Pronunciation Anxiety of Indonesian Pre-Service English Teachers: Do Study Duration and Gender Matter?

Introduction

Second/foreign language (L2) learning requires particular procedures involving public practices and experimentation. This repetitive trial-and-error process in L2 learning may put pressure on learners, leading to anxiety (Tran, 2012). Therefore, it is quite common to see learners who perform well in other subjects encounter challenges and feel anxious in L2 learning (Subekti, 2018b). Studies suggested that anxiety can negatively impact second/foreign language (L2) learning (Ahsan et al., 2020; Alimorad & Adib, 2022; Limeranto & Subekti, 2021). When communicating in a foreign language, language learners may experience anxiety due to fear of making mistakes, being misunderstood, or being negatively evaluated by others (Ali & Anwar, 2021; Daud et al., 2019; Pan & Zhang, 2021; Subekti, 2018a). Anxiety can lead to low self-esteem and confidence, negative self-talk, and self-doubt, all of which may lead to a lack of motivation and interest in language learning (Aubrey et al., 2022).

One of the forms of anxiety in language learning is L2 pronunciation anxiety (PA), which refers to the fear learners experience when pronouncing words in a particular language, particularly when they are unsure of the ‘correct’ pronunciation (Baran-Lucarz, 2014; Kralova et al., 2018). It is a common occurrence for language learners, and it can lead to a lack of confidence and even embarrassment inhibiting the development of their communication skills (Baran-Lucarz, 2014; Plailek & Essien, 2021).

Studies have been conducted investigating PA, for example in Poland (Baran-Lucarz, 2014), Iran (Alimorad & Adib, 2022), Pakistan (Ahsan et al., 2020), Korea (Baran-Lucarz & Lee, 2021), Japan (Harumi, 2021), and China (Yang, 2017). For example, a study by Baran-Lucarz (2014) involved 151 Polish learners of English. It reported that learners' PA strongly predicted their willingness to communicate (WTC). It was also found that the-low anxiety group had a significantly higher level of WTC than the high-anxiety group. The finding could suggest the negative effect of PA. Another study was conducted recently by Alimorad and Adib (2022) involving 134 Iranian L2 learners of English at two private language institutes. The study reported that PA significantly predicted willingness to communicate negatively. It means that the higher the PA, the more unlikely learners would be willing to use L2 in communication. Earlier, involving 90 Chinese preservice English teachers, a study by Yang (2017) investigated the relationship between learners’ self-efficacy and pronunciation performance. It reported that only the mastery experience dimension of self-efficacy had a significant predictive impact on learners’ pronunciation. It suggested that learners’ experience of success in pronunciation was paramount in shaping their confidence in pronunciation. Regarding this, in Pakistan, a study involving 200 university learners by Ahsan et al. (2020) reported that learners' anxiety became a major barrier to learners' English pronunciation, suggesting the detrimental effect of PA on language learning.

In the Indonesian context, though not very extensive, at least two studies have been conducted to investigate PA (Aulia, 2018; Maharia et al., 2021). A study by Aulia (2018) investigated the effectiveness of the use of a WhatsApp group as a medium in which 18 learner participants posted their voice records for feedback in the Pronunciation Practice course. Though learners' achievements varied by the end of the course, 14 learners (78%) acknowledged that the WhatsApp group helped make pronunciation learning more fun and enjoyable. Furthermore, in a study involving 48 Indonesian preservice English teachers by Maharia et al. (2021), it was found that learners’ PA correlated with their learning motivation. However, the relationship was not significant, suggesting that other factors were probably at play affecting learners’ PA. Other studies mentioned learners’ anxiety associated with poor pronunciation as a factor contributing to low willingness to communicate (Muamaroh & Prihartanti, 2013), and speaking anxiety in general (Subekti, 2018a).

