Willingness to Communicate and Second Language Proﬁciency: A Correlational Study

: This study attempts to answer one straightforward question: “what is the relationship between students’ proﬁciency level and their willingness to communicate?”, i


Introduction
Developing English (or any L2) learners' willingness to communicate (WTC) is important because it is a precursor to the development of communicative skills and has a major influence on language acquisition in general.In EFL (English as foreign language) settings, a major (perceived) impediment is the lack of access to the target language and opportunities to use it to communicate because language is learned through meaningful communication [1].Communication is important for L2 learners; through the process of interacting, they negotiate meaning with their interlocutors, and this process facilitates acquisition because the learners receive feedback from their use of the L2 [2].Through using the language, the learners have the opportunity to use communication strategies, which are regarded as important for producing language output [3].In Thailand, as in many other EFL settings, language learning outcomes are poor, in part perhaps because students do not develop the ability to engage actively in finding opportunities for interacting in the target language.Most students are reliant on the language input provided by the teacher and the opportunities for interaction with other learners in class.Whether students take up these opportunities is at least in part dependent on their willingness to participate and actively communicate in class.In Thailand, students are reticent, and this may in part explain why Thai students' proficiency ranks a disappointing 64 out of 88 non-native English-speaking countries [4].There are many reports of Thai learners being particularly quiet and passive in language learning compared to other learners (e.g., [5,6]).In addition, many Thai learners have few opportunities to develop and express their autonomy in the foreign language context of an educational system in which English is a compulsory subject for graduation.This severely impacts their motivation, of which their WTC is one expression.In addition, too much emphasis is placed on grammar, which may affect learners' ability to improve their speaking skills [7].Perhaps the Thai education system plays a role in this.Although there has been an attempt to adopt a communicative approach in teaching, there are still many cases where traditional grammar-translation is used [8].Even in communicative tasks such as role playing and oral presentations, speaking opportunities are often minimal, with activities being rigidly planned and controlled by the teacher.Another issue which may affect the English proficiency of Thai students is that assessments focus heavily on grammatical knowledge, reading, and writing at the expense of listening and-in particular-speaking.Therefore, teachers place more emphasis on grammar teaching in order to help students pass their exams.As a result, many Thai learners receive little opportunity to interact in English and do not develop confidence and fluency.It is possible that there is a threshold level of proficiency, below which students lack confidence to engage in L2 communication, and that most Thai learners simply do not achieve this.In order to investigate this hypothesis, we conducted a correlational study to establish the relationship between L2 proficiency and WTC, in order to better understand the specific needs of different groups of students and thus to better align educational practices in Thailand and other EFL settings.This study involved a large number of participants used in quantitative studies having been conducted on WTC as they were taking a compulsory English course.Knowing students' WTC can help us identify ways of developing it, especially in a context that is not conducive to learner autonomy.Because language use and language learning happen at the same time, it is necessary to identify factors which constrain and encourage learners' opportunities to communicate and acquire the target language through communication and interaction.

