Learners ’ Attitudes and Perspectives towards English Pronunciation Abilities with Different Religious Backgrounds in Thailand

A plethora of studies has examined EFL learners ’ attitudes towards and perceptions of English pronunciation, yet little has been discussed about the influence of religious backgrounds on one ’ s pronunciation abilities, especially in the Thai context. This study aims to extend the research area by studying Buddhist and Muslim EFL learners ’ attitudes and perspectives about their English pronunciation abilities in Thailand. Using a mixed-method design, it collected survey data from 60 undergraduate students (50% Buddhist, 50% Muslim) at a university in south Thailand. An English pronunciation test was conducted to gather data on the students ’ English pronunciation performances. The quantitative findings revealed that Buddhist and Muslim Thai EFL learners possessed moderate levels of attitudes towards their English pronunciation and perceived indirect influences of their religion on their English pronunciation. Although the learners ’ attitudes and perceptions were connected, they were not significant predictors of their actual English pronunciation. Religion and gender had no direct impact on the learners ’ attitudes; however, their frequent religious practices have led to the awareness that students from some religious backgrounds can have an advantage in pronouncing some


INTRODUCTION
While there have been an increasing number of studies looking into learners' attitudes and perceptions of English pronunciation (e.g., Huensch & Thompson, 2017;Sardegna et al., 2018;Tsunemoto & McDonough, 2020), little is known about how sociolinguistic variables (race, ethnicity, and religion) influence EFL learners' attitudes and perceptions of their English pronunciation abilities. According to Jarosz (2019), sociolinguistic factors cannot be overlooked since they might influence learners' acceptance and success in learning English pronunciation. Learners' foreign language intake and processing are socially mediated, and each individual's social identities influence their linguistic usage, choice, and growth in communicating in the target language (Tarone, 2007). Thus far, empirical studies have indicated that EFL learners from different countries possess different attitudes and perceptions towards English accents; for instance, Malaysian EFL learners highly appreciated their accented English, while the Japanese and Koreans disapproved of their own varieties of English and expressed their preference for native English pronunciation-such disparities were thought to occur as a result of the historical and political contexts of the learners' societies, which may have influenced the process of establishing EFL identity (Tokumoto & Shibata, 2011). Recently, these sorts of disparities in attitudes and perceptions about English pronunciation have been discovered among EFL learners living in the same country (Huang & Hashim, 2020), suggesting that sociolinguistic identities may affect EFL learners at the individual level.
Religion is one of sociolinguistic variables. Routine religious practices can cause speech variance and language use (Yaeger-Dror, 2014). Existing empirical research, nevertheless, has only looked at language differences among individuals within a rigorous religious framework and practice (Yaeger-Dror, 2015). As an example, Versteegh (2017) contrasted the setting and role of Christian Arabic with Christian Greek and Latin, which are thought to have served as in-group varieties in the Roman empire; the findings would presumably be useful to those who study religious circumstances. The current study, on the other hand, attempts to broaden the research field to include educational contexts, such as learning English as a foreign language, which might help us better understand Individual Differences (IDs) in foreign language learning and acquisition studies. The significance of the IDs study is in the possibilities it provides for the creation and adaptation of language learning instructions (Robinson, 2001). In a basic instance, learners come to an English classroom with individual differences but with the same goal: to learn and master it. Course design and teaching, as well as the instructor, may fail to meet and accommodate individual difficulties and requirements to be proficient English learners if they are unaware of the variables that underpin learners' IDs.
Religious beliefs are frequently established at a young age, and regular practice of these principles can have an impact on how learners acquire and adopt a foreign language. The religious origins of learners and their English learning strategies have been found to have a substantial relationship (Liyanage, 2004). Specifically, Liyanage et al. (2010) investigated the contributions of religions and ethnicity to ESL learners' language learning strategy choices in Sri Lankan and Japanese high schools and discovered that religious rather than ethnic identity influenced the students' English learning strategy choices. In another study, Shaaban and Ghaith (2003) discovered that learners' religion affected Lebanese EFL learners' linguistic attitudes towards foreign languages such as English and French. Nonetheless, Mohammadi and Izadpanah (2019) found a negative link between learners' sociocultural identity incorporating religion and their EFL learning in a recent study. Although religious backgrounds appear to have an impact on learners' English language learning, the magnitude of that impact appears to differ across contexts. Research on this topic is still limited in the existing body of literature; little is known about how students with two different religious backgrounds progress in their English learning and to what extent their religious beliefs and practices affect their English learning, which is the research gap the present study attempts to address.
Thus, this study is interested in examining how Muslim and Buddhist EFL students in Thailand feel and perceive the influence of religious backgrounds on their English pronunciation abilities and how these attitudes and perceptions influence their English pronunciation performances. Thailand is commonly recognized as 'the Land of Buddhism' although it is relatively diverse in terms of ethnicity, language, and religion. About 93.6 % of the Thai people acknowledge Theravada Buddhism as their religion, whereas Muslims embody about 4.6 % of the population and are recognized as the largest religious minority group (Knodel et al., 1999) and the rest are Christian, Hindu, and traditional religions. In Southern Thailand, most Thai Muslim EFL university students, commonly coming from three major provinces, including Yala, Pattani, and Narathiwat, are bilinguals as they can speak both Southern Thai and Patani Malay. These learners are mostly raised by Muslim parents who speak Malay as well as being able to read Al-Quran which is in Arabic. Malay and Arabic have some similar phonemes to English which do not exist in Thai.
This study addresses the following research questions: 1. What are the attitudes of Buddhist and Muslim Thai EFL learners towards their English pronunciation abilities? 2. How do Buddhist and Muslim Thai EFL learners perceive the influence of their religious backgrounds on their English pronunciation abilities? 3. How do their attitudes and perceptions of their English pronunciation abilities influence their pronunciation performances?