Several studies have discussed possible causes of PA in L2 learning (Alimorad & Adib, 2022; Baran-Lucarz, 2016; Baran-Lucarz & Lee, 2021; Harumi, 2021; Thompson & Lee, 2014). Baran-Lucarz (2016) argued that PA is a multifaceted construct composed of four sub-components, which though independent, interact with each other dynamically (Baran-Lucarz, 2016), and thus can be seen as factors influencing PA. These components are learners' fear of negative evaluation from potential interlocutors, pronunciation self-efficacy, pronunciation self-image, and beliefs concerning the difficulty of the phonological system of the target language. A study in Korea by Baran-Lucarz and Lee (2021) investigated potential determinants of PA. The study revealed that learning experiences with native-speaker teachers, the experience of studying abroad and L2 learning motivation reduced learners’ PA. In a similar vein, in her study investigating PA among 64 Japanese L2 learners of English, Harumi (2021) identified sources of PA among the participants. These sources of PA were the need to be a part of the Japanese community, the need to impress people in the immediate context, accuracy-focused teaching, social expectation, and language ownership. For example, some learners were unsure about how they should speak or pronounce words in English in front of their Japanese friends. Thus, they deliberately used heavily accented Japanese English in their pronunciation. The study suggested that learners’ PA likely stems from their self-presentational concerns. Furthermore, previous studies reported the relationships between PA and the experience of living abroad (Alimorad & Adib, 2022; Baran-Lucarz & Lee, 2021; Thompson & Lee, 2014). The findings of these two previous studies suggested that learners’ PA decreased as time spent abroad increased.

Furthermore, though not specifically investigating PA, research indicates that when language learners believe that the main goal of L2 learning is to become as proficient as a native speaker, it can cause them to feel anxious (Sadeghi & Abdi, 2015). This is because many L2 learners feel pressure to attain a high level of proficiency and to sound like native speakers, which can lead to inflated expectations and feelings of inadequacy or frustration when these goals are not met (Zhang & Zhong, 2012). Exposure to recordings and videos depicting interactions among native speakers may prompt learners to regard native speaker proficiency as the benchmark standard for language proficiency. This may lead to anxiety if they cannot meet this standard (Elaldi, 2016; Quintos, 2022).

Although various factors determining PA have been quite extensively discussed, there has been a limited number of studies investigating the relationship between study duration and PA (Kafes, 2018; Tekten, 2020). Involving 596 university learners in Türkiye, Tekten (2020) found that PA was higher in learners who had never been abroad or in learners learning English for a shorter period. Earlier, another study was conducted by Kafes (2018) involving 75 freshmen at a Turkish university. The study reported that learners who had intensive English courses before enrolment at university had a higher level of PA than learners who did not. This finding, interestingly, suggested a positive association between learners' awareness attributed to exposure to the language and PA. It means that exposure to the language, instead of lowering learners' anxiety, increased it. Despite the contributions of these studies, it may still be worthwhile to conduct a study investigating whether PA differs based on study duration in the under-researched Indonesian L2 context. Besides, in the context of English education where learners are prepared to be future English teachers, the possible effect of study duration in the department on learners’ PA can be intriguing to be investigated.

Another plausible factor in PA is gender. The role of gender has been extensively investigated in language learning. And though the effects of gender on speaking anxiety have been investigated in several studies (Alsowat, 2016; Arnaiz & Guillén, 2012; Dewaele & MacIntyre, 2014; Mohtasham & Farnia, 2017), very few studies have specifically investigated the role of learners’ gender on their PA (Kafes, 2018; Tekten, 2020). Kafes (2018) reported that learners’ gender was not one of the determinants of their PA. In contrast, Tekten (2020) reported that female learners had significantly a higher level of PA than their male counterparts. These findings suggested that the role of gender in PA is still inconclusive. It is realized that many studies on speaking anxiety seemed to suggest that female learners tended to have higher anxiety levels than their male counterparts (Arnaiz & Guillén, 2012; Dewaele & MacIntyre, 2014; Mohtasham & Farnia, 2017). Nonetheless, acknowledging pronunciation, although it may be linked to speaking anxiety, as a separate concept (Baran-Lucarz, 2017), it is still crucial to conduct a study specifically examining the potential relationship between gender and PA.

