Chinese University Students’ Beliefs about English Language Learning and Self-efficacy

Learners’ beliefs on language learning and perceived self-efficacy are important to the success of their second/foreign language (SL/FL) learning. To reveal the general profiles of and relationship between Chinese students’ beliefs about English learning and self-efficacy, the present study examined beliefs about English learning and self-efficacy held by Chinese university EFL (English as a FL) learners at differing English proficiency levels. A total of 1,698 students from a top university in Beijing answered a battery of questionnaires. The re - sults revealed a general overview of the students’ beliefs about the nature of language learning and the roles of teachers, feedback and learning strategies, and self-efficacy. Another major finding was that participants at different English proficiency levels differed significantly from one another in beliefs about language learning and self-efficacy.


Introduction
Generally concerned with beliefs about the nature and process of language learning, including perceptions of mistakes, self-efficacy and the role of feedback, language learning beliefs are often measured by questionnaires and interviews in specific contexts.It is the same with self-efficacy beliefs which refer to beliefs in one's abilities to do something such as learning a second/ foreign language (SL/FL).High self-efficacy seems to motivate students to study harder and achieve more (Woodrow, 2011).Since what learners believe and think what they can do often determines what efforts they will make to study a SL/FL (Gao, 2016;Horwitz, 1988;Wenden, 1991), it is necessary to research what beliefs are often held by language learners.Although diverse belief patterns have been revealed in learners in different contexts (Al-Roomy, 2015;Daif-Allah, 2012;Yang, 1999), more research is needed considering the large number of learners and variety of contexts.This is especially so in China which houses a large foreign language learning population, while studies on language learning beliefs and self-efficacy are far from adequate.Moreover, few studies have compared beliefs about language learning and self-efficacy between students at different proficiency levels.Hence, the present study intended to explore beliefs about English learning and self-efficacy in Chinese university EFL (English as a FL) students at differing English proficiency levels, hoping to reveal the general profiles of and relationship between their beliefs about English learning and self-efficacy.

