Willingness to Communicate in Iranian EFL Learners: The Effect of Class Size

Willingness to communicate can be considered as one of the important factors in modern language pedagogy which put emphasis on meaningful communication. The present study investigated the effect of class size on the Iranian EFL students’ willingness to communicate among three different class sizes. The researcher collected the data through observation of three classes in terms of students’ turn of talk and talk time. Descriptive statistics including mean and standard deviation and one way ANOVA were run to analyze the data. The results of the study indicated that class size had a substantial effect on the students’ willingness to communicate. Students were found to be more willing to communicate in small classes where they had more opportunity to practice oral skills and communicate. The current study had some pedagogical implications for both second language teaching and learning at the end.


Introduction
One of the main purposes in learning second or foreign languages for many second language learners is the use of target language for communication.Language use also can indicate successful second language acquisition.Modern language teaching has emphasized the significance of cultivating communicative competence in second language learners (Canale and Swain, 1980).By the emergence of the communicative language teaching (CLT), "authenticity, real-world simulation and meaningful tasks" were the main characteristics of classroom conversation (Brown, 2001).Therefore for CLT to be effective there should be interaction between students and teacher as well as learners and their peers.To this end, students may need motivation and willingness to initiate interaction.Some researchers have argued that a fundamental goal of second language instruction should be the creation of willingness to communicate (WTC) in the language learning process (e.g.McIntyre et al., 2003).Willingness to communicate is a potentially fundamental concept for effective interaction and language production.According to MacIntyre et al. (1998), the proper goal of the L2 pedagogy should be to instill WTC in learners and language learning process to produce learners who are eager and willing to seek out opportunities and communicate.
One of the crucial factors that can influence students' willingness to communicate is class size.In the following sections the current study primary focused on this issue.

WTC
The WTC concept was first developed in the first language communication by McCroskey and Richmond (1990).McCroskey and Richmond (1990) treated WTC as a personality trait and defined it as the individual's intention to initiate communication and participate in conversation in various situations.McCroskey et al (1985) investigated communication apprehension among Japanese students in speaking both their native language (Japanese) and L2 language (English).The findings of the study showed a high level of communication apprehension in both English and Japanese languages among the students.
Based on the findings of McCroskey et al. (1985) and by combining native language communication and L2 motivation, MacIntyre et al (1998) developed a comprehensive model of WTC in SLA research.The proposed model of WTC integrated social-psychological, linguistic, educational and communicative dimensions of language to predict, describe and explain the students' WTC in the L2.MacIntyre et al (1998) defined WTC as "the probability of engaging in communication when free to do so."In fact they considered WTC as a state of readiness to initiate communication in L2 at a particular time and in specific situations.Thus, it can be concluded that the MacIntyre et al's WTC model sees willingness to communicate in target language as a situational factor which includes both "enduring" and "temporary" effects.The pyramid or heuristic model of variables influencing WTC in L2 communication tries to illustrate the twelve variables that influence WTC (MacIntyre et al, 1998).
Based on MacIntyre et al.'s model of WTC, Yashima (2002) investigated the effects of individual differences on Japanese students' WTC.The results showed that an attitude toward international community, motivation for learning English and English communication confidence affected WTC in English in Japanese context.Matsuka (2004) conducted a study on 180 nursing Japanese college students to investigate the relations among individual differences, WTC and English proficiency.The results of the study indicated that introversion and communication apprehension were direct negative predictor and that self-efficacy, including perceived competence and motivational intensity, was the direct positive predictor of WTC in English.He reported that compulsivity and internal postures were indirect postures.MacIntyre et al ( 2011) explored ambivalence about communication among adolescent French immersion students, and found interrelations among linguistic developments in L2, self-development and non-linguistic issues.They reported that perceived competence and error correction were the most important issues concerning WTC.Hashimoto (2002) partially replicated a study by MacIntyre and Charos (1996) to examine affective variables as a predictor of L2 use in Japanese ESL learners' classroom.The results of the study indicated that motivation and WTC affect communication frequency in classroom.Furthermore, the finding showed that the variables underlying WTC, including perceived competence and L2 anxiety, were the causes of WTC.Perceived competence exerted a strong influence on WTC in L2 communication and led to more L2 use.L2 anxiety was found to be negative predictor of perceived competence and WTC, consequently.
Among few studies that have been conducted on WTC in Iran Birjandi and Amiri (2011) explore WTC in first language acquisition.The result of this study revealed that the results of the study revealed that WTC could be considered both a personality factor and a social characteristic which affects L1 development among kids.Alemi, Daftarifard and Pashmfroosh (2011) investigated The Impact of Language Anxiety and Language Proficiency on WTC and found that Iranian university students' WTC is directly related to their language proficiency but surprisingly higher proficient learners showed to be less communicative than lower proficient ones outside the classroom.