The Present Study

The present study intends to answer the following research questions. First, do Indonesian English language education learners experience pronunciation anxiety (PA)? Second, does PA significantly differ across learners from different durations of study? And third, is there any significant difference in PA between male and female learners?

The rationale for conducting this study is as follows. In the Indonesian context, home to one of the largest L2 speakers of English in the world, studies on PA are still very limited and mainly conducted qualitatively involving certain classes or courses only. Therefore, it is necessary to carry out an empirical study on PA involving more participants in the context. Furthermore, there are limited studies contemplating the possible association between study duration and PA. Regarding this, it is strategic to involve, in the present study, learners from the English language department. It is because these learners likely obtain more English exposure in their study than typical L2 learners in such English as a Foreign Language (EFL) settings as Indonesia. Next, it is also worthwhile to carry out an empirical study delving into possible gender roles in PA considering the scarcity of such empirical studies in the literature. As a whole, the present study may set the stage for further investigations on PA and the potential interplay with study duration and gender in Indonesia and, potentially, beyond.

Methods

Research Design

The present study employed a survey method of distributing an online questionnaire via a Google Form. The study was conducted quantitatively because of two reasons. The use of a survey study conformed with the research objectives of this study. That was to investigate Indonesian pre-service English teachers’ level of PA and the extent to which learners’ PA differed across study duration and gender. Secondly, in the Indonesian context, studies on PA have thus far been conducted qualitatively or involving a certain class only (Aulia, 2018; Maharia et al., 2021). For this reason, a quantitative study may produce findings that can provide an overview from an eagle-eyed perspective on Indonesian pre-service English teachers’ PA.

Participants and Ethical Considerations

This study employed convenience sampling, indicating that participants were recruited from a conveniently available group of prospective participants (Gray, 2014). The participants of this study were 101 learners studying at an English Language Education department at a university in South Sulawesi, Indonesia. They were typically multilingual, knowing how to speak at least the Indonesian language and Torajan language apart from English they were formally learning at the university. The range of their age was from 17 up to 30 with an average of 20.14 (SD=2.02). Of these 101 participants, 87 participants (86.1%) were females whilst 14 participants (13.9%) were males. The gender proportion of this sample, though seemingly unbalanced, mirrored the gender proportion of the learners in the department, dominated by female learners. In terms of their study duration, 23 participants (22.8%) were in the first year of their study, 36 participants (35.6%) were in their second year, 19 participants (18.8%) were in their third year, and 23 participants (22.8%) in their fourth year.

In this study, ethical principles of conducting research were strictly adhered to. Before the study was conducted, gatekeeper consent (Ramrathan et al., 2016) was secured from the head of the English language education department where the prospective participants studied. In the process of data collection, the study employed the autonomy principle (Israel & Hay, 2006) where prospective participants had the freedom to participate in this study. Through the consent form at the beginning of the online questionnaire, the prospective participants were informed of the objectives of the study and their rights and responsibilities if they decided to participate. The instruments were also designed in such a way that they were easy to complete and did not require much time to fill, suggesting the implementation of non-maleficence and justice (Creswell, 2014). Confidentiality of the participants is also maintained throughout this report.

Instruments

This study used an online, Google Form questionnaire to collect the data. The questionnaire set consists of the consent form, a demographic questionnaire such as age, gender, and semester of their study, and a questionnaire on PA.