Literature Review
Learning beliefs are notions, myths or misconceptions in contrast with "truth" (Alanen, 2003), or subjective and individual understandings that are held to be true (Alexander & Dochy, cited in Wenden, 1998), or the way we think we learn (Riley, 1994).Language learning beliefs are beliefs about how to learn a second language that emerge through one's own experience and the influence of others (Horwitz, 1987).They are generally concerned with beliefs about the nature and process of language learning, including perceptions of mistakes, the role of the teacher, and the role of feedback (Dai & Wang, 2002;Song, 2013;Wenden, 1991;Zhong, 2012).To tap into the profile of students' language learning beliefs, Horwitz (1987) developed the Beliefs About Language Learning Inventory (BALLI), which consists of five parts: foreign language aptitude, the difficulty of language learning, the nature of language learning, learning and communication strategies, motivation and expectations.Soon after, many other questionnaires have been developed based on the BALLI to fit different contexts, such as the Language Learner Factors Questionnaire designed by Wen and Johnson (1997), the Belief Survey designed by Sakui and Gaies (1999), the Language Learning Beliefs Questionnaire developed by Liu and Dai (2003), and the Language Learners' Beliefs Scale (LLBS) designed by Birjandi and Mohammadi (2014).
Studies targeting various learners have revealed differing belief patterns among students of different cultures.For example, Korean students are high in their motivation to learn English, place spoken English in a higher place than formal English, while they are not confident enough to speak English with others (Park, 1995).Taiwanese students report to be positive in self-efficacy about learning English, foreign language aptitude, the communicative function of English and repetitive practice (Yang, 1999).Saudi students seem to be more realistic in beliefs about learning and communication strategies and foreign language aptitude (Al-Roomy, 2015;Daif-Allah, 2012).Mainland Chinese students generally have high beliefs in self-management and foreign language aptitude, great instrumental motivation, and frequent use of learning strategies, but are rather low in self-efficacy, they also rely heavily on their mother tongue to learn English and stress functions more than forms of language learning (Liu & Dai, 2003).Kern (1995) examined the beliefs of 180 Berkeley freshmen from various ethnic groups, including Asians (40%), Caucasians (30%), Hispanics (17%), African-Americans (6%), and American Indians (1.2%), and found that they placed more focus on grammar and were more cautious of making mistakes after having completed 15 weeks of first-year level French.Zhang and Cui (2010) explored beliefs held by distance English language learners in a highly prestigious Chinese university and the differences in beliefs between beginner distance learners and those with more distance education experience.Analyses of 90 questionnaires revealed that most participants regarded insufficient communication with teachers and peer students as the dominant difficulty in distance learning, and that learners with more distance learning experience believed more strongly in the benefits of an autonomous approach to learning.Genç, Kuluşaklı, and Aydın (2016) examined perceived self-efficacy and beliefs on English language learning in 210 Turkish EFL undergraduate English majors.The findings showed that the students had medium scores in their English self-efficacy and strongly believed that motivation factors had a great role on their learning process.The study also showed that the student's beliefs about language learning were affected by their English self-efficacy.
The importance of self-efficacy in SL/FL learning has been well recognized (e.g., Bandura, 1997Bandura, , 2006;;Raoofi, Tan, & Chan, 2012;Sağlam & Arslan, 2018).Grounded within the framework of social cognitive theory of human behavior (Bandura, 1997), self-efficacy beliefs refer to "beliefs in one's capabilities to organize and execute the courses of action required to produce given attainments" (Bandura, 1997, p. 3).And the efficacy belief system is "not a global trait but a differentiated set of self-beliefs linked to distinct realms of functioning" (Bandura, 2006, p. 307).For example, a learner may believe that they will get a high score in the upcoming English test, another learner may believe that they are better than others in the language class.Thus, " […] there is no all-purpose measure of perceived self-efficacy" (Bandura, 2006, p. 307).Understandably, different self-reported surveys and questions have been used to measure self-efficacy related to different aspects of SL/FL learning in specific contexts, such as general self-efficacy, academic self-efficacy, and web-based learning self-efficacy (Alegre, 2014;Baleghizadeh & Masoun, 2013;Bandura, 2006;Doménech-Betoret, Abellán-Roselló, & Gómez-Artiga, 2017;Genç, Kuluşaklı, & Aydın, 2016;Kuo, Tsai, & Wang, 2021;Ozer & Akçayoğlu, 2021).For example, Sağlam and Arslan (2018) developed a 29-item English Language Skills Self-Efficacy Scale to measure students' self-efficacy beliefs in four basic language skills.Baleghizadeh and Masoun (2013) divided 57 Iranian learners in an English-language institute into the experimental and control groups to investigate the continuous influence of self-assessment on their self-efficacy in their current English language class.The participants answered the English as a Foreign Language Self-efficacy Questionnaire.The results showed that self-assessment significantly improved the experimental group's self-efficacy.Kuo et al. (2021) administered the selfdeveloped web-based learning self-efficacy questionnaire to 608 university students from Taiwan.The results showed that the components of web-based learning self-efficacy led to different types of engagement: General Internetbased learning self-efficacy contributed to behavioral and emotional engagement, whereas functional Internet-based learning self-efficacy contributed to emotional and cognitive engagement.
These studies, as well as others, generally reveal that self-efficacy helps improve SL/FL learning and is related to many other variables such as academic success, strategy use, and motivation (e.g., Blumenthal, 2014;Bong, 2002;Doménech-Betoret, Abellán-Roselló, & Gómez-Artiga, 2017;Kao et al., 2020;Mills, 2009;Ozer & Akçayoğlu, 2021;Pan & Chen, 2021;Wong, 2005;Woodrow, 2011;Zhan et al., 2021).For example, Woodrow (2011) found that students with high self-efficacy tend to study harder and are more intrinsically motivated.Mills, Pajares, and Herron (2007) found that intermediatelevel French learners' confidence beliefs in their ability to attain a particular grade were highly related to their academic success.Mills's (2009) study of 47 university students of French showed that project-based learning significantly improved students' self-efficacy in the areas of communication, cultures, connections, comparisons, and communities. Heidari, Izadi, and Ahmadian (2012) explored the relationship between 50 Iranian EFL juniors' self-efficacy beliefs and their employed vocabulary learning strategies.The results revealed that students had fairly high level of self-efficacy and that self-efficacy was significantly positively correlated with their use of vocabulary learning strategies.The results indicated that learners' self-efficacy beliefs had great impact on successful learning experiences and achievements.Doménech-Betoret et al. (2017) administered the self-efficacy and expectancy-value beliefs questionnaires to 797 Spanish secondary education students.Structural equation modeling analyses revealed that students' expectancy-value beliefs mediated the relationship between students' academic self-efficacy and achievement/ satisfaction.The findings were partially consistent with those in Zhan et al. (2021), which explored the impact of self-efficacy and learning motives  Ozer and Akçayoğlu's (2021) study.Analyses of the data revealed a medium negative correlation between foreign language self-efficacy and anxiety in addition to a small positive correlation between foreign language self-efficacy and self-regulated learning.Pan and Chen (2021) recruited 197 first-year Chinese university students to investigate relations among teacher supports, technology acceptance, technological self-efficacy, and self-directed language learning.The results showed that technological self-efficacy mediated the relationship between teachers' affective supports and students' self-directed language learning as well as the relationship between teachers' behavior supports and students' self-directed language learning.