Literature Review
The term willingness to communicate (WTC) originated from research on unwillingness to communicate in the first language, where WTC was regarded as a trait-like construct, which is stable in an individual and across communication situations and types of interlocutors (McCroskey and Baer, 1985 cited in MacIntyre [9]).WTC in second language learning (L2) is defined as "readiness to enter into discourse at a particular time with a specific person to person, using a L2" [10] (p.547).The ability to express oneself in the target language is widely considered a primary goal in L2 learning.WTC is a prerequisite for successful communication and thus ongoing development, because it involves an individual's psychological preparedness to use the L2 when there is an opportunity, as well as the learner's ability to act on this in the moment [9].WTC is considered to be a predictor of the frequency with which learners communicate in the target language and can thus facilitate the process of learning [10].WTC is affected by a wide range of factors, both individual and situational, and is considered a dynamic construct because it "can vary according to interlocutor(s), topic, and conversational context, among other potential situational variables" [11] (p. 291).The L2 WTC heuristic pyramid model proposed by MacIntyre et al. [10] integrates psychological, linguistic, educational, and communicative factors, which were previously researched independently [9].
The L2 WTC heuristic pyramid model proposes that the L2 communication process is made up of six layers, which include twelve variables, covering linguistic, communicative, and psychological aspects.Layer VI, which is the lowest one, includes intergroup climate and personality.Level V includes a typical affective and cognitive context which shows the tension between an individual's desire to approach people from the target language and fear to do so.Level IV includes highly specific motives and self-related cognition.The motives include intergroup motivation, which comes from the individual being a member of a certain social group, and interpersonal motivation, which comes from the social role which (s)he plays in the group.At this level, roles, motives, and L2 self-confidence interact.These three layers involve enduring influences because they are more stable and predictable in most situations compared with the top three layers.Layer III is situationally oriented because an individual's desire to communicate depends on the person and his or her state of self-confidence.The person represents a different social group.Layer II shows the personal psychological preparedness for communication; an individual may be willing to communicate or choose to remain silent.Layer I represents communication behaviour, which is the use of the L2.These variables are regarded as transient, i.e., dependent on the time and place in which they occur.According to the model, learners feel more ready to communicate when they move up the pyramid and they can make use of the L2.This model suggests that the choice and decision made by an individual at a particular moment contribute to his/her initiation to communicate.
The review of previous studies reveals how WTC has been correlated with various variables.The variables in the heuristic pyramid model have been studied in a variety of educational contexts (e.g., [12][13][14][15][16]), in particular communication anxiety, self-perceived communicative competence, motivation, and personality [13,15,[17][18][19].Anxiety and selfperceived communicative competence were the first two variables to have been studied and were found to be good indicators of L2 WTC [15,[17][18][19][20].With regard to communication anxiety when learning or using L2, research results show that it contributes to low levels of WTC, i.e., learners who have high anxiety about language communication tend to keep silent [10].However, a study by Alemi et al. [21] revealed that the interaction between WTC and anxiety was not that significant.The studies on WTC and motivation show that highly motivated learners have higher levels of WTC both in class and outside class [10,22].Even though there is not always a direct correlation between the two variables such as in the study by Yashima [15], the findings show that motivation influences self-confidence in L2 communication, and this leads to an increase in L2 WTC.Various motivational constructs have been investigated with WTC, such as integrative motivation and WTC [13,22].More recently, Ghanizadeh et al. [23] conducted a study in Iran to investigate the relationship between WTC and four components of the Motivational Self System proposed by Dörnyei, i.e., criterion measure, ideal L2 self, attitudes to L2 culture, and community and family influence.The findings reveal a significant and positive correlation between all motivational factors except family influence.
The relationship between personality and WTC has been investigated by focusing on learners who are extroverted because it is a personality trait that is linked with communication [18].MacIntyre et al. [10] found that introversion/extroversion and emotional stability are linked with WTC through communication apprehension and perceived language competence.The "big five" personality factors, which include openness to experience, conscientiousness, extraversion, agreeableness, and neuroticism, have been investigated with WTC, such as in the study by Oz [24], which found that among Turkish students included in the study, extraversion, agreeableness, and openness to experience could predict WTC in English.Adelifar et al. [25] conducted a study in Iran with participants who have the same proficiency level, using the NEO Five Factor Model of personality, which includes (a) neuroticism, or the tendency to experience negative experiences; (b) extroversion; (c) openness to experience; (d) agreeableness; and (e) consciousness or one's level of self-control in planning and organisation.The results revealed that extroversion had no relationship with WTC.