The questionnaire on PA consists of fourteen Likert-scale statements and was adapted from a questionnaire named Foreign Language Pronunciation Anxiety Scale (FLPAS) designed by Kralova et al. (2018) for their study in Slovakia. It was inspired by the Foreign Language Classroom Anxiety Scale (Horwitz et al., 1986),  the Phonetics Learning Anxiety Scale (Baran-Lucarz, 2013), the experiences of Kralova et al. (2018) on teaching phonetics course, and their reflections of learners’ opinions on their pronunciation weak points and worries. In the present study, for each statement, there were five possible responses, Strongly Agree, Agree, Undecided, Disagree, and Strongly Disagree. Minor adjustments were made to contextualize the statements in the context of the present study. For example, the item ‘It seems to me that I cannot get rid of my Slovak accent in English’ was modified into ‘It seems to me that I cannot get rid of my Indonesian or Torajan accent in English’.

The modified questionnaire items were tested for validity using correlation and reliability using Cronbach’s alpha and McDonald’s omega. The correlation values between each of the items and the overall construct were examined. It was to ensure that each of the items measured the intended variable. All items but items 7 and 9 produced a statistically significant correlation with the overall construct of PA. The correlation values were also compared to the critical value for Pearson’s correlation coefficient (N = 101, df = 99). All items but items 7 and 9 produced values higher than the critical value, indicating validity. Items 7 and 9 producing values lower than the critical value were omitted and excluded from the analysis. The Cronbach’s alpha coefficient of the twelve valid items was .81 whilst the McDonald’s omega coefficient was .82, both indicating high reliability.

Data Collection Procedure and Data Analysis

The data collection process was conducted online for approximately three weeks from 4 October 2023 up to 20 October 2023. The questionnaire link was distributed in the class WhatsApp groups with the help of the teachers at the university. After that, the obtained data were recorded into SPSS 25 for further analysis. The responses to the Likert-scale statements on PA were recorded as follows, 5 points for Strongly Agree, 4 points for Agree, 3 points for Undecided, 2 points for Disagree, and 1 point for Strongly Disagree. However, four of the fourteen items were negative statements where the Strongly Agree response indicated a high level of PA. Hence, the responses in these four items were reverse-scored where Strongly Agree was recorded as 1 point and Strongly Disagree as 5 points. Furthermore, gender was recorded as a nominal variable where females were indicated as zero (0), and males were indicated as one (1). Likewise, duration of study was also recorded as a nominal variable where 1 point indicated the first year of study, 2 points indicated the second year, 3 points indicated the third year, and 4 points indicated the fourth year.

After the validity and reliability tests were executed, statistical formulas were executed to answer the research questions. To answer the first research question on learners' PA, descriptive statistics were employed. The data were presented in the forms of means, percentages, and standard deviations. To answer the second research question on the possible role of study duration on PA, a chi-square test was performed with two variables: duration of study and PA. Finally, to answer the third research question on whether there was a significant difference in PA across genders, a t-test was performed with two variables: gender and PA.

The sequence of data collection and analysis can be observed in Figure 1.

Figure 1. The Sequence of Data Collection and Analysis

Findings

The Participants’ Pronunciation Anxiety Level

The present study found that the mean of the participants' PA measured using a 12-item Likert-scale questionnaire was 38.86, with the minimum being 21 and the maximum being 56 (SD=7.23). The overall mean score of the participants' total PA indicated that on average, the average mean score was 3.24 on a scale of 1 up to 5. This suggested that the participants experienced PA at a moderate-high level.

To allow for more thorough and detailed analyses of the findings on each of the questionnaire statements, the fourteen items are presented in three different categories: Pronunciation Anxiety related to Negative Evaluation from Potential Interlocutors and Peer Comparison, Pronunciation Anxiety related to Perception of Competence, as well as Pronunciation Anxiety related to being English as Foreign Language (EFL) Speakers. The findings on the first category, comprising items 1, 2, 4, 5, and 6 can be seen in Table 1.

