Self-perceived communication competence and communication apprehension: A study of Indonesian college students

Considering the important role of Individual Differences (IDs) factors in variations in the ultimate success of learners learning second language (L2, this study investigated Indonesian college students’ two ID components, self-perceived communication competence (SPCC) and their communication apprehension (CA), in English as L2. Several rationales become the basis of the present study. Despite the importance of learners’ SPCC in L2 learning as suggested by several empirical studies outside Indonesia, there have not been any specific study investigating learners’ SPCC in the Indonesian context despite the country being the home of one of the largest number of L2 speakers of English in the world. As a seminal study in the context, the study using quantitative methods enables generalisation which could pave a way for further studies in the field in the Indonesian context. A total of 276 non-English major college students participated in the study in a survey using probability random sampling. Using descriptive statistics in SPSS 21, the study found that learners had medium level of SPCC and medium level of CA. Through correlation formula, the study also found that there was moderate significant association between learners’ SPCC and their CA. Through regression analysis, it was found that their SPCC could predict 23% of variance in their CA. The findings urge teachers to provide as many learning opportunities as possible in enjoyable learning environment to increase learners’ perception of their capability, in turn boosting their confidence to speak in English. Based on the findings and several limitations of the study, several suggestions of future studies are presented.


INTRODUCTION
That there are variations in the ultimate success and achievement of learners learning second language (L2) may have become a common intuitive knowledge among L2 educational practitioners and researchers. Dornyei (2005) attributed this variation of learners' achievement in learning an L2 to learners' individual differences (IDs). He defined IDs as individual learner's characteristics that are unique and differ him or her from other learners. These characteristics explain why learners acquire an L2 differently from one another (Dornyei, 2005;Ortega, 2009). In line with that, Andreou, Vlachos, and Andreou (2006) explained that IDs influence the way learners learn and succeed in their L2 learning. IDs include various learners' characteristics, which include age, gender, language aptitude, learning styles, learning strategies, willingness to communicate, motivation, anxiety, and perceived competence (Ortega, 2009).
One of IDs factors is learners' self-perceived communication competence (SPCC). As the name suggests, SPCC refers to learners' perceptions on the extent of their "adequate ability to pass along or give information" (McCroskey & McCroskey, 1988, p. 109). Taken in the L2 context specifically, SPCC means learners' views on how well they have the capability to communicate in L2. Highlighting the importance of SPCC in L2 communication, Clement, Baker, and MacIntyre (2003) stated that even though learners' "actual competence might influence communication, it is the perception of competence that will ultimately determine the choice of whether to communicate" (p. 192). In other words, at some point, learners' SPCC plays a role in determining whether or not they make actual communication regardless of their real competence.
Acknowledging this role of learners' SPCC in L2 learning, several studies on or related to learners' SPCC have been conducted in various contexts (Bakx, Sanden, Sijtsma, Croon, & Vermetten, 2019 in the Dutch context; Cao, 2011 in the Chinese and Korean context; Denies, Yashima, & Janssen, 2015;Ferla, Valcke, & Schuyten, 2019 in the Dutch context; Jamshidnejad, 2010 in the Iranian context; Joe, Hiver, & Al-Hoorie, 2017 in the Korean context; Shahbaz, Khan, Khan, & Ghulam, 2016 in the Pakistani context, to name a few)). Ferla et al. (2019), for example, reported that the participants scoring high in SPCC were found to be more persistent, less anxious, to "process the learning material at a deeper level, and achieve better study results" (p. 519). In a similar vein, Denies et al. (2015) and Shahbaz et al. (2016) found that learners' SPCC is attributed to their courage to use L2 in class and contribute in class discussions. Shahbaz et al.'s (2016) further stated that as learners have a positive perception on their ability to make meaningful communication in one language, they will likely be more willing to communicate in that language. This confirmed Bandura's (1988) reiteration more than three decades ago that learners' perception of their competence influences their performance in which perception of low competence to perform a task can lead them to be anxious while performing that task and vice versa.