Research Questions
As reviewed above, most current literature shows that beliefs about language learning and self-efficacy affect how students behave to a great extent and that these beliefs interact with other variables during the process of SL/FL learning.Although the current literature has revealed interesting findings about learners' beliefs of language learning and self-efficacy, more research is called for given the large number of language learners, the huge variety of learning contexts, the complicated nature of language learning and the roles of learning beliefs and efficacy in language learning.It is the same in China with a large number of foreign language learners.Moreover, limited studies can be found which have examined beliefs about language learning and self-efficacy between students at various proficiency levels.Thus, to reveal the general profiles of and relationship between Chinese students' beliefs about English learning and self-efficacy, the present study aimed to examine beliefs about English learning and self-efficacy in Chinese university EFL students at differing English proficiency levels.The following questions were of particular interest: 1.What are the profiles of Chinese university students' language learning beliefs and self-efficacy? 2. Are there any belief and self-efficacy differences among students at varying English proficiency levels?3. How are the students' learning beliefs related to self-efficacy?

The Present Study
This study was conducted in a top state-owned university in Beijing, which often accepted outstanding high school graduates in each province of China.In this university, all first-year non-English majors had to take the standard English placement test prior to the start of formal classroom teaching when they first registered in the university.The test consisted of listening, reading, and writing, the results of which put the testees into bands 1-3 groups.Students with lower scores went into lower band groups, which signified lower proficiency in English.They then selected English language courses accordingly.

Participants
A total of 1,698 (776 male, 918 female, and four missing genders) students participated in the study, of whom 1,171 were freshmen, 481 sophomores, 41 juniors and four seniors.833 (324 male, 505 female and four missing) were in the band 1 group, 444 (245 male and 199 female) and 421 (207 male and 214 female) were in bands 2 and 3 groups respectively.With an average age of 19 (SD = 1.10) and an age range of 16 to 24, the participants came from various disciplines such as architecture, civil engineering, medicine, business management, and psychology.

Instruments
The survey consisted of three parts: background information questionnaire, English Learning Belief Inventory, and Self-Efficacy Belief Questionnaire.The background information questionnaire covered such items as gender, age, discipline, and English band level.The 29-item English Learning Belief Inventory (ELBI) (Cronbach alpha a = .749)and the 14-item Self-Efficacy Belief Questionnaire (SEBQ) (a = .644)were adopted from the questionnaire used in Zhang and Cui (2010).The ELBI asked students to report their beliefs about the nature of language learning, the role of the teacher, the role of feedback and language learning strategies.The SEBQ had three parts: The 3-item Self-Efficacy Questionnaire invited students to comment on their language learning abilities, the 8-item Self-evaluation Questionnaire asked students to assess the importance of feedback, opportunities, practice, teachers' help and personal effort in language learning, and the last three items asked students to judge what type of learners they were.The ELBI used a five-point Likert scale with 1 referring to 'strongly agree' and 5 'strongly disagree,' and the SEBQ employed 'no' (1) and 'yes' (2) choices for each item.

Data Collection Procedure and Analyses
The survey was translated into English and back-checked by a teacher researcher with Ph.D. in Applied Linguistics.Then the survey, together with a consent form, was administered to students by their course teachers in class and was answered in about 15 minutes during class break.The collected data was analyzed via SPSS 20.Means and standard deviations of ELBI items as well as frequency and percentages of SEBQ items were computed to explore general patterns of students' language learning beliefs and self efficacy.Posthoc ANOVA (Duncan) was then run to examine differences in language learning beliefs and self-efficacy between students in different band groups.Finally, correlation analyses were conducted to reveal relations between language learning beliefs and self-efficacy.