Other studies have explored variables that relate to L2 WTC, such as the classroom environment, including tasks, group size, or teacher; learning contexts such as immersion; attitudes towards the international community; cultural factors; and gender and age [13,16,19,[26][27][28][29][30].Some of the variables are similar to the variables in this study.With regard to the Asian context, Wen and Clément [14] critically assessed the heuristic pyramid model when it was applied to the Chinese setting, where English is mainly used as a foreign language; communication happens only in class between the teacher and the students.They suggested that cultural influence, i.e., Confucianism and the teaching of Confucian classics, might affect how learners perceive ways of learning.Teachers play an authoritative role, and students submit to authority.In addition, there are other cultural aspects which may hinder the students' willingness to communicate, such as a face-protected orientation, which affects their openness to judgement of their L2 performance, and a group-oriented culture that may make Chinese learners less willing to communicate with foreigners who are considered outsiders.To Wen and Clément, because the heuristic pyramid model originated in the West, the variables that are presented in the model do not fit with the Chinese setting, especially the relationship between desire to communicate, which is in Layer III, and WTC in Layer II of the model.They define desire as the preference of the learners or their choice to communicate, whereas willingness shows learners' readiness to communicate.The Chinese students who are quiet and do not seem to be willing to communicate do not necessarily lack desire to communicate.Additionally, those who have desire to communicate may not be willing to do so if they are affectively unprepared because they may be anxious as a result of situational variables in the environment.The variables in their proposed framework include social context, personality factors, motivational orientation, and affective factors.Those variables create a positive communication environment which encourages learners to engage in communication and reduce anxiety.However, this proposed theoretical framework requires further empirical confirmation.
The studies on the relationship between proficiency and WTC include how the learners use various language skills to perform language tasks or correlating the scores of proficiency tests and WTC.For example, Alemi et al. [21] investigated the WTC of 49 Iranian university students and how this interacted with their anxiety and proficiency by correlating TOEFL scores with WTC questionnaire results [20] and language anxiety questionnaire responses using the questionnaire constructed by Horwitz, Horwitz, and Hope (2001).Alemi et al. [21] focused on the WTC of four language skills areas, and they separated WTC into WTC in the classroom (when the participants were assigned to do communicative tasks) and WTC outside the classroom, orientations for language learning, and social support.Because WTC is regarded as both a state and a trait construct, the researchers wanted to see whether some students might be able to communicate in class but might not be willing to communicate outside the classroom.The data revealed a relationship between WTC and proficiency in different situations, i.e., highly proficient learners had a higher willingness to communicate in the classroom than low proficient learners.However, low proficient students had higher WTC outside the classroom than highly proficient students.The researchers thought that this might stem from the fact that low proficient students do not like to be evaluated, or highly proficient students might receive more support from the teachers in class, which makes them more communicative in class.However, the authors may have somewhat overstated their claims as the data were only based on students' scores on the reading and structure sections of the TOEFL test, whereas WTC involves all four language skills.
Another study on the relationship between WTC and proficiency was conducted by Rostami et al. [31].The participants were 60 female students between 15 and 22 years of age who were studying English in schools and universities.The study was conducted over five weeks with students from six classrooms.The researchers used TOEFL scores to represent the students' proficiency by having the participants take the grammar, vocabulary, and reading comprehension sections.They also included classroom observations to determine how the participants were willing to communicate in a real classroom context.The observation was a checklist of behaviours which showed WTC.The data revealed that there was a relationship between proficiency level and WTC, but level of education or age did not correlate with WTC.The observation also revealed that advanced learners were more willing to communicate than intermediate students.
With reference to the Thai educational context, a study by Tan and Phairot [32] on the relationship between WTC and proficiency is relevant.The participants were 375 twelfthgrade Thai EFL students from two Southern Thai government secondary schools.The researchers tried to address the weakness of the previous research on proficiency which has tended to use self-report to represent the participants' perceived communication competence or proficiency.Instead, they used the scores of the standardised English examination in Thailand to report the participants' proficiency and separated the participants into high, moderate, and low proficiency.The researchers adapted the WTC questionnaire from Pattapong [33] cited in Tan and Phairot [32], who developed the questionnaire from Weaver [34].Weaver used the questionnaire to check the WTC of Japanese learners in writ-ing and speaking.However, Pattapong only focused on speaking and replaced Japanese with Thai [33].Tan and Phairot chose the items which were relevant to their situations and simplified and combined similar items.The questionnaire they used in their study thus included 26 items covering communicative situations that the participants generally encountered in class or outside class with teachers, classmates, or strangers, in the case of outside class communication.The results showed a higher level of WTC inside the classroom where the students were under the control of the teacher.There was a correlation between English proficiency and WTC, i.e., proficiency could predict WTC both inside and outside class.However, their results were based only on proficiency scores on general English knowledge, reading, and vocabulary; speaking and writing competence were not included.
The previous research on the relationship between WTC and proficiency has demonstrated a clear difference between WTC inside and outside class.Proficiency has been mostly represented by scores on standardised tests but has only covered receptive skills, such as reading and grammar, with skills more relevant to WTC, such as speaking, not included.Therefore, this study aims to investigate if there is a correlation between WTC and proficiency across all four skills and to do so with a large number of students in the context of a compulsory English course.