Furthermore, learners' SPCC is believed to play a big role in affecting learners' communication in L2 where the L2 is spoken as a foreign language (FL)  like the condition in Indonesia. In such contexts where most learners are still developing their language skills and where the communication demand is not very complex, learners are more likely to be self-conscious on how well they think they can communicate ideas in L2 despite their still developing language capability (Baker & MacIntyre, E-ISSN: 2528-4479, P-ISSN: 2477 2003; Yashima, 2002). Bakx et al.'s (2019) longitudinal study found that learners gradually gained capability to better assess their language competence by being engaged in communication in that language frequently over long period of time. Bakx et al.'s (2019) findings may be closely related to their participants' prior experiences in using the language during their social work programmes. They noted that their participants probably began to assess their own social-communicative competence more appropriately with more detailed criteria to do so, thus a more "accurate" SPCC.
Regarding this, several other studies indicated that learners' SPCC is indeed greatly affected by learners' prior experiences in using L2. Pleasant experiences in using L2 is believed to be very important to raise learners' SPCC (Dewaele & Dewaele, 2018;MacIntyre & Legatto, 2011). An empirical study by Joe et al. (2017) in Korea, for instance, revealed that learners feel more confident in using L2 when they feel they have had certain achievement in L2 in the past, for example, being successful in English examination. In comparison, when learners feel that they have had negative experiences, such as upsetting, difficult, or stressful ones, they would likely withdraw from future communication engagement in L2 (Cao, 2011). A participant in Cao's (2011) study confessed that she felt angry and frustrated whenever she recalled her negative experience in using L2 and this made her feel inferior on her ability to communicate.
It is posited that learners' SPCC is also affected by the number of interlocutors and learners' perceptions on the communication competence of their interlocutors. Kang (2005) and Kitano (2001) argued that the more interlocutors the learners have, the more likely they would feel apprehensive and anxious. They tend to be more relaxed in communicating when they have fewer interlocutors (Kitano, 2001;Subekti, 2018a). Furthermore, participants in Jamshidnejad's (2010) study reported that as they considered their interlocutors more proficient in communicating in L2, they would rather give up talking because they were afraid of any negative evaluation (Horwitz, Horwitz, & Cope, 1986) in which learners are afraid of being ridiculed and laughed at if they make any errors (Subekti, 2018b). A participant in my recent qualitative empirical study on Willingness to Communicate (WTC) in L2 also reported that his distressing experience in L2 decreased his SPCC and increased speaking anxiety, which in turn hindered him from even trying to communicate in L2 (Subekti, 2019a). Interestingly, this study also found that learners who had low SPCC felt that despite their low SPCC, they were willing to communicate in front of interlocutors they were already familiar with, for example, their close friends in class.
Another IDs factors besides SPCC is communication apprehension (CA). As the name suggests, CA is learners' anxiety when they have to communicate in front of other people (Horwitz et al., 1986). Brown (2000) stated that learners' inability to convey ideas which are too complex can be the source of this anxiety. Gusman (2004) asserted that language learning which involves a lot of practices, trials, and errors can at times instil embarrassment on learners' part, which in turn is attributed to their apprehensive behaviours. CA could be triggered by several factors, the most prominent of which is public speaking situations (Dilbeck et al., 2010;Subekti, 2018b). Dilbeck et al.'s (2010) study, for instance, found that the participants feel apprehensive most in situations requiring them to speak in public. Dwyer and Fus (2002), however, found that learners' apprehension on public speaking can be remedied through continuous practices in Public Speaking class where learners are constantly required to do public practice balanced with public speaking instructions. This finding may give some kind of support to the notion of the influence of successful learning experiences in reducing apprehension (see Grant, 2018 in immersion programme contexts). Besides, my recent empirical study (Subekti, 2019a) also found that learners with pleasant experiences in using L2 tend to be braver speakers whilst those having upsetting experiences tend to be apprehensive.