Thai EFL Learners' Difficulties in English Pronunciation
The difficulties that learners frequently encounter during their English learning process are one factor that might impact their attitudes toward and judgments of English pronunciation (Maretha & Waluyo, 2022;Tokumoto & Shibata, 2011;Tsunemoto & McDonough, 2021;Waluyo, 2020). Therefore, this section provides a review of Thai EFL learners' difficulties in English pronunciation in general, and for Buddhist and Muslim EFL students in particular. In his comprehensive introduction to English in Southeast Asia, Low (2019) classifies Thailand within the Expanding Circle together with Cambodia, Myanmar, Vietnam, Indonesia, and Laos, where English is primarily used as a foreign language. In the aspect of English pronunciation, Thai EFL learners are argued to face a great level of difficulty since their first language is greatly different in phonology, such as distinguishing and producing English vowels (Trakulkasemsuk, 2012), intonation, grammar (e.g., noun modifiers, reduplication, and cohesive devices), L2 to L1 transfer, translation, and lexical borrowing. Thai EFL learners experience major challenges in their English pronunciation, for instance, the letter 'h' in combination with another consonant, whether in the initial position, e.g., 'think, them, their, although, the' or final position, saying 'how mud' instead of 'how much' (Khamkhien, 2010).
Both the English alphabet and sounds that do not exist in Thai phonology commonly cause a major difficulty for Thai students in pronunciation. Wei and Zhou (2002), who explored the English pronunciation of Thai learners, elaborate that Thai EFL learners fundamentally have pronunciation problems with consonants and vowels. They summarize, "words with /ei/, usually pronounced as /e/; words with /r/, usually pronounced as /1/; words with /v/, usually pronounced as /f/; and words with /z/, usually pronounced at /s/ or voiceless" (p. 1). Thai has more vowels than English which tends to confuse Thai EFL learners, resulting in error pronunciation in terms of length which is recognizable in intonation problems such as in yes-no questions and wh-questions as well as in stress problems (Trakulkasemsuk, 2012). In addition, Sarmah et al. (2009), who studied the rhythm and vowel system of native speakers of Thai found that Thai people subconsciously use their L1 rhythmic characteristic, which is mostly high when speaking English, particularly when the spoken English words have, "…. Another cause of difficulty comes from the linguistic features of Thai Buddhist, which are likely to influence Thai EFL students who practice Buddhism. Thailand is one of the world's most heavily Buddhist countries, and the majority of Thai people are Theravada Buddhist. Beginning in India in the fifth century BCE, early texts were written in Sanskrit, Prakrit, and Middle-Aryan; however, the Buddhist sacred texts have been translated into many languages (Spolsky, 2003). The largest available collection is in Pali, which is the language of Theravada Buddhism. Nevertheless, Pali was originally a spoken language only without an alphabet of its own. As an alternative, Tipitaka was introduced by Sri Lankan monks as a written form, and it has been widely used in Thailand, Burma, etc. (Bullitt, 2005). Recently, Pali is also called Tipitaka, and it has been used interchangeably and contains understanding and appreciation of Buddha's teaching. Due to the limitation in Buddhist religion's alphabet and sounds, there seems to be an adjustment in which Thai standardized alphabets and sounds have been included in the teaching of Buddhism among Thai people. According to Ronakiat (2002), there are 21 Thai consonant phonemes and 24 English sounds as a result of Thai Buddhist linguistic features; whenever Thai EFL students encounter sounds that do not exist in Thai, they will have difficulty pronouncing the letter or word correctly using standard English sounds.
