 WTC theory (MacIntyre et al., 1998) and some empirical studies, such as Hashimoto (2002).In the current study, as well as in a classroom, students with high L2 competence and who are less apprehensive tend to be more willing to communicate in an L2.
However, the final model accounted for 38% of the variance of WTC in English (Table 1; R 2 converted to percent), thereby revealing that the joint effect of the use of Language learning communication strategies, self-perceived communication competence, Communication apprehension, and motivation predicted the variation of WTC in Malaysian classrooms to a certain extent.Further discussion, such as learners' personalities and learners' self-esteem in using an L2, should be conducted centred around the fact that there are other significant variables, apart from Language learning communication strategies that play a vital role in L2 learning in the Malaysian classroom and have a compelling influence on students' WTC in English.The high proportion of the variance of self-perceived communication competence (62%; Table 1) suggests that students' perception of their ability to communicate tends to be influenced by a combination of their awareness in using of Language learning communication strategies, their Communication apprehension, and motivation.Similarly, the high proportion of the variance of motivation (51%; Table 1) suggests that learners' motivation to learn English tends to be influenced by their awareness of using Language learning communication strategies, thus indicating the vital role of this variable in increasing students' motivation to learn an L2.In contrast, the low proportion of the variance of Communication apprehension (15%; Table 1) suggests that students' motivation to learn an L2 and their awareness in using Language learning communication strategies were the least predictable.Students' tendency to have a sense of apprehension, whether in a group discussion, meeting, or even through interpersonal conversation in the classroom may be attributed to other factors such as classroom environment, learners' belief about English learning, and communication.

Conclusion
The current study is heuristic in that it is the first to examine the influence of using Language learning communication strategies on WTC in an ESL classroom.In order to attain the level of English proficiency required, there is a need to encourage Malaysian ESL university students to achieve higher WTC in English.
Motivation and L2 WTC are coherent variables embodied in SLA (Yashima 2002).Language learning communication strategies is an important variable in the current study, as no previous study has ever examined its relationship with students' WTC in a second/foreign language utilizing structural equation modelling.The results of the current empirical study support MacIntyre et al.'s (1998) heuristic WTC model in determining the variables influencing WTC in an L2 communication setting.Moreover, the results reinforce the significance of the WTC heuristic model by enhancing Gardner's (1985) socio-educational model.However, the final model shows that Language learning communication strategies has a significant direct impact on students' WTC in English in the classroom and is considered a very effective factor in influencing students' WTC in English.MacIntyre et al.'s (1998) heuristic model can strengthen future research on the ultimate purpose of language learning.For a more thorough investigation of L2 WTC in the Malaysian context, future research may address a similar topic and use other methods, such as classroom observations and interviews.In addition, by adopting different instruments to examine students' WTC in English, further research must seek to determine students' WTC, not only in the spoken mode, but also in other modes.Further research is recommended in order to determine the effect of other variables such as personality traits as well as a person's desire to communicate with a specific person and WTC in English.Owing to Malaysia's multicultural and multilingual perspective, future research must determine the influence of race and gender on their WTC in an L2.In summary, "By learning how to use communication strategies appropriately, learners will be more able to bridge the gap between pedagogic and non-pedagogic communication situations" (Faerch & Kasper, 1983: 56).
work on unwillingness to communicate.Through the extended works of MacIntyre and Charos (1996), MacIntyre et al. (1998) built a heuristic model of WTC in an L2.MacIntyre et al.'s heuristic model proposes WTC to be an interaction between cognitive affective variables and social factors.Researchers have conducted substantial works on WTC as a predictor in learning an L2 (MacIntyre, Baker, Clément, & Conrod, 2001; MacIntyre, Baker, Clément, & Donovan, 2003; MacIntyre & Doucette, 2010; Peng & Woodrow, 2010).Since the model was introduced in 1998, researchers and scholars have been working on testing its different forms.Focusing on the educational context of the classroom, MacIntyre et al. (

Figure 1 .
Figure 1.Structural model of WTC in English in the Malaysian ESL classroom Indicators: PC1 = communication competence in form focus activities, PC2 = communication competence in meaning focus activities, WTC= willingness to communicate

Table 1 .
Standardized parameter estimates for the structural model