Materials and Methodology
This section describes the context of the study, participants, research question, and analysis.

Context
The study was conducted at King Mongkut's University of Technology Thonburi, a public university specializing in science and technology, in Bangkok, Thailand.Similar to other universities in Thailand, students take compulsory English courses in the first three semesters of their university study.In this university, the students are required to take an English Proficiency Test (EPT) before the first semester starts.Those who score less than 30% attend an intensive English preparation course for two weeks before taking LNG 101 (the first compulsory English course, which focuses on general English for communication).Those whose scores are between 30% and 40% take LNG 101 right away, and those who score 41% and higher take LNG 102, which is more advanced than LNG 101 and focuses on technical English.Both LNG 101 and LNG 102 teach the four skills in an integrated manner.Commercial course books are used in both courses, but a final group work task is added to have the students apply what they have learned throughout the course.Generally, there are 40-45 students in class, and the teachers try to use English in class, especially to ask questions, but they may switch to Thai if the students' proficiency is not high enough.Regardless, all tasks and oral presentations are completed in English.

Subjects
The study involved the administration of a WTC questionnaire to all first-year undergraduate students who studied in the regular programmes mainly using Thai as a medium of instruction.A total of 2237 students completed the questionnaire, 1836 of whom completed every section.In addition, 20% of the students (468 students) were selected by using stratified sampling, based on the percentage of the students from different departments, to obtain representation of the whole population in order to take a more comprehensive language test, called TETET (see below), which includes a speaking component.A total of 365 students registered for the test, and 299 students completed it, as well as the WTC questionnaire.

Measures
The instruments in this study include a WTC questionnaire, an English Placement Test (EPT), and the Test of English for Thai Engineers and Technologists (TETET).The WTC questionnaire was adapted from the questionnaires used in the WTC studies by Cao and Philp [35], Freiermuth and Jarrell [36], Léger and Storch [37], MacIntyre et al. [10], and MacIntyre et al. [20] because the questionnaires used in those studies were similar to the context of the study, but the questions about the communicative situations that the students would be likely to encounter in LNG 101 and LNG 102 classes were modified.The items are intended to measure WTC in a compulsory classroom context.The questionnaire consists of three sections: perceptions of willingness to communicate (Appendix A), communicative self-confidence (Appendix A), and frequency of English use (Appendix A).The Thai version of the questionnaire was distributed to the students a month after they started their English courses.
The English Placement Test (EPT) is a gap-fill and multiple-choice test given to all the students upon enrolling at the university.It consists of 100 items that test students' knowledge of listening, the structure of English, vocabulary, reading comprehension, functional English (used to test speaking ability), and written expressions (used to test writing ability).The content in the EPT reflects the language students encounter in the compulsory English courses.The test takes up to two hours, and the maximum score is 100.
TETET was developed at the university to measure students' English proficiency in the workplace and is based on TOEIC.The test simulates various English tasks which are common in the workplace.It is a computer-based test which consists of four skills: reading, writing, listening, and speaking.With regard to reading, the students are tested on survival reading skills, reading from the Internet, reading technical manuals, and email reading.Writing includes report writing, writing memorandums, and email writing.Listening includes listening to what is going on in a meeting, information conversations, and following instructions from an AVRS (automated voice response system).Speaking skills are tested through short questions in an interview context and leaving phone messages.A range of test types is used in TETET, including multiple-choice, drag-and-drop, writing short answers, etc. TETET is scored on a seven-band scale.As TETET includes tests of oral skills, it was decided to add this as an instrument for measuring L2 proficiency.

Research Questions
This study aims to answer the following research questions: What is the relationship between Thai first-year students' language proficiency and self-perceived WTC?
What is the relationship between Thai first-year students' language proficiency and self-perceived confidence?
What is the relationship between Thai first-year students' language proficiency and self-perceived language use?

Statistical Analysis
SPSS version 13.0 was used to calculate the degree of correlation between proficiency and WTC using the Pearson correlation coefficient.Descriptive statistics are used to show mean scores of proficiency test.