Table 1. Pronunciation Anxiety Related to Negative Evaluation from Potential Interlocutors and Peer Comparison

Some findings in Table 1 can be highlighted. Out of 101 participants in total,
72 participants (71.3%) felt embarrassed talking to people with good English pronunciation and 78 participants (77.2%) got nervous when they had to pronounce English words in front of other people. Furthermore, 76 participants (75.3%) felt embarrassed when realizing their pronunciation mistakes in front of others, 66 participants (65.4%) were afraid of being judged negatively due to their poor pronunciation, and 68 participants (67.4%) were afraid of being less competent in pronunciation than their future students. The mean scores of the items in these categories ranged from 3.41 up to 3.64, suggesting learners’ PA concerning interlocutors and peer comparison was generally at a moderate-high level.

Moreover, the findings on the second category, Pronunciation Anxiety related to Perception of Competence, comprising items 3, 8, 11, and 12, can be observed in Table 2.

Table 2. Pronunciation Anxiety related to Perception of Competence

Some findings presented in Table 2 warrant further comments. As seen from items 11 and 12, respectively, out of 101 participants in total, ‘only’ 33 participants (32.6%) thought they could never master good English pronunciation and 68 participants (67.4%) did not think that English pronunciation was difficult. Interestingly, as seen in items 3 and 8, respectively, 54 participants (53.5%) reported their dissatisfaction with their pronunciation and 84 participants (83.2%) were worried about not being understood because of mispronunciation. These findings perhaps indicated that despite learners’ optimism in improving their pronunciation, they were generally still dissatisfied with their achieved levels and afraid that their pronunciation had poor intelligibility. The mean scores of the items in this category suggested ambivalence. On one hand, items 11 and 12 produced mean scores lower than 3.00, suggesting low PA and a certain degree of pronunciation self-efficacy. On the other hand, items 3 and 8 produced mean scores higher than 3.00, indicating a moderate-high level of anxiety and a certain degree of negative perception of pronunciation competence.

As for the findings on the third category, Pronunciation Anxiety related to being EFL Speakers, comprising items 10, 13, and 14, they can be seen in Table 3.

Table 3. Pronunciation Anxiety Related to Being EFL Speakers

From Table 3, the findings can be highlighted. Items 10 and 14 produced mean scores higher than 3.00, suggesting a moderate-high level of PA concerning comparison between learners’ pronunciation and that of native speakers. As seen in items 10 and 14, respectively, out of 101 participants in total, 57 participants (56.4%) reported they could not get rid of their Indonesian/Torajan accent and 72 participants (71.3%) reported it was very difficult for them to pronounce English words like a native speaker. However, as seen in item 13, 78 participants (77.2%) expressed their disagreement on the rules of English pronunciation being incomprehensible, suggesting low PA. This item produced a mean score of 2.39.

The Participants’ Pronunciation Anxiety: Interactions with Study Duration and Gender

Using a chi-square test of independence, the study found that there was no significant difference in learners’ PA level attributed to the duration of their study, ꭓ² (90, 101) = 84.24, p > .05. This finding suggested that the participants’ PA barely had any interaction with the duration of their study at the English language education department.

Furthermore, through an independent sample t-test, the study found that female learners reported a higher level of PA (M= 39.46, SD=6.95) than male learners (M=35.14, SD =8.06) and there was a significant difference between female learners’ PA and that of male learners, t(99) = 2.11, p < .05. These results can be observed in Table 4.

Table 4. Learners’ Pronunciation Anxiety by Gender

Discussion

The Participants’ Pronunciation Anxiety Level

Using descriptive statistics, this study found that learners generally had a moderately high level of PA. Learners reported a fairly consistent level of moderate-high PA concerning interlocutors and peer comparison. This finding was in line with the reiteration of Baran-Lucarz (2016) stating that learners’ fear of negative evaluation from potential interlocutors leads to PA. Though not extensively addressed in the PA literature, several studies on L2 speaking anxiety have suggested that the fear of negative evaluation from peers and teachers instilled anxiety (Kurakan, 2021; Subekti, 2018a) and hindered them from using L2 (Lee, 2016). In the case of the Indonesian participants in the present study, cultural aspects may also be at play. Joe et al. (2017) suggested that his Asian learner participants considered ‘face’ very important and thus they tended to avoid situations that may embarrass them in language learning such as making mistakes in front of others. The Indonesian participants in the present study may not be very different, leading to the prevalence of learners' PA associated with fear of negative evaluation from others.