Furthermore, a large body of research suggests that the two aforementioned ID variables, SPCC and CA, should be seen as inter-related constructs rather than standalone ones (see Blood, Blood, Tellis, & Gabel, 2001;Donovan & MacIntyre, 2004;Dwyer & Fus, 2002;Fahim & Tabatabaeian, 2013;Fushino, 2019). Blood et al.'s (2001) study, for instance, found that learners who stutter severely, thus very possibly experiencing bad experiences in speaking, tend to be afraid of speaking more than their less severe counterparts. This higher apprehension could be attributed to their perception of inability to perform well (Subekti, 2018a(Subekti, , 2018b or low SPCC. Blood et al. (2001) further stated if learners perceive themselves to be incompetent in communication, "they are less likely interact with either peers or teachers" (p. 173). Such apprehension could be the result of frequent past negative experiences with fluency breakdowns (Blood et al., 2001). Additionally, a study in Iran by Fahim and Tabatabaeian (2013) found that learners' SPCC is negatively correlated with communication. In a similar tone, Fushino's (2019) firmly asserted that if a student has low CA and high SPCC in L2, the person is considered to have high L2 communication confidence, necessitated for L2 communication.

Rationale of the present study
The afore-mentioned elaboration on a bulk of literature regarding SPCC and CA may give some kind of support that these two ID variables are indeed very crucial in L2 learning. Despite the existence of these numerous empirical studies, however, some points still need to be further addressed, which prompts the necessity for a further study in the field. First, despite Indonesia having one of the biggest number of L2 English speakers in the world (Kachru & Nelson, 2006), empirical research on the Indonesian learners' SPCC in English is generally still very rare. Several studies did indeed touch this topic as an additional results of their investigation of other topics such as willingness to communicate (e.g. : Muamaroh & Prihartanti, 2013), Foreign Language Anxiety (FLA) (e.g.: Marwan, 2008;Muamaroh & Prihartanti, 2013;Subekti, 2018a), but to the best of my knowledge, there has not been any research investigating learners' SPCC in the Indonesian context despite its crucial role in affecting L2 learning (Clement et al., 2003). Furthermore, whilst it may be true that the Indonesian learners' FLA has been investigated by several previous studies, a specific study investigating the relationship between CA, one of the FLA constructs (Horwitz et al., 1986), and SPCC is worth further investigation as several studies in other contexts suggested the association between the two (e.g.: Blood et al., 2001;Fahim & Tabatabaeian, 2013;MacIntyre & MacDonald, 2009).
The present study was the continuation of my earlier quantitative study focusing on Willingness to Communicate (WTC), often regarded as the primary goal of language instruction (MacIntyre, Clement, Dornyei, & Noels, 1998), on the exactly same participants as those of the present study (see Subekti, 2019b). Subekti (2019b) found that learners reported a high level of WTC and their perceived communication competence was found to be a strong predictor of their WTC in a negative way. Furthermore, in another earlier study of mine (see Subekti, 2019a) with seven selected participants of Subekti's (2019b) study on situational WTC, it was found through qualitative semi-structured interviews that several learners had low SPCC and their low SPCC made them feel anxious to initiate conversation in class, indicating low WTC. One participant in the study commented "[I stay quiet in English class…because I am not very proficient in English.
[I] feel that [I am] still unable [to speak English well]." (Subekti, 2019a, p. 383). This indicated, qualitatively, that SPCC may also have strong relationship with CA. The results of these previous two studies on WTC became the immediate basis of conducting the present study as these two studies suggested the crucial role of SPCC in influencing L2 communication and its possible relationship with apprehension in communication in the Indonesian context. Additionally, the quantitative nature of the present study would allow generalisable findings (Gray, 2014) which can pave a way for further research in the field of SPCC in the Indonesian context.