Results
The results from the Pearson Correlation test are statistically significant between WTC and performance on both the English Proficiency Test (EPT) and TETET (p < 0.01), both overall and for each of the three individual parts of the questionnaire (perception, self-confidence, and frequency of English use), as shown in Table 1.The correlation is somewhat low (0.221 between WTC and EPT and 0.300 between WTC and TETET).Thus, to summarise, for RQ 1, the results show the correlation between proficiency and self-perceived WTC to be 0.089 between WTC and EOT and 0.125 between WTC and TETET.For RQ 2, the correlation between proficiency and self-confidence is 0.267 between WTC and EPT and 0.350 between WTC and TETET.Finally, for RQ 3, the correlation between proficiency and self-perceived language use is 0.130 between WTC and EPT and 0.181 between WTC and TETET.Although the data from both proficiency tests accord with each other, we can see that the correlation between WTC and TETET, which focuses on the real use of English, is a little stronger than the EPT.
Because the items in the WTC questionnaire are related to speaking, we further analysed the scores of TETET.The results in Table 2 show that the lowest scores are for speaking.This might reflect why communicative self-confidence shows the strongest correlation between WTC and TETET, and the correlation is stronger than that between WTC and EPT.

Discussion
The findings of this study show statistical significance between WTC and students' English proficiency, as in previous studies (which were conducted in other contexts).However, the correlation between the two variables was only moderate to weak.MacIntyre et al. [20] agree that even though language proficiency is not one of the factors included in the heuristic pyramid model described above, the degree of L2 proficiency has a remarkable impact on the learners' WTC as shown in the study by Freiermuth and Jarrel [36].The study by Liu and Jackson [38] also reveals a link between proficiency and WTC in that low proficient learners do not take risks in speaking the L2 in class.In addition, Cao's study shows that learners' lack of linguistic competence affects their listening and speaking comprehension, in turn affecting their willingness to talk [27].
This study focused on WTC in oral communication in the EFL class through investigating the participants' WTC with their friends and their teachers while they were in the English class.In the context where English is studied as a compulsory subject and reading and grammar are more familiar to the learners because they are contained in the examinations, it is difficult for most of the learners to develop speaking competence.Therefore, we found that proficiency is related to their willingness to communicate at a moderate level.Because willingness to communicate is related to speaking, which is a productive skill, but proficiency was measured based on receptive skills such as reading, vocabulary, and grammar, as in the English placement test, the correlation between these two variables was not strong.However, when productive skills were measured to show proficiency, correlation was stronger, as seen from the mean score of speaking being the lowest among the four skills measured by TETET (See Table 2).The findings also reflect the role of English in the EFL context, in that students have less opportunity to communicate in the classroom; therefore, they have less confidence.
According to the heuristic pyramid model, Thai students have less exposure to the variables which help them to develop WTC because the only time they can communicate in English is when answering teachers' questions and role playing in front of the class.It is possible that the context in which they are required to communicate is rather limited, so their proficiency might not count.
The context in which the students' WTC was measured is the fundamental English course, which is a compulsory course for all undergraduate students.Classes have up to 40 students, and therefore communication is usually limited to teachers asking some display questions or the types of questions that are used to encourage language practice by having students give their knowledge on a subject matter, rather than real communication.Therefore, there is little to no opportunity for learner interaction.This may explain why the correlation between the "frequency of use" and students' proficiency is weak.The same applies to "perception", which investigates students' intention to communicate with teachers and friends in class in various contexts, such as when teachers elicit ideas from students or students ask questions when their friends give a presentation.Class size also contributes to the frequency of language use.Not all the students had the chance to use English to interact with their friends or the teacher.Even though the teachers want to encourage the use of English in class, when the students work in groups, they prefer to use Thai.Therefore, for low proficient participants, there is no correlation between proficiency and WTC.
The results also show a moderate to weak correlation between "communicative confidence" and proficiency.This may be because in order to be confident to communicate in English, the students need a certain amount of knowledge of the language in order to perform in the target language comfortably.Once the students master the language, they can overcome affective barriers more easily.However, of course, not all proficient students can use English to communicate with confidence in the EFL context; the two are related but not symmetrically or linearly so.
Another possibility that may explain the results is the context of the study.The limited opportunity to develop and exercise their autonomy in language learning may have affected students' motivation, as seen from the moderate to weak correlation between proficiency and WTC.In other words, English proficiency may not affect their motivation to learn, as shown from the results for WTC in class.
Since the ultimate goal of any language teacher is enabling students to communicate in the target language, the results of this study are helpful for EFL teachers to manage their English classes in ways that expose their learners to the target language more frequently, until they can use the language more naturally and are willing to use it when needed.Those classes that create situations where the learners can move from the lower layers of the model we discussed above, and overcome affective barriers, will more likely help them to be more motivated to use the target language and then develop willingness to communicate and use the L2 when they want to.In the EFL context, where interaction normally happens in class, creating an environment which facilitates the use of the L2, as well as encouraging out of class exposure to the L2, is necessary to develop willingness to communicate.
From the findings of this study, we can draw a number of implications.At the classroom level, because proficiency is related to students' confidence, the teacher can try to help students who have a low level of proficiency to be more confident by having successful experiences in communicating in "real" situations where the risks are low.In the WTC heuristic pyramid model, learners' intergroup and interpersonal motivation occur together with L2 self-confidence; therefore, helping the less proficient learners to feel that they belong to the group by providing language tasks that require each member to communicate meaningfully, rather than depending on the proficient ones, would create a "mastery experience" [39] for low proficient learners.This, in turn, helps the learners to be more efficacious in the tasks they are completing.Communicative confidence also affects highly proficient students' WTC.Such students may have more knowledge of the target but may not be confident enough to use it to communicate.Providing the opportunity for this group to use the language to communicate is important, as language is learned through interactive, meaningful communication.Therefore, providing the opportunity for the students to communicate more would also help them to be more confident in using the language and willing to use it.
In a low-autonomy EFL context, such as this study, in order to motivate students, they could be given choices for the tasks they participate in, as this may encourage their WTC.As per self-determination theory, learners need to have autonomy, relatedness, and competence in order to feel motivated [40].Taking this into consideration, teachers may be able to foster learner autonomy at various stages of their teaching by enabling students to make decisions in the tasks they are doing, making sure that they have a sense of belonging in a group work task by having each of the group members know their roles and by scaffolding, so as to overcome their lack of competence.
Even though this study attempted to fill the gap in previous research by including productive skills in measuring proficiency, there is still a limitation of the study in that the context focused only on classroom communication; although language use is likely to be limited outside of class, its extent and its relationship with the development of WTC deserves further investigation.In addition, the correlation was conducted mainly between the scores of proficiency tests and WTC with no confirmed data from other sources.Therefore, further research to identify causes of correlation would make the study more complete.confidentially, and only the researchers will have access to the information you provide.Although I ask for your name, I do so only because I want to associate your answers to this questionnaire with your other data.There are no right or wrong answers.