Class Size
Studies have shown that students placed in smaller classes benefit more academically than those placed in larger classes.Glass and Smith (1980) found that student morale, achievement, attitude, and satisfaction were higher in "smaller" classes.However, Hanushek (1988), in a review of literature on this topic, concluded that differences are always scientifically small.Additionally, Evertson and Folger (1989) argue that students in small classes have more opportunity to talk to the teacher about problems.Kumar (1992) found that it is the nature of teaching-learning activities and the teacher's role and attitude which influences the nature of learner participation and the patterns of interaction rather than class size per se.Interestingly, Hess (2001) takes the view that more communication and interaction can occur in a large classroom where students benefit from peer-teaching through group tasks.It also has been suggested that Students in smaller classes spend more time on tasks, misbehave less frequently, and have higher test score gains (Achilles et al, 2002).The interaction between the teacher and students improves when classes are smaller and student participation increases, which results in a more positive attitude toward learning (McCluskey, 2002).The underlying factor which affects the success of interactional negotiation in the classroom is the extent to which students actually talk.Wells & Chang-Well, (1992) claim it is true that small groups stimulate more student talk-time, student utterances.
The challenge of creating this ideal environment is likely to be easier in smaller classes (Glass & Smith, 1980) than larger ones.It seems that students can have better interaction when they are in smaller classes.
Piles of studies have been conducted to investigate different variables affecting second language learners' WTC, particularly in Western countries.As far as the researcher knows, there are not many studies which takes the effect of class size on WTC into consideration in Iranian EFL learners.Therefore, this study aims at investigating the effect of different class size on Iranian learners' WTC who are learning English in an EFL context.
To fill the gap of literature the present study proposed the following research question: Does WTC differ across different sized classes in Iranian EFL learners?

Participants
The participants in this study were 30 male and female adult Iranian intermediate EFL learners.All of them were attending the same institute in Kermanshah, Iran.In the current study, the participants were selected from three separate classes: one class containing five students which was regarded as small class in this institute and we call it class A, the other class consisted of ten students and we named it class B, and in the large class there were fifteen students that is class C in this study.

Observation
The researcher estimated the turn of talk and talk time for each student during the observations.Observations took six weeks to be collected each one lasted for one hour and half.

Data Collection Procedure
The data were collected in fall 2012.Three classes with varying sizes were used for data collection procedure.Class A included 5 students, class B consisted of 10 students, and class C contained the largest number of students which was 15.
The researcher observed all three classes to record the amount of time each student talk and participant in activities.Furthermore, the number of talk turn during communication and conversation for each student was calculated.

Data Analysis
The current study employed descriptive statistics, namely mean and standard deviation with regard to each student's turn of talk and talk time.Moreover, one way ANOVA was run to investigate the effect of class size (three classes) on students' willingness to communicate.

Results and Discussion
In this section, the researcher indicated the results of the research question.