An early pronunciation study conducted at Thammasat University Language Institute by Kanokpermpoon (2007) involving Thai university students concluded that students face difficulties in these sounds /γ/ as in 'good', /ϖ/ as in 'van', /Τ/ as in 'thin', /Δ/ as 'rhythm', /ζ/ as in 'zoo', /Σ/ as in 'shoe', /Ζ/ as in 'genre', /τΣ/ as in 'church' and /δΖ/ as in 'George' because they do not occur in Thai phonology, for example 'ship', 'ball', 'focus', etc. In addition, syllable position in English, for instance, /φ/ as in 'philosophy' and /σ/ as in 'spy', and the interchangeability of /λ/ as in 'league' and /Ρ/ as in 'breath' in Thai were also challenging for Thai students. However, the author did not classify whether American or British sounds were compared in this study. Another interesting research conducted using the Thai inherited Sanskrit consonant chart was used to determine middle, high, and low sounds. This chart allows to determine the effect of syllable tone, which is useful when analyzing both statements and questions. Generally, Thai consonants consist of 34 of 35 Sanskrit equivalences plus 10 invented ones. Thai consonants are classified into three classes-namely, high, middle, and low consonants-which can affect the syllable tone when functioning as an initial sound (Karoonboonyanan, 1999). Besides, Thai is a tonal language. There are five tones in Thai syllables, for instance, Saman (middle), Ek (low), Tho (falling), Tri (high) and Chattawa (rising) (Karoonboonyanan, 1999).
Meanwhile, for Thai EFL students in the Southern part of Thailand, cultural and religious identities, as well as geographic location, may have an impact on their multilingualism. Students from the three southern border provinces, including Pattani, Yala and Narathiwat-the border between Thailand and Malaysia-can speak both Thai and Patani Malay. From a religious perspective, the majority of them are Muslim and can read and pronounce basic Arabic from the Al-Quran. It is assumed that as they are born in Thailand, speak Thai, and are considered as Thai residents, they may share the same difficulties in acquiring English pronunciation as other Thai students from across Thailand. Notwithstanding, their upbringing as Muslims and religious practice of reading the Al-Quran aloud may have affected their pronunciation skills, and these factors may have influenced their attitudes and perceptions towards a foreign language such as English, compared to other Thai students of different religions, which the present study seeks to illuminate. Basically, the English alphabet starts with the letter A and finishes with the letter Z. It is always written and read in the same order from left to right, while the Arabic alphabet consists of 28 letters, and it is written and read from right to left. Meanwhile, non-Muslim Thai EFL students may only read in the Thai language, which does not share these alphabets, has different tones, and is written in different styles.