Furthermore, this study found that learners’ PA concerning their perceived pronunciation competence seemed to be rather ambivalent. They reported moderate to high levels of anxiety due to concerns about their pronunciation being misunderstood. However, they seemed to have a certain level of pronunciation self-efficacy, believing that they could master English pronunciation. This self-efficacy may partly be attributed to the extensive English pronunciation exposure at the English language department. Living abroad experiences, which academic life at English education departments may slightly resemble in terms of language exposure, were reported to be associated with reduced PA levels (Alimorad & Adib, 2022; Thompson & Lee, 2014). Hence, it can be speculated that this exposure to the language may influence learners’ pronunciation self-efficacy. Besides, as learners studying at an English language department, the participants may see their L2 possible selves as English teachers or proficient English-using selves in the future. This may partly explain their optimism in their pronunciation learning despite their fear of their pronunciation being misunderstood. Another angle to see this particular finding was that learners may be overly critical of their pronunciation considering they had chosen English as their future career path. A study involving 239 Iranian high school learners by Ghorbandordinejad and Nasab (2013) found a significant association between learners' perfectionism and anxiety, suggesting that when they were overly critical of their language performance, they tended to be more anxious. Nonetheless, in the case of the present study, qualitative investigations may be necessary to delve further into the issue from the perspectives of learners and their teachers in the department.

This study further found that learners generally had moderate-high PA related to being EFL speakers. Most of them reported they could not get rid of their Indonesian/Torajan accents and did not believe they could pronounce words the way native speakers do. Baran-Lucarz (2016) mentioned that one of the components of PA is learners' beliefs concerning the difficulty of the phonological system of the L2. When learners believe the L2 phonological system is difficult to master, they tend to be more anxious. In a study in Korea, it was also found that exposure to native-speaker teachers lessened learners' PA (Baran-Lucarz & Lee, 2021), perhaps suggesting that learners looked up to native speakers as the benchmark of their L2 competence. In English language departments, learners’ language skills are typically trained using academic sources oriented towards native speaker norms, including pronunciations. Hence, it was not surprising that most of the learner participants in the present study expressed their concern about being unable to attain pronunciation performance like that of native speakers.

Interestingly, however, with the emergence of English as Lingua Franca (ELF) as opposed to EFL, there has been a growing acknowledgment of varieties of English spoken by L2 speakers from various L1 backgrounds (Rose et al., 2020) as long as intelligibility is achieved (Rose & Galloway, 2019). Besides, L2 speakers of English with distinctive cultural and linguistic backgrounds should be seen in their own right rather than being evaluated merely as 'lacking' monolinguals through comparison with native speakers of English (Matsuda, 2020). As far as L2 PA studies are concerned, this may call for revisiting the construct in light of the more globalized use of English by fellow L2 learners and speakers.

The Participants’ Pronunciation Anxiety: Interactions with Study Duration and Gender

Using a chi-square test of independence, this study found that the participants' duration of their study at the English language education department barely had any influence on their PA. This finding was rather surprising considering that previous studies suggested L2 exposure was reported to be associated with a lower level of anxiety (Alimorad & Adib, 2022; Baran-Lucarz & Lee, 2021; Thompson & Lee, 2014). These studies reported that experiences of living abroad led to lower levels of PA. This particular finding of the present study was also different from a finding of one of the few studies specifically contemplating the possible influence of exposure to English in academic contexts on PA (Tekten, 2020). A study in Türkiye by Tekten (2020) suggested that learners who studied English for a shorter period tended to be more anxious, indicating that study duration indeed made a difference in PA. It is speculated that the learner participants in the present study had high expectations for themselves and there may be discrepancies between the level of pronunciation mastery they wished to achieve and their current mastery of pronunciation. Another angle to examine the issue was that the English education curricula may have insufficiently facilitated learners to develop their pronunciation confidence over the years of their study. This could be from insufficient pronunciation practices, little use of English in classes by teachers, excessive use of first languages, and many other relevant factors. However, further investigations involving learners, teachers, and curriculum developers are necessary to mitigate unfounded conjectures about the issue.