Beliefs about the Nature of Language Learning
As shown in Table 1, the participants scored more than 3 on items 1-6, with a score of nearly 4 (mean = 3.83 ~ 4.00) on items 1-4.Alternatively, 70% to 80% of the participants (strongly) agreed that making mistakes is a natural part of learning, that different people learn languages in different ways, that language learning takes a long time, that it is easier for children than adults to learn a foreign language, and that I need to know language rules before I can communicate in English; and around 60% believed that women are better than men at learning foreign languages.Meanwhile, around 40% of the participants endorsed item 11 that making mistakes is harmful in language learning (mean = 2.10), further confirming their agreement with item 1 that making mistakes is a natural part of learning.In addition, a score of 2.996 on item 10 suggested that around half of the participants agreed that it is possible to learn a language in a short time.The participants scored 2.42 to 2.76 on items 7-9, indicating that around one third of the participants assumed that they can communicate in English without knowing the rules, that learning a foreign language is mostly a matter of learning a lot of grammar rules, and that learning a foreign language is mostly a matter of translating from Chinese.
A similar score pattern for items 1 to 11 was observed for all the three bands students.Moreover, as reported in Table 1, band 1 students scored the lowest on items 3-4, 6 and 10 yet the highest on items 7 and 11; band 2 students scored the highest on items 1-6 and 10 but the lowest on items 8-9 and 11; band 3 students scored the lowest on items 1-2, 5 and 7 but the highest on item 8. ANOVA analyses revealed significant differences in items 1 (F = 6.18, p = .002), 2 (F = 4.02, p = .018),5 (F = 4.02, p = .018),8 (F = 13.22,p = .000),9 (F = 11.12,p = .000),and 11 (F = 3.71, p = .025)(Table 1).Alternatively, band 2 students agreed significantly more strongly than their bands 1 and 3 peers that making mistakes is a natural part of learning but significantly less strongly than the latter two groups that learning a foreign language is mostly a matter of translating from Chinese, and significantly more strongly than their band 3 peers that different people learn languages in different ways; band 3 students were significantly less in line with the statement that women are better than men at learning foreign languages than their bands 1 and 2 peers; band 1 students believed significantly more strongly than their band 2 peers that making mistakes is harmful in language learning; and the students of three bands differed significantly from one another in their agreement with item 8 that learning a foreign language is mostly a matter of learning a lot of grammar rules.

Role of Teachers
The participants scored 2.98 to 3.78 on items (12-18) on the role of teachers (Table 1).Namely, around 80% of the participants believed that the role of teachers was to help them learn effectively or to offer to help them.Around half of them acknowledged that the role of teachers was to tell them what to do, what their difficulties were, and what progress they had made, to set their learning goals and give them regular tests.
A similar score pattern was observed for students of all three bands.Table 1 shows that bands 1-3 students had nearly similar scores on items 12-18.Results of ANOVA analyses (Table 1) revealed significant differences in items 15 (F = 2.62, p = .073)and 18 (F = 2.55, p = .078).Namely, band 2 students believed significantly less strongly than their band 3 peers that the role of teachers was to tell them what their difficulties were; and band 1 students were significantly more in line with the role of teachers being to set learning goals than their band 2 peers.

Role of Feedback
The participants scored 2.97 to 3.33 on items 19 to 21, indicating that more than half of them acknowledged: having my work evaluated by others is helpful and I know best how well I am learning, and that around half of them considered themselves good at language learning.
A similar score pattern was observed for students of all three bands.Meanwhile, band 1 students scored the lowest on items 19-20; band 2 students scored the highest on items 19-21; and band 3 students scored the lowest on item 21.Results of ANOVA showed that significant differences occurred in items 20 (F = 10.87,p = .000)and 21 (F = 2.57, p = .077).Namely, band 2 students agreed significantly more strongly that they were good at language learning than their bands 2 and 3 counterparts and that they knew best how well they were learning than their band 3 peers.