Thai EFL Learners' Attitudes and Perceptions towards English Pronunciation
Compared to other Asian EFL learners such as Chinese, Japanese, and Malaysians, empirical research, particularly investigating Thai EFL learners' attitudes towards and perceptions of their English pronunciation abilities, is relatively few. One of the few studies is Snodin and Young's (2015) investigation of 251 Thai EFL learners' views and attitudes about native-speaker English variations. The study found that American English was the most preferred native accent by Thais, and it was considered the accepted model of English pronunciation compared to other accents encompassing British, Australian, and New Zealand. Meanwhile, Kalra and Thanavisuth (2018) observed that Thai EFL students were able to recognize non-native accents from Asian nations such as China, India, Vietnam, Japan, and Myanmar with ease. Indian accent was considered preferable, while Japanese and Burmese accents were negatively perceived; however, the participants still valued native-like accents more than non-native ones. Thais' fondness for native accents and desire to acquire them stems from a sense of intelligibility, English ownership, and identity concerns (Prakaianurat & Kangkun, 2018). Teachers' pronunciation and accent can sway Thai EFL students' views and attitudes regarding their own English pronunciation as well as an approved model (Pathomchaiwat, 2019;Phothongsunan, 2017). By gender, Female Thai EFL students showed more esteem for Thai English speakers than their male counterparts (McKenzie et al., 2016).
Among the four main English skills, Thai students' perceptions of English reading and writing skills are generally favorable, but they lack confidence in speaking due to a lack of knowledge, apprehensiveness, and a personal passive preference as listeners (Tananuraksakul, 2017). As for Thai university students in Southern Thailand (especially those from the Thailand-Malaysia border), the perception and ability in English speaking skills may vary as they grew up speaking both Thai and Malay. Students from the three southernmost provinces (Pattani, Yala, and Narathiwat), as well as some areas of Songkhla and Satun, speak both Patani Malay (PM) and Southern Thai (ST), whereas those without a Patani Malay background only speak Southern Thai (ST). This is noteworthy since both groups study English in public schools from elementary to university levels. Southern Thai students who speak Patani Malay (PM) have a cultural and geographical advantage while speaking English, as some Malay language sounds are comparable to English sounds, as shown in the acoustic characteristics of Malay and English beginning (p, b, t, d, k, g) as pronounced by Malay speakers (Shahidi & Aman, 2011). Besides, the majority of students from Southern Thailand are Muslims who can read the Al-Quran, which has certain Arabic consonant sounds that are extremely close to English sounds (s, z, t, d). A study by Phadung et al. (2012) revealed that students from this region are largely influenced by their Malay language and Muslim identity rather than by their Thai language. The teacher's accent and pronunciation can also significantly affect Thai learners' attitudes and perceptions of their own English pronunciation and the accepted model of English pronunciation (Kusuma & Waluyo, 2023;Pathomchaiwat, 2019;Phothongsunan, 2017;Waluyo & Rofiah, 2021).