Based on the afore-mentioned rationales, this study seeks to answer these research questions. First, how is college students' self-perceived communication competence (SPCC)? Second, how is the students' level of communication apprehension (CA)? Third, to what extent does their SPCC affect their CA?

METHODS
This study used quantitative research design involving 276 non-English major college students from a total of 402 students. It used SPSS 21 to conduct descriptive and regression analyses to answer the research questions. Probability random sampling was used to select these participants as it was believed that the population of the research was homogenous with regard to the research questions (Gray, 2014). As the present study was the continuation of Subekti's (2019b) study, both shared the same instruments. They were fifteen questionnaire items on learners' SPCC and fifteen items on learners' CA. Several items used in Matsuoka's, Matsumoto's, Poole's, and Matsuoka's (2014) study in the Japanese context became the primary source of the present study's questionnaire items on SPCC and CA. The items on SPCC were adapted from Matsuoka et al.'s (2014) and Peng's and Woodrow's (2010) studies. The other fifteen items on CA were adapted from the works of Horwitz et al. (1986) and Matsuoka et al. (2014).
The choice of using combination of items from different established questionnaires was based on sound rationales. It was expected to combine relevant items in accordance with the research context. Specific to questionnaire on SPCC, it was fully realised that McCroskey's and McCroskey's (1988) questionnaire on SPCC is by far the most widely used in SPCC literature (e.g.: Blood et al., 2001;Dilbeck et al., 2010). However, the items may be more related to SPCC in L1 rather than in L2 as they measure learners' SPCC in talking to stranger, acquaintances, or friends. Such items as "Present a talk to a group of strangers" and "Talk in a large meeting with friends" found in the questionnaire, for examples, may not be quite relevant to the present study's Indonesian contexts, in which English communication may be constrained largely to classroom context, and the participants, who may unlikely encounter such situations in L2 communication. Thus, providing questionnaire items relevant to the participants' classroom communication, such as "I am able to do a role-play standing in front of the class in English (e.g., ordering food in a restaurant)" and "I am able to give a short selfintroduction without notes in English," was deemed important to obtain more valid data (Gray, 2014).
Participants were to answer the questionnaire items using four possible responses: "Strongly agree", "Agree", "Disagree", and "Strongly disagree", each of which would be numerically scored afterwards for data analysis. For items on SPCC, the possible responses were: "Strongly agree", equal to four (4), "Agree", equal to three (3), "Disagree", equal to two (2), and "Strongly disagree", equal to one (1). In comparison, for items on CA, "Strongly agree" was scored one (1), "Agree" two (2), "Disagree" three (3), and "Strongly disagree" four (4). The reverse scoring was based on the assumption that learners' CA would likely be in contrast with their SPCC in which learners' belief on their communication competence to perform a task can affect the extent to which they feel anxious while performing that task (Bandura, 1988;Shahbaz et al., 2016).
There were 276 students participating in this study. These participants were students of General English course at a college in Java, Indonesia. They were, in general, in upper beginner level or lower intermediate level of English. There were around 400 students taking the course at the time of data collection. This study employed probability random sampling in which these selected participants were selected randomly with the belief that the population was in general homogeneous in relation with the problems of the research (Walliman, 2011).
The questionnaires were distributed from 7 May 2018 to 18 May 2018 in the even semester of 2017-2018 academic year after consent was granted by the Director of the Language Centre as the gatekeeper (Gray, 2014). The questionnaire was distributed along with individual consent form on the first page (Israel & Hay, 2006). Each participant was given time to understand the contents of the informed consent form detailing the researcher's identity, the purpose of the study, their rights whether or not to participate in the study, and data confidentiality shall they decide to participate (Israel & Hay, 2006).

RESULTS AND DISCUSSIONS
The present study has the same demographic data of the participants with that of my previous work (see Subekti, 2019b) as the two used exactly the same participants and research context with two different research foci. From 276 students participating in the present study, 163 were female students (51.9%), whilst 113 others were male students (40.9%), with the youngest being 17 years of age, and the oldest being 24 years of age (SD = 1.09).