Use of Strategies
The participants scored 3.01 to 3.69 on all items (22-29) on the use of strategies except for item 26 (mean = 2.57).This meant that around 50% to 70% of the students believed that they could explain why they needed English, knew how to set their own learning goals, identify their strengths and weaknesses, find their own ways of practicing, plan their learning, measure their progress and check their work for mistakes.Around 40% of them thought they would go up to practice speaking English when hearing someone speaking English.
A similar score pattern was observed for students of all the three bands.Band 1 students scored the lowest on items 22, 24-25 and 29 yet the highest on item 26; band 2 students scored the highest on all the items; and band 3 students scored the lowest on items 23 and 26-28.Results of ANOVA analyses revealed significant differences in items 24 (F = 6.10, p = .002),25 (F = 6.20, p = .002),27 (F = 2.93, p = .054)and 29 (F = 5.51, p = .004).Alternatively, band 2 students believed significantly more strongly than their bands 1 and 3 counterparts that they knew how to identify their strengths and weaknesses, find their own ways of practicing and check their work for mistakes.They also believed significantly more strongly than their band 3 peers that they knew how to plan their learning.

Self-efficacy
As shown in Table 2, more than 68.9% of the participants acknowledged that they had the ability to learn a language successfully and to get the score they desired in their next English test, and that they knew how to find an effective way to learn English.A similar pattern existed for students of all three bands.Band 2 students had the highest percentages of agreement, followed by bands 3 and 1 groups respectively.Results of ANOVA analyses revealed significant difference in items 31 (F = 7.31, p = .001)and 32 (F = 2.70, p = .067).Namely, band 2 students believed significantly more strongly than their bands 1 and 3 peers that they would get the score they desired in their next English test, and significantly more strongly than their band 3 counterparts that they knew how to find an effective way to learn English.

Self-evaluation
As many as 67.1%, 54.6%, and 41.6% of the participants believed that feedback they gave themselves, feedback from teachers, and feedback from other people helped them most, respectively.62.7%, 81.9%, 80.3%, and 83.1% of the students held the view that feedback, opportunities to use the language, practice and their own effort played the most important role in successful language learning, respectively.Only 39.8% believed that the language teacher played the most important role.Similar patterns existed for students of all three bands.Scoring higher or lower in different items, the three band groups differed significantly from one another in all self-evaluation items except for items 37-38, with F values ranging from 2.82 (p = .06)to 11.71 (p = .000).Alternatively, band 2 students believed significantly more strongly than their bands 1 and 3 peers that the feedback given by themselves helped them most, yet significantly less strongly than the other two groups that the feedback from the teacher or other people helped them most and that the language teacher played the most important role in successful language learning.In addition, band 1 students believed significantly more strongly than the band 2 group that feedback played the most important role in successful language learning and significantly more strongly than their band 3 peers that their own effort played the most important role in successful language learning.TAPSLA.12797p. 10/25 M. Liu

Type of Learner
62.1%, 59.1%, and 76.5% of the participants considered themselves to be learners who liked to learn with other people, a teacher and on their own, respectively.A similar pattern was observed for students of all the three bands.With similar scores on items 41-43, band 3 students significantly differed from band 2 students in item 42 (F = 2.03, p = .131)and band 1 students in item 43 (F = 2.65, p = .071).Namely, band 3 students rated themselves as learners who liked to learn with a teacher significantly more strongly than their band 2 peers and themselves as learners who liked to learn on their own significantly less strongly than their band 1 peers.
Furthermore, 182, 131, 103, and 94 significant coefficients existed for the whole sample, band 1, band 2, and band 3 samples, respectively.It seemed that the larger the sample size, the more significant correlations there were between English learning belief statements and self-efficacy items.

Findings and Summary
Statistical analyses showed that both the ELBI and SEBQ were fairly reliable in the present study, which also found that most of the ELBI items were significantly related to SEBQ items for both the whole sample and the three band groups.In particular, the SEBQ items were generally significantly correlated with belief statements about strategy use, as found in similar studies (Genç et al., 2016;Heidari et al., 2012;Raoofi et al., 2012;Zhan et al., 2021).In addition, as shown in Table 3, the larger the sample size, the more significant correlations there were between ELBI and SEBQ items.These findings might be largely attributed to the large size of each specific sample.Thus, the results need to be confirmed with samples of varying sizes.