Research Design
This study used a mixed-methods design, which enabled researchers to gain a better understanding of Buddhist and Muslim EFL learners' attitudes towards and perceptions of their English pronunciation abilities through survey data analyzed quantitatively and qualitatively. Creswell (1999) emphasizes that a mixed-method design is applicable to understanding social phenomena, in which insights from respondents are enhanced by descriptive and inferential statistics. In this study, as illustrated in Figure 1, the examinations involved descriptive statistics (means, standard deviation, percentages, etc.) and inferential statistics, including independent variables (t-test, bivariate correlation, and regression). In language acquisition research, quantitative research methods offer the opportunity to capture insights from bigger sample sizes and conduct more complex statistical analyses than may be covered by qualitative methods (Field, 2018;Henning, 1986). The variables of interest in this study involve learners' attitudes toward their English pronunciation and their scores on pronunciation tasks. This study also conducted a theme-based analysis for open-ended questions to delve into learners' perceptions of the influence of their religious backgrounds.

Context and Participants
Following the objectives of the study, undergraduate students at a university in the south of Thailand were recruited. A purposive sampling method was employed. To be participants, students had to meet three criteria: 1) pursuing an undergraduate degree, 2) being a Muslim or Buddhist, 3) studying a general English course during the time of the study. As a result, 60 participants (23.2% male; 76.7% female) were recruited in which 30 of them were Buddhist and 30 of them were Muslim. The students majored in Medical Technology, Physical Therapy, Accounting, Law, Political Science, Tourism, Thai studies, Engineering and Management. While ranging from 18 to 20 years old, the age mean was 18.92 years old. All these students were born in Thailand.

Likert-scale item survey
The questionnaire was divided into three sections. The first section gathered demographic information from students, such as religion, age, birthplace, gender, and major. The second section had 12 items that were adapted from Tokumoto and Shibata (2011). These items measured students' self-assessment of their English pronunciation which indicated their attitudes. The items basically consisted of cognitive components (6 items), affective components (2 items) and behavioral components (4 items). The responses ranged from Strongly Disagree (1) to Strongly Agree (5). In detail, the cognitive components are to measure students' self-assessment of 'accentedness' (1 item)-the extent that students think of the difference between their English pronunciation and native-speaker pronunciation, 'intelligibility' (2 items)-the ability to recognize and decompose phonological sequences into a word when speaking and 'acceptability' (3 items)-the extent that students accept their English pronunciation accent. Then, the affective components assessed students' self-assessment of their attachment to their English pronunciation (2 items), while the behavioral components (4 items) explored students' self-assessment of their behavioral intentions when speaking English.
The third part of the questionnaire consisted of three items that investigated students' perceptions of the influence of their religious backgrounds on their English pronunciation. The items included, 'In general, I think I can pronounce some English letters, sounds or words better than my friends with different religions', 'Because I practice my religion at home, I can pronounce some English letters, sounds or words better than my friends with different religions' and, 'I think that my religious background influences my English pronunciation'. The responses also ranged from Strongly Disagree (1) to Strongly Agree (5).

Qualitative survey
The researchers attempted to conduct interviews with the participants. After a few interviews, it was realized that the interview sections were not effective because the students seemed to misunderstand the questions and worried a little bit about their responses. The language barrier was obviously observed. Therefore, the interviews were canceled, and the researchers decided to employ an alternative method. To gain a deeper understanding of students' responses, this study created four open-ended questions in a survey form that asked the participants about their perceptions of the influence of their religious backgrounds. The questions consisted of: 1) Do you think that you can pronounce some English letters, sounds, or words better than your friends with different religions? Why? 2) Do you think your religious background influences your English pronunciation? Why? 3) What English words can you pronounce? 4) What English words can you not pronounce?
The questions were distributed to the participants through Google Forms after the first survey was finished.

English Pronunciation Test
The participants were given a pronunciation test that required them to pronounce English words in recorded videos. The videos were then posted in the class's Facebook group. In this task, the participants were asked to pronounce twelve English words, such as 'fan', 'van', 'leaf', 'five', 'sun', 'six', 'zebra', 'rose', 'three', 'tooth', 'father' and 'bath', and six English sentences including statements and questions: 1) The homework is on page six, exercise seven. 2) You are going to work in pairs. Language and Education, 10(2), 943-961, 2023 3) What's the matter? 4) What does this word mean in your language? 5) Can I help you? 6) Have you all finished?

| Studies in English
Although they seem simple, all of these words and sentences were considered to involve some difficulties in pronunciation for Thai EFL learners based on the literature review and the researchers' teaching experiences in Thailand. The examiners were two foreign English lecturers who finished their graduate studies in the U.S. and had more than two-year teaching experiences in Thailand. Students' pronunciation performances were assessed using an assessment rubric mapped onto the Common European Framework of Reference (CEFR) for Languages ranging from A1 to C2.