To ensure the internal reliability of the questionnaire items, reliability analysis was conducted in which questionnaire items could be said to have a high internal reliability if their Cronbach's alpha coefficient is close to one (1) (Bryman, 2012). The fifteen items of learners' SPCC had .87 Cronbach's alpha coefficient, indicating a high internal reliability. The fifteen items of learners' CA .92 Cronbach's alpha coefficient, indicating an even higher internal reliability.

Research question 1: How is college students' SPCC?
The mean score of the questionnaire items on SPCC was 41.32 (SD = 5.45) indicating, in broad sense, learners' medium level of SPCC. The detailed results of the participating students' responses could be seen in Appendix 1. As observed in Table 1, interestingly, with "Strongly agree" scored four (4.0) and "Agree" scored three (3.0) indicating high SPCC for each of the questionnaire item, there were only four items of fifteen yielding mean scores more than three, indicating learners' high SPCC. Nine items yielded mean scores between 2.5 to 2.9, indicating medium SPCC as the majority, with the other two items yielding mean scores lower than 2.50, indicating low SPCC. The items yielding high mean scores were items number four, seven, fourteen, and fifteen. The highest SPCC was from item number fourteen "I know I can speak English if I prepare for speech well," in which the mean score was 3.17. This statement was endorsed by 96% of the respondents. This massive response may closely be related to item number fifteen, in the third position, "I can manage to make a speech if I prepare the draft," endorsed by 86.6% of the respondents. This result confirmed several previous studies' findings that learners need to have preparation time before speaking in L2 (e.g.: Kang, 2005;Marwan, 2008;Subekti, 2018a). In the Indonesian context, this may especially be the case as learners' exposure to English is fairly limited to classroom contexts. As such, they may not feel confident enough to speak spontaneously without any prior preparation. The second highest mean score, at 3.17, was from the fourth item, "I am able to give a short selfintroduction without notes in English." 92.7% of the participants either strongly agreed or agreed to the statement. This result may be attributed to the possibility that doing self-introduction was considered relatively simple and uncomplicated for the Indonesian participants, who were already in their upper beginner or lower intermediate level of English. This may also be the case in explaining why item number one, "I am able to give my peer sitting next to me directions to my favourite restaurant in English" also obtained favourable responses from 78.3% of the participants. These participants may consider giving directions relatively simple task. The result may be attributed to learners' pleasant experiences in using L2 (see Dewaele & Dewaele, 2018;Joe et al., 2017;MacIntyre & Legatto, 2011) for accomplishing simple communication tasks.
Another item with a high mean score, at 3.05, was the seventh, "I know I can manage to speak English using words I know even if it is not correct," suggesting that most of the respondents perceived themselves as capable speakers provided that language accuracy was not the benchmark. This result may give some kind of support that excessive emphasis to accuracy may hinder learners from talking. This could also be related to learners' negative and upsetting experiences in using L2 (Cao, 2011), especially its grammar. Indonesian EFL instruction which generally pays attention to grammatical accuracy may contribute to these upsetting experiences. In general, these results support some authors' ideas that in EFL contexts, such as Indonesia, where communication demand is not very complex, learners tend to be selfconscious on how well they could communicate (see Yashima, 2002). Furthermore, despite the participants' medium to high SPCC level in general, there were two items worth commenting as they yielded low mean score indicating low SPCC. 62.7% of the respondents either strongly disagreed or disagreed with item number eight, "I am able to speak without being nervous" whilst 72.8% of the respondents also opposed item number eleven, "I feel more comfortable in speaking English in public". These results were consistent with some authors' reiteration that among language classroom activities, speaking worries learners most (e.g.: Horwitz et al., 1986;Subekti, 2018aSubekti, , 2018b, let alone speaking in front of many people (Dilbeck et al., 2010).