General Profiles of and Differences in Students' Language Learning Beliefs
The findings of this study presented a general overview of English learning beliefs and self-efficacy held by Chinese university EFL learners.The majority of the students, in spite of their English proficiency level, demonstrated similar beliefs and perceptions as learners of foreign languages.For example, to most of them, making mistakes was a natural part of learning; different people learned languages in different ways; learning a language took time; feedback, personal efforts and others' help played important roles in learning foreign languages successfully; people should employ various strategies accordingly to learn foreign languages well.All these beliefs were generally in line with language learning theories.For instance, according to the behaviorism theory (McLeod, 2017;Skinner, 1953) and the input theory (Krashen, 1982(Krashen, , 1985(Krashen, , 2008)), practice, interaction, reinforcement, mistakes, and feedback are crucial to the acquisition of a foreign language, especially to adult learners.Moreover, according to the input theory, knowing language rules helps learners monitor their output, which justifies the belief held by many learners that they could not communicate in English without knowing the rules and that learning a foreign language was mostly a matter of learning a lot of grammar rules.Since learning a foreign language takes time and is more than learning grammar rules and vocabulary, learners may experience various emotions such as anxiety and enjoyment, encounter numerous challenges and thus be under varying degrees of pressure during the learning process, as discussed in Horwitz, Horwitz, and Cope (1986) and Dewaele and MacIntyre (2014).All these findings attest to TAPSLA.12797p. 12/25 M. Liu the importance of learner-internal and external factors in SL/FL learning and acquisition (Skehan, 1989).
In addition, quite many students expected much of their teachers who should help them learn (more) effectively, tell them how to set goals, plan their learning and provide feedback on their progress.These beliefs might be influenced by traditional Chinese culture which regards teachers as authority figures of knowledge who have supreme power to educate and plan for their students (Cortazzi & Jin, 1996;Scollon & Scollon, 2000).
Moreover, many participants agreed that learning a FL was mostly a matter of translating from Chinese.This indicated that they were aware of other factors involved in SL/FL learning in addition to grammar and vocabulary, as discussed in Horwitz et al. (1986) and Skehan (1989) and evidenced in empirical research (e.g., Botes, Dewaele, & Greiff, 2020;Gardner, 1985;Liu, 2020).
Concurrently, the present study revealed that most participants knew their motives to study English as well as their strengths and weaknesses, and knew how to plan and assess their learning, implying that they were largely autonomous and independent learners, as found in Zhang and Cui (2010).This might be attributed to the fact that the participants were from a top university in China, which generally accepted outstanding senior high school graduates in each province of the country.These students might have got accustomed to learning on their own to become exceptionally good.
The present study also revealed significant differences in certain beliefs between students in different band groups, especially between band 2 students and their bands 1 and 3 peers.For example, band 2 students agreed significantly more strongly than their bands 1 and 3 peers that making mistakes was a natural part of learning but significantly less strongly than the latter two groups that learning a foreign language was mostly a matter of translating from Chinese.They believed significantly less strongly than their band 3 peers that the role of teachers was to tell them what their difficulties were, and significantly less strongly than band 1 students that the role of teachers was to set learning goals.They also reported to be more significantly independent in learning foreign languages.It was hard to explain these differences, which needs further research.
In addition, band 3 students were significantly less strongly in line with the statement that women are better than men at learning foreign languages than their bands 1 and 2 peers.This might be because the band 3 group had the lowest male to female ratio.Moreover, band 1 students believed significantly more strongly than their band 2 peers that making mistakes was harmful in language learning.Furthermore, students of three bands differed significantly from one another in their agreement with the belief that learning a foreign language is mostly a matter of learning a lot of grammar rules, indicating that Chinese University Students' Beliefs about English Language… TAPSLA.12797p. 13/25 the participants had mixed attitudes towards grammar learning.The following two reasons might have contributed to this: (a) English language education has long focused on grammar and continues to be so even now in many places in China, and (b) various teaching approaches have been introduced into classroom teaching and learning, especially communicative teaching and learning, which emphasizes the importance of effective communication.Nevertheless, all these explanations remain assumptions, which need to be confirmed in future research.