Data Analysis
After the data collection, data preparation and cleaning were conducted. Then, all the clean data was computed into SPSS. Descriptive data was interpreted in three categories: Mean scores 0-2.49 (Low), 2.5-3.49 (Moderate) and ≥ 3.5 (High). To answer the first and second research questions, students' responses on their attitudes towards their English pronunciation were analyzed by using descriptive statistics and the independent t-test. For the third research questions, bivariate and regression analyses were performed. Data from the Likert-scale surveys and English pronunciation tests were utilized at this stage.
Moreover, a thematic analysis was conducted on the data collected from the open-ended questionnaire. The analysis process followed Braun and Clarke (2006) as presented in Table 1. In the analysis, a deductive approach was employed in which themes and codes were based on the findings of previous studies, as explained in the literature review section. The focus was on investigating what the students thought about their pronunciation abilities compared to others with different religions and how the students perceived the influence of their religious backgrounds on their pronunciation abilities.  Braun & Clarke, 2006).

No. Phase
Description of the process 1 Getting to know the data Transcribing data, reading and re-reading the data, and writing down initial thoughts 2 Generating initial codes Coding important data characteristics in a systematic manner across the whole data collection and compiling data pertinent to each code 3 Searching for themes Organizing codes into possible topics and collecting all essential data for each prospective theme 4 Reviewing themes Creating a thematic 'map' of the analysis by checking if the themes function in connection to the coded extracts (Level 1) and the whole data set (Level 2) 5 Defining and naming themes Continuous analysis to fine-tune the specifics of each topic as well as the overall story the analysis conveys, resulting in unambiguous definitions and titles for each 6 Producing the report The last chance for analysis.
As explained previously, the measures for learners' attitudes consisted of three components, including cognitive, affective, and behavioral aspects. For cognitive components, both Buddhist (M=3.04, SD=.75) and Muslim (M=3.06, SD=.85) Thai EFL learners rated the extent that they think of the difference between their English pronunciation and native-speakers' pronunciation 'accentedness' at the moderate level; similarly, both of them respectively rated their ability to recognize and breakdown phonological sequences into a word when speaking 'intelligibility'

Students' perceptions toward the influence of their religious backgrounds
Generally, Buddhist (M=2.3, SD=.68) and Muslim (M=2.4, SD=.63) Thai EFL learners had a low level of perception with regards to the influence of their religion on their English pronunciation and no significant differences were observed across religion (t(58)=-.89, p=.38) and gender (t(58)=1.21, p=.23), but the effect size was insignificant 3) ⁄.656=.153). Compared to Buddhist, Muslim Thai EFL learners had a stronger belief that they can pronounce some English sounds, letters, or words better than their friends with different religions (M=2.6, SD=.77) because they practice their religion at home (M=2.6, SD=.72); however, they equally believed that their religious backgrounds had a low level of influence on their English pronunciation (M=2.0,SD=.98;M=2.1,SD=.85). The results are presented in Tables 4  and 5.

The influences of students' attitudes and perceptions of their English pronunciation abilities on their pronunciation performances
Buddhist and Muslim Thai EFL learners' attitudes were positively related to their perceptions (r=.411, p<.001), yet neither of these variables had any association with their English pronunciation performances. Significant and positive relationships were observed among the components of learners' attitudes: cognitive and affective components (r=.549, p<.001), cognitive and behavioral components (r=.540, p<.001), and affective and behavioral components (r=.681, p<.001). All three of these components were strongly associated with the learners' attitudes, indicating that these components appropriately measured learners' attitudes toward their English pronunciation. The strongest relationship was seen between learners' attitudes and cognitive components (r=.889, p<.001), followed by learners' attitudes and behavioral components (r=.839, p<.001), and learners' attitudes and affective components (r=.798, p<.001).