There may be several factors attributed to this result. First, learners may not feel confident when talking in public as this task could make them lose "face" in front of their peers shall they make errors (Horwitz et al., 1986;E-ISSN: 2528-4479, P-ISSN: 2477 Subekti, 2018b). This may especially be true in Asian culture, such as Indonesia, which tends to consider how others consider themselves very important (Joe et al., 2017;Subekti, 2018b). They may likely be more concerned about how they look especially when they do not feel familiar with the interlocutors (Jamshidnejad, 2010). This, however, as Dwyer and Fus (2002) suggested, could be remedied with continuous practice allowing learners to go beyond their comfort zone through repeated classroom instruction on public speaking.

Research question 2: How is the students' level of CA?
The mean score of the questionnaire items on CA was 33.72 (SD = 7.26), indicating medium level of CA. The detailed results of the students' responses could be seen in Appendix 2. Table 2 shows the means of their responses on each of the questionnaire item in which low mean scores indicated higher apprehension. As observed in Table 2, only two items yielded means scores less than 2.0 with twelve items being between 2.0-2.50 range and an item at 2.74. These results suggested that the participants were in general in the range of moderate apprehension. Several items in which fast majority of the participants reported their apprehension, however, may need further discussions. Statement number six, "I feel a little nervous if my English is wrong when speaking in public," was confirmed by 85.9% of the respondents, whilst statement number eleven, "I feel very self-conscious about speaking English in front of other students," was confirmed by 71.7% of the respondents. These two results may suggest that public speaking situation is problematic for the participants. This apprehension of public speaking situation may be attributed to hesitation in speaking in class and possible perception of inability to perform well (Subekti, 2018a(Subekti, , 2018b. In relation with this, Gusman (2004) asserted that L2 learning requires learners to engage in public practices including trials and errors, which very often cause embarrassment and frustration leading to apprehensive behaviours. This possibility may increase if learners are to speak in front of interlocutors they are not really familiar with (Jamshidnejad, 2010;Subekti, 2019a) and as L2 learning often requires learners to do such public practices (Dilbeck et al., 2010), this apprehensive tendency seems to be frequent in L2 situations (McCroskey et al., 1985).
Furthermore, that 83.4% of the respondents either strongly agreed or agreed to statement number one, "I felt tense in presentation," amplified the notion that many Indonesian L2 learners of English are apprehensive of presentation situations (Marwan, 2008;Subekti, 2018a). That 70.6% of the respondents also agreed to statement number seven, "No matter how many times I may make a speech, I feel tense," may support the prevalence of notion. In addition, this may also have something to do with the big number of interlocutors the participants were typically required to speak in front (Kang, 2005;Kitano, 2001). Kitano (2001) posited that learners are more likely to be less apprehensive if they are required to speak in front of less number of people, which may unlikely happen in typical presentation or speech situations in language classes. Furthermore, the respondents also reported high apprehension by approving statement number three, "I feel worried that I might use strange English," (75.7%) and number nine, "It frightens me when I don't understand what the teacher is saying in English" (71.4%). This can solidify previous studies' findings of learners' fears of negative evaluation (e.g.: Horwitz et al., 1986;Kitano, 2001;Subekti, 2018b). Learners may be concerned of being considered stupid or of being ridiculed if they speak English "differently" or if they ask questions shall they not understand their teachers' explanations. Asian culture, like the Indonesian culture, considering "face" very important may strengthen Asian learners' fear of doing such things (Joe et al., 2017;Subekti, 2018b).

Research question 3: To what extent does students' SPCC affect their CA?
The study found a statistically significant negative association between learners' SPCC and their CA. The correlation strength was at moderate level, r (274) = -.48, p < .01. It means that the higher learners' SPCC, the lower their CA tended to be. The correlation result could be seen in Table 3.