General Profiles of and Differences in Students' Self-efficacy Beliefs
As presented above, the participants, in spite of their English proficiency level, acknowledged that they had the ability to learn a language successfully and knew how to learn English effectively.They were also aware of the importance of feedback, opportunities to use the language, practice, their own efforts and the language teacher, and knew what kind of learners they were.All these findings indicated that they had great self-efficacy beliefs, as found in similar studies (Blumenthal, 2014;Mills, 2009;Pan & Chen, 2021;Yang, 1999;Zhan et al., 2021;Zhang & Cui, 2010).These findings not only further supported the participants' beliefs about language learning, but further pinpointed that they were confident and autonomous learners, like their peers in Zhang and Cui (2010).
Likewise, the present study also revealed significant differences in certain self-efficacy beliefs between students in different band groups, especially between band 2 students and their bands 1 and 3 peers.For example, band 2 students believed in their ability to achieve their goals and the supreme importance of their own feedback significantly more strongly than their bands 1 and 3 peers, and in their ability to learn English effectively significantly more strongly than their band 3 counterparts.Band 1 students believed in the supreme importance of feedback in successful language learning significantly more strongly than the band 2 group and in the supreme role of personal efforts in successful language learning significantly more strongly than their band 3 counterparts.These findings seem to contradict with the belief that self-efficacy beliefs positively affect academic success (Alegre, 2014;Mills, 2009;Woodrow, 2011).Thus, they need to be further researched.

Limitations
The present large-scale study provided an overview of Chinese university EFL learners' beliefs and self-efficacy about English learning and revealed the TAPSLA.12797p. 14/25 M. Liu general relationship between the two, thus contributing to a better understanding of the two issues.Even so, some limitations existed in the study.The biggest limitation was that the study solely relied on survey data.If complemented with qualitative data, the study might be able to better account for the significant differences in certain English learning beliefs and self-efficacy between students at varying English proficiency levels.In addition, the participants of the present study solely came from the same highly prestigious university, who had probably formed the habit of being confident and autonomous learners earlier in secondary schools.Students with other backgrounds might display different belief and self-efficacy patterns, which will be examined in future research.Moreover, as discussed in Amuzie and Winke (2009), Zhang and Cui (2010) and Ozer and Akçayoğlu (2021), language learning beliefs and self-efficacy are dynamic and change at different stages of the learning process.The present study only captured students' English learning beliefs and self-efficacy at a certain moment.Future research should be directed to examine whether, how, and why their beliefs and self-efficacy about language learning change over time.

Conclusions
The present study examined English learning beliefs and self-efficacy held by Chinese university EFL learners.The results revealed a general overview of the students' beliefs about the nature of language learning and the roles of teachers, feedback and learning strategies, and self-efficacy.The whole sample as well as those at varying English proficiency levels demonstrated similar beliefs and perceptions as learners of foreign languages.They generally believed that making mistakes was a natural part of language learning, that learning a foreign language took time, that students should resort to various strategies according to individual needs, and that teachers, feedback, practice, and personal efforts all played important roles in successful language learning.They also believed that they had the ability to learn a foreign language well.All these indicated that the students were generally confident and autonomous learners.
The study also showed that students at different English proficiency levels differed significantly from one another in English learning beliefs and selfefficacy.Though mixed findings occurred, band 2 students had stronger beliefs about the value of independent and autonomous learning and greater selfconfidence in successful language learning than the other two groups.Even so, in terms of overall readiness for autonomous and independent learning, there is still room for development.This is especially so for students at lower profi-ciency levels.Considering the great impact of long-history traditional Chinese culture and classroom education, it may take Chinese learners a lot more efforts to transit from passive to active, independent, and autonomous learners.The underlying mechanisms for this transition deserve research in the future.Concurrently, teachers can help students with the transition by sharing learning experiences, co-setting goals, assigning both individual and collaborative projects and other means.
Meanwhile, as found in Woodrow (2011), students with high self-efficacy tend to study harder and are more intrinsically motivated.Heidari et al.'s (2012) study indicated that self-efficacy beliefs had great impact on successful learning experiences and achievements.Thus, it is beneficial to develop positive self-efficacy beliefs in learners.For example, a positive and encouraging attitude can help facilitate students' positive self-efficacy beliefs (Woodrow, 2011).Teachers can help students hold correct beliefs about foreign language learning to motivate them to study the target language (Genç et al., 2016).Moreover, as discussed in Zhong (2012), learners might be more likely to act upon their beliefs when they are high in their self-efficacy.Therefore, it is necessary to boost students' self-efficacy, in that when they believe in their ability to perform they would invest more efforts into language learning and thus achieve better learning outcomes.

Table 1
Means, Standard Deviations and ANOVA Results of English Learning Belief Inventory Items

Table 2
Frequencies, Percentages and AMOVA Results of Self-Efficacy Belief Questionnaire Items The first number is frequency and the second is percentage. Note.