Qualitative Findings
The purpose of the open-ended questionnaire was to have a better grasp of what the students believed about the impact of their religious backgrounds on their pronunciation abilities. In total, there were 884 words collected from Buddhist students and 437 words from Muslim students for the first question while 840 words from Buddhist students and 407 words from Muslim students for the second question. Buddhist students were classified BS and Muslim students were coded MS in the findings' presentation, respectively, based on the researchers' allocated numbers.
The first analysis explored what the students thought about their pronunciation abilities compared to others with different religions. The analysis yielded conflicting results with three themes: 'Yes', 'No', and 'Unsure'. Subsequently, the second analysis delved into how the students perceived the influence of their religious backgrounds on their pronunciation abilities.

Yes, I can pronounce some English letters, sounds, or words better than my friends with different religions
Those who said 'Yes' cited: 1) personal experience with a friend whose religion was different who could pronounce some English words better, 2) personal belief that their friends could pronounce some words close to English accurately due to frequent religious practices, and 3) personal knowledge that Thai and English have different alphabets. Below are the sample responses.
(1) My friends of different faiths, I believe, talk more clearly. Because other religions' languages have a similar accent to English. (BS 1) (2) Some religious friends may be able to express themselves more clearly than I can. (BS 2) (3) There are variances in pronunciation as a result of religious and societal differences in practice.

(MS 1)
(4) Yes, because Islam teaches how to spell Arabic, and when you read the Alquran, you will see that Thai doesn't have the same alphabets. (MS 2)

No, I cannot pronounce some English letters, sounds, or words better than my friends with different religions
Meanwhile, students who responded 'No' thought that their ability to pronounce English depended on their level of practice, while others said that their religion had no bearing on their ability to pronounce English. These are shown by the replies below.

I am not sure if I can pronounce some English letters, sounds, or words better than my friends with different religions
On the other hand, some students were unsure, indicating that while some of their religions' alphabets were similar to those in English, their pronunciation abilities were still based on their daily practices in English; some students could benefit from their religious backgrounds, while others could underperform regardless of their religions.

Recognition of the influence of religion
Most of the students revealed their recognition of the influence of religion on their English pronunciation, as seen in these responses.

The influence of personal efforts
Nevertheless, some students perceived that personal efforts in practicing English pronunciation account for better ability in English pronunciation. Additionally, there was a mention of the effects of the environments where the students live in. The following excerpts confirm these personal efforts and the effects of students' environments (17) No, because the ability to speak English words correctly is based on practice rather than faith. (BS 9) Meanwhile, the third and fourth questions inquired about the words that the learners could pronounce, for example 'fan', 'van', 'six', 'love', 'sleep', 'thing', and 'diary' and could not pronounce, for example 'zebra', 'three', 'matter', 'eleven', and 'father', etc.