To see the extent to which learners' SPCC could predict their CA, regression formula was used. The result could be seen in Table 4. As seen in Table 4, the value of R 2 was .23, indicating that learners' SPCC could predict 23% of their CA. The afore-mentioned findings need further comments. First, the result was slightly in line with the result of Fahim's and Tabatabaeian's (2013) study in Iran. They found that learners' SPCC could predict 17.80% of variance in learners' FLA, one aspect of which was CA. The slightly similar result of the present study and that of Fahim and Tabatabaeian (2013) in Iran may suggest the consistency of the moderate inter-relatedness of learners' SPCC and their CA. Many authors argue that perception of inability to perform well, perhaps through series of upsetting experiences (see Blood et al., 2001;Cao, 2011), often leads to higher level of apprehension (Donovan & MacIntyre, 2004;Subekti, 2018b). Hence, reducing their apprehension and increasing their SPCC, learners could attain high L2 communication confidence necessary for L2 communication (Fushino, 2019).
However, considering the moderate effect of SPCC on CA, at 23%, this study could not overlook other variables, making up the remaining 77%, that may affect learners' CA. That is to say that a lot more than just learners' SPCC was at play to affect learners' CA.

CONCLUSION
The present study has several implications related to L2 instruction. First, considering the effect of SPCC on CA, it is suggested that teachers design activities boosting the growth of learners' SPCC. Classrooms should be made into enjoyable environment where learners can feel safe that eventually will lead to risk-taking behaviours in communication. Subekti (2018a) posited that it may be a good idea for teachers to focus more on their efforts in learning than their speech outcome per se as this may boost learners' confidence to speak despite their limited competence. Teachers are expected to give learners more opportunities to speak in front of fewer number of audience. Shall they ask learners to speak in public, learners should be given enough time to prepare for their performance.
Despite the possible contribution of the study in the field of SPCC and CA, especially in the Indonesian context, several limitations are acknowledged. First, self-report questionnaires used in the study may inherently bring the consequence that the findings depended on learners' report. Additionally, considering the quantitative analysis and the large number of participants involved, it is persuasive to say that the results of this study could be generalised to a wider population. Nevertheless, they should be seen within the population context, Indonesian non-English major college students. Learners from different age groups, different level of education, or different educational contexts, may have different characteristics. As such the findings of the present study may not be able to represent them. Inherently, unlike qualitative studies, this quantitative study was also unable to capture learners' experiences in depth.
Furthermore, several suggestions for future studies could be offered. First, this study used two questionnaires, each of which contained items taken from different studies. The items on learners' SPCC had .87 Cronbach's alpha coefficient and the items of learners' CA had .92 Cronbach's alpha coefficient, indicating the newly-built questionnaires have high internal reliability. Hence, it may be worthwhile to conduct further studies in the same field using these questionnaires. Secondly, it can be useful to investigate learners' SPCC in lower educational levels in Indonesia as these learners typically obtain very little exposure to communicative activities in L2. Additionally, considering there are possibly still a number of unknown factors affecting learners' CA other than their SPCC in the present study, it is suggested that future studies conduct quantitative research in Indonesia on the possible inter-relationship among IDs variables such as SPCC, motivation, apprehension, personality, and WTC, using Structural Equation Modelling. Finally, considering the participants' apprehension and low SPCC in public speaking situations, future studies could also investigate several selected learners' experiences in dealing with public speaking practices during a period of one semester to know whether their SPCC and apprehension fluctuate across time in response to different experiences and L2 instructions.

ACKNOLEDGEMENT
I would like to thank the English Language Education Department of Universitas Kristen Duta Wacana for funding this research. My gratitude also goes to the Director of the Language Training Centre where I conducted this study and all the participants. Lastly, I would also like to thank the anonymous reviewers for their valuable feedback. 11. I feel very self-conscious about speaking English in front of other students.
14.5 57.2 26. 4 1.8 12. I get nervous and confused when I am speaking in my English class.