Perceptions of Willingness to Communicate
Instructions: Below you will read a number of different communication tasks in which you might engage in while studying in English class.I would like you to tell me how willing you would be to do each of these in English.By 'willing' I mean 'showing strong intention', so please put an 'X' in the box that describes the level of your willingness, using the following scales.Answer the teacher during the elicitation stage.
Ask for clarification when you are confused about the task you must complete.
Ask questions of friends who do an oral presentation in front of the class.
Talk with the teacher during class.

Communicative Self-Confidence
Instructions: I am interested in your anxiety about communication and self-perceived communicative competence when communicating in English in the classroom.Put an 'X' in the box that represents the degree to which you agree or disagree with each statement, using the following scales: I feel nervous about using English while participating in class activities.
I say what I want to say in English.
I think my friends/teacher cannot understand me because of my poor English.
I feel comfortable sharing my ideas/feelings/opinions with my friends/teacher in English.
I know the words required for communicating in English.
In general, I find communicating in English in classroom situations relaxing.
I think participating in class activities helps me develop my fluency (i.e., with hesitation and pause.)

Frequency of English Use
Instructions: I am interested in the frequency of communication in the classroom.Please put an 'X' in the box that describes how often you use English, using the following scales: This is the end of the questionnaire.Thank you for your cooperation.

I
am not afraid of making mistakes.I find it difficult to communicate in English.I am worried that I will not understand what my friends say in English.

I
use English to communicate with my friends.I use English to communicate with my teachers.I use English to answer the teacher's questions I use English to check meaning (e.g., 'What does it mean?','I do not understand.')I use English to ask questions I use English for simple interactions (e.g., How are you today?)I use English only when I participate in class activities.

Table 1 .
Correlations between EPT and WTC and TETET and WTC.