DISCUSSION
This study aimed to examine whether EFL learners from the same country but with different religions had different attitudes and perceptions toward their English pronunciation. It also delved into the associations and predictive roles that learners' attitudes and perceptions may have on English pronunciation performance. The first result of this study, answering the first research question, confirms that moderately positive attitudes were reported among Buddhist and Muslim Thai EFL learners, yet religion and gender did not play an influential role in their attitudes. Buddhist and Muslims had moderate levels of positive attitudes on their 'accentedness' (item 1), 'intelligibility' (items 2 and 3), and 'affective components' (items 7 and 8), but they did not fully have a higher level of 'attitudes' (item 4) towards 'acceptability' and 'behavioral components' (item 9 and 11). At this point, the results indicated that regardless of their religion or gender, Thai EFL learners did not think that their English pronunciation would be accepted in international business, they did not consider their English pronunciation to reflect a native-like accent, and they did not have the intention to keep their English pronunciation accent. Nonetheless, Buddhist EFL learners held a higher level of acceptance that their English pronunciation would be acceptable for the profession of English teachers than their counterparts. Thai EFL learners' preference towards native accents has been identified by previous studies (Kalra & Thanavisuth, 2018;Prakaianurat & Kangkun, 2018) and is reflected by both Buddhist and Muslim Thai EFL learners in this study. However, unlike the findings from previous studies (Chan, 2018;McKenzie et al., 2016), the first result of this study did not sustain that gender plays a role in learners' attitudes towards their English pronunciation.
The second result of this study, answering the second research question, suggests that Muslim Thai EFL learners had a stronger belief that they could pronounce some English sounds, letters, or words better than their friends with different religions due to their frequent religious practices at home. Nevertheless, neither Buddhists nor Muslims believed their religious backgrounds affected their English pronunciation. While previous studies have confirmed positive associations between religion and English learning strategy (Liyanage, 2004;Liyanage et al., 2010), the second result of this study adds to the knowledge of learners' awareness of the influence of their religious backgrounds on their English pronunciation. It partially supports the findings from Shaaban and Ghaith (2003) but does not follow the findings from Mohammadi and Izadpanah (2019), who noted a negative relationship between learners' sociocultural identity involving religion and their EFL learning. For a special case of Thai EFL learners, Buripakdi (2012) discovered, "the participants' (Thai EFL learners) conceptualization of their English illustrated that English use in Thailand was situated in a hierarchy of language and was deeply embedded in internal colonization" (p. 1).
Then, the last result of this study, answering the third research question, validates that Buddhist and Muslim Thai EFL learners' attitudes were closely associated with their perceptions of their English pronunciation. Yet, these two variables did not have predictive roles on the learners' English performances. Furthermore, Buddhist and Muslim Thai EFL learners seemingly shared the same thoughts regarding the words they could and could not pronounce correctly. They believed they could pronounce words like 'fan', 'van', 'six', 'love', 'sleep', 'thing', and 'diary' and felt they could not correctly pronounce words such as 'zebra', 'three', 'matter', 'eleven', 'father', etc. However, despite such reported beliefs of the words they could pronounce correctly, the report from the teacher assessing the learners' pronunciation videos noted that some of the learners still could not pronounce the words like 'van' and 'six' correctly. At this point, this may be a sign of a discrepancy of knowledge between learners' beliefs of their ability in English pronunciation and the acceptability of their ability in English pronunciation, which this study would suggest for future studies to explore. The result of the English pronunciation videos also indicates that although both learners generally shared difficulties in English pronunciation, as pointed out by Wei and Zhou (2002), Sarmah et al. (2009), andTrakulkasemsuk (2012). Muslim Thai EFL learners performed better and did not seem to be having a noticeable difficulty in pronouncing most of the assigned English words. The close connection between Arabic and English pronunciation might have enhanced Muslim Thai EFL learners' English pronunciation, making their tongue more flexible than their counterparts who only speak Thai.

CONCLUSION
Based on the results of the study, it is concluded that in the case of Buddhist and Muslim Thai EFL learners, religion and gender did not have significant direct influences on their attitudes toward their pronunciation. However, their frequent religious practices had grown the awareness that students from some religious backgrounds could have an advantage in pronouncing some English sounds, letters, or words over their friends with a different religion. The difference seemed to be not so much the difference in religious beliefs as the different languages used in their religious practice. The findings of this study inform English teachers that Buddhist and Muslim Thai EFL learners basically did not hold negative attitudes and perceptions towards their English pronunciation. EFL teachers should prioritize teaching accepted pronunciation accents as their own pronunciation and accent may impact Thai EFL learners' attitudes towards English pronunciation and the accepted model.
The findings of this study also indicate that there are more aspects to be explored with regards to the influences and roles of religious backgrounds on learners' English pronunciation. Since this study does not provide deep qualitative findings, future studies utilizing qualitative research methods are recommended. It is acknowledged that the findings of this study are limited to Buddhist and Muslim EFL learners in Thailand; studies exploring different religions or the same religions in other contexts may or may not attain the same results, yet the findings of this study would be of usefulness.