Furthermore, using an independent sample t-test, this study found a statistically significant difference between female learners’ PA and that of their male counterparts. Female learners were significantly more anxious than male learners. This particular finding was in line with a finding of a study involving 596 L2 learners in Türkiye by Tekten (2020) who also found that female learners had a higher level of PA than their male counterparts. Though not in PA literature, many studies on speaking anxiety also reported that female learners tended to be more anxious in language learning than their male counterparts (Arnaiz & Guillén, 2012; Dewaele & MacIntyre, 2014; Mohtasham & Farnia, 2017). It is also important to see the influence of learners’ gender not merely as that of a neurobiological factor but rather as that of socio-psychological and cultural factors. For example, in a society where gender roles are relatively equal, learners’ gender may have little influence. In the case of the present study where the society still holds patriarchal beliefs, females may have more pressure to perform well and to avoid mistakes. That being said, the unbalanced number of male and female learners in the present study may, to a certain extent, affect the eventual finding. Hence, this particular finding should be interpreted with caution, and conducting further investigations involving a more balanced number of male and female learners could be necessary to better provide representative findings.

Conclusion

This study aimed to explore PA among L2 learners and examine whether their study duration and gender had an impact on it. Using a survey as a method of collecting data, this study found that learners generally experienced a moderate level of PA. This study further found that learners' duration of study in English education did not seem to influence their anxiety. Furthermore, female learners were found to be significantly more anxious in pronunciation than their male counterparts.

This study has several possible contributions. It can provide an overview, albeit perhaps exploratory, of Indonesian preservice English teachers' level of PA and its degree of interaction with their study duration and gender. This study can pave the way for further studies on PA in the Indonesian context. The findings indicate that the duration of the study, and therefore the degree of exposure to English, had little effect on PA may inform developers of English education curricula. They possibly need to examine whether such a finding had anything to do with the English education curricula encompassing many aspects such as learning materials and activities, teaching methods, assessment, and evaluation, as well as the goals and objectives. Next, the finding that male learners had lower PA levels suggests that male learners tend to be more confident and assertive in L2 learning and this can be leveraged in class. In L2 classes, teachers can organize learners into groups involving male and female learners. This mix can potentially create a situation where more confident learners could support more anxious learners encouraging them to participate in risk-taking behaviours during L2 learning. It may especially be the case in a population of learners dominated by females such as in typical English education departments.

Despite the possible contributions, this study has limitations that should be acknowledged. It employed a survey as the only method of data collection. In itself, it inherently brought the weakness that the data solely depended on the participants’ degree of honesty in responding to the questionnaire items. Furthermore, the number of female and male learners was not balanced because of the convenience sampling employed in this study. This may have, to a certain extent, affected the finding on the role of gender in influencing PA. Furthermore, as the participants in this study were from an English language department, it may be unpersuasive to say that the findings of this study can be generalized to a wider context. Nonetheless, replications may be found in a context sharing similar characteristics to that of this study.

Last but not least, possible directions for future studies can be suggested. The finding suggesting little influence of study duration on anxiety may warrant qualitative investigation involving learners, teachers, and possibly curriculum developers on what could be attributed to such a finding. It could be intriguing to contemplate the possible interactions between learners’ attitudes towards English as Lingua Franca (ELF) and their PA considering that ELF advocated varieties of English spoken by speakers from various first languages.

Declaration of Conflicting Interests

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this study.