Does WTC differ across different size class in Iranian EFL learners?
Class observation revealed that as class size increased, WTC decreased; this held true with regard to both indicators, namely turn of talk and talk time.Thus, the greater the number of students in class, the less willing the students would be.As shown in table 2, the mean of talk-time in class A is 15.37 and the standard deviation is 1.33.The mean of turn of talk in class A is 146 and the standard deviation is 12.The mean of talk-time in class B is 8.06 and the standard deviation is 0.64.The mean of turn of talk in class B is 81 and the standard deviation is 6.The mean of talk-time in class C is 5.37 and the standard deviation is 1.47.The mean of turn of talk in class C is 51 and the standard deviation is 4. As table 3 shows mean of talk-time in 3 different classes which in class A in comparison to 2 other classes is high (15.37).This is shown in figure 1.As table 5 shows, the results of ANOVA indicated that there was statistically significant difference at the .05significance level in WTC mean score for the three classes: (F (2,27)=15.12).The findings of current study shed light on the fact that class size has a crucial effect on an individual's WTC in a second language classroom.Furthermore, the results support to the Wells and Chang-Well's (1992) statement that students in smaller classes are more motivated to participate in communicative activities and conversation.The results of current study were in line with those of Aubery (2010) in which the effect of class size on students' WTC was confirmed.In fact, in classes with great number of students, learners experienced higher level of anxiety during interaction and initiating communication.Thus, they perceived the situation as anxiety provoking and avoid communicating.On the other hand, students in smaller classes had enough opportunity to practice and build their self-confidence which in turns lead to higher perception of the capability in communicating in second language.Coleman (1989) argued that one way to overcome the negative influences of larger classes on students' WTC is that to employ more communicative tasks and activities in classrooms that support participation and interactions.
The result of current study revealed that students in class A had more time to express themselves and personalize their answers.Thus, these students overcome their anxiety level and feel free to initiate conversation.On the other hand, due to fewer number of students in class A, the students had more chance to take turn during meaningful communication.
Students in class B found to be less willing to communicate in compare to the students in class A because of fewer opportunities they had to take turn and to be engaged in communicative activities.
Finally, the results indicated that students in class C had the least degree of willingness to communicate due to the greater number of students.In fact, these students had the least amount of time to express themselves and to take turn during communication.

Conclusion
Results from this study present how strongly class size can affect the participation of learners in class discussions and WTC.Teachers in larger classes should be aware that to employ more communicative strategies in order to encourage the students to participate and to be involved in class interaction.Further study can probe students' willingness to communicate not only in oral mode but also in other modes.Different instruments can be employed to examine learners' willingness to communicate in writing and their comprehension of written and spoken English.

Figure 1 .
Figure 1.Comparison of talk-time mean for 3 classes of different size

Figure 2 .
Figure 2. Comparison of turns of talk mean for 3 different class sizes

Future
studies can be carried out to probe both what affective variables can be added to MacIntyre et al.'s model of WTC in L2 (1998) and what affective variables should be excluded from the model in throwing light on the complicated nature of WTC in L2.

Table 1 .
Classification of Classes

Table 2 .
Descriptive statistics of WTC-Talk-time and turns of talk

Table 5 .
One-way ANOVA willingness to communicate by length of studying

,
Post-hoc comparison using the Tukey test indicated that the mean score for class A was significantly different from class B and class C; however, Class B and class C did not differ significantly from each other.In fact, it can be concluded that the more the number of students, the less willing the learners would be.

Table 6 .
Post-Hoc Scheffe's Tests WTC by class sizeSince class A included lower number of students, they found to be more willing to initiate conversation in L2 WTC.In fact, students in Class A created more opportunities for themselves to be cooperative in the class due to the lower number in compare to class B and C. The results suggested that Students in class A (M=15.37, SD=1.33) had more time to talk in L2, followed by students in class B (M=8.06, SD=.64) and students in class C (M=5.37, SD=1.47).On the other hand, learners in class A found to have more turning of talk (M=146, SD=12) than the other students in class B (M=81, SD=6) and class (M=51, SD=4).