Investigating Undergraduate ESL Learners' Readiness for Autonomous Learning

: The paper reports on a study which investigated undergraduate ESL (English as a Second Language) students' readiness .fbr autonomous learning of English. The main objectives of the study were to get a better understanding oJ. Bangladeshi undergraduate ESL learners' perceptions about autonomous learning and teacher-centered learning, and to identify whether they are ready for outonomous learning. The study is based on a five-point Likert Scale questionnaire sut"vey followed by a semi-structured interview conducted with the Jirst year students of a major public university in Bangladesh. Quantitcttive data were analyzed using the Statistical Package for Social Sciences f,SP.S,S version 16.0). Descriptive statistics were calculated for each variable in the questionnaire in order to describe the respondents' perceptions about autonomous learning and teacher-centered learning. Paired-samples t tests were calculated to identify any significant dilference (p>0.05) in the respondents' preference for autonomous learning to teacher-centered learning. The results show that the participants in the study prefer autonomous learning to teacher' centered learning which may imply their readiness for autonomous learning. Though it is a commonly-held belief that teaching in Bangladesh is teacher-centered and students are heavily dependent on tea.chers at all levels, the study suggests the potential for implementing autonomy in undergraduate English language clqsses in Bangladesh.


Introduction
In autonomous learning, learners are in charge of their own learning and they take all decisions about their learning (Holec, 1981). If learrrers are given opportunities for autonomy, learling can be more focused and meaningful for them (Little, 1991;Dam, 1995;Camrlleri, 1997;Chan, 2007Chan, , 2003. Again, the development of autonomy in foreign tanguage classrooms can contribute to achieving optimal success in language learning (Little,2OO4a). Therefore, it is important to encourage learners and give them opportunities to be autonomous in their language learrring. But before implementing learner autonomy in the classroom, teachers need to know whether students are ready for autonomy because it depends on learners' "capacity and willingness" (Dam, 1995) to become autonomous. Learnershave to have both ability and keenness simultaneously to be autonomous (Sinclair, 2000); otherwise it may be diffrcult for teachers to develop autonomy among the leamers.
In Bangladesh, teaching is generally viewed as teacher-centered (Choudhury, 2006) and students at schools are heavily dependent on teachers. As a result, students may have developed a habit of teacher-dependence which they may carry with them to their undergraduate classes. But, do these students really prefer depending on teachers to depending on themselves for their learning? Are they ready for learrrer autonomy?
To answer questions like these we need research into autonomous learning at the tertiary level in Bangladesh, which so far remains largely unexplored. Considering this . Associate Professor, Deparlment of English, University of Dhaka z need, the present study was undertaken with a view to getting a better understanding of Bangladeshi undergraduate ESL learners' perceptions about autonomous learning and teacher-centered learning, and exploring whether they are ready for leaming English autonomously.
In line with the objectives, the study addresses the following research questions: (l) What are the most commonly held perceptions about autonomous learning and teacher-centered learring among the Bangladeshi undergraduate English language learners? and (2) Do the Bangladeshi undergraduate ESL students prefer autonomous learning to teacher-centered learning?
Learner autonomy and language learning The notion of learner autonomy was first developed in the early 1970s by Holec and his colleagues (Smith, 2008). In language learning, autonomy means "the ability to take charge of one's own learning" (Holec, 1981, p. 3).By "taking charge of one's learning" what Holec (2001) means is "to have, and to hold the responsibility for allthe decisions concenring all aspects of this learrring." According to Littlewood (1996, p.97), autonomy is the "learners' ability and willingness to make choices independently." Although learner autonomy empowers learners to take all decisions regarding their learrring, it does not mean that they will do it alone without a teacher. As Little (1991) mentions, learner autonomy is not learning without a teacher, nor leaming without interaction. It is not even a perrnanent state. The leamer's willingness of being autonomous may also vary from time to time. So, teachers need to know whether learners are ready to learrr autonomously in the classroom as "learner autonomy is characterised by a readiness to take charge of one's own learning in the service of one's needs and purposes" (Dam 1995, p.1). Dam (1995) also added a new dimension to the concept of leamer autonomy which requires leamers to act not only independently but also "in cooperation with others, as a socially responsible person" (p.1). So, we can consider learner autonomy as comprising both independent and collaborative learning.
According to Trim (as cited in Norilda et a1.,2004), a school or even a university cannot provide its students with all the skills and knowledge they would need throughout their lives. Therefore, it is important for teachers to teach their learners how to learn on their own so that they can use their learning outside the classroom. If learners are reflectively engaged in planning, monitoring and evaluating their learning in the class, their learning will be more successful than otherwise, and they should be able to use their knowledge and skills acquired in the classroom in the world beyond (Little,2000).
Before teaching learner autonomy, a question may be raised whether it can be implemented in all cultural contexts as the concept of leamer autonomy emerged in Europe and it is often treated as a westem concept (Sinclair, 1997;Jones, 1995).
Though learner autonomy is a western concept, there is no reason to think that only western learners are capable of developing learner autonomy. As Smith (2008) points out, "the exercise and development of learner autonomy can be seen as an educational goal which is cross-culturally valid" (p. 396) and, according to Little(2Oo4b), there is enough evidence to support leamer autonomy as a psychological phenomenon which can transcend cultural differences. Again, learler autonomy has multiple interpretations which are not solely westerrr but universally appropriate and accepted as an important general educational goal (Sinclair, 1997). Therefore, we can also consider the promotion of learrrer autonomy in context of Bangladesh because a proper implementation of learrrer autonomy would make learners autonomous and help them take responsibilities of their own learning of English' Research on learners' perceptions on autonomous learning In order to implement learner autonomy, it is important for teachers to investigate whether the learners are ready for autonomy' Many studies have been done to understand learrrers' perceptions on autonomous learning in L1, ESL and EFL contexts. But I will review below iome studies conducted in ESL and EFL contexts as they resemble the Bangladeshi context of teaching and learning English.
A research srudy by Thang and Azarina (2007) in the Malaysian ESL context observed 756 undergraduate ESL learners of three Malaysian public universities. The study revealed that host of these learners preferred teacher-centered learning. This finding was supporled by Thang's earlier studies (2003Thang's earlier studies ( , 2005 which reported Malaysian undergraduate students' lack of autonomy in their learrring of English as a second language and lack of awareness of their language learrring processes. Malaysian undergraduate learners' teacher-centeredness was also supported by Thang's later study (2009, where she compared between public and private university ESL learners. The study reported that boih learner groups showed their preference for a more teachercentlred approach, but the private university students were likely to move towards a more autonomous learning approach. Thang and Azartna (2007) concluded that though the learners were teacher-centered, it cannot be said that they are not able to be autonomous. They "desire the freedom and responsibility to decide what, where, when an<l how to learn. They prefer to employ their own learning styles and are confident jn themselves..." (Thang & Azarina, 2007, p. l4).
On the other hand, a study by Vanijdee (2003) in the Thai EFL context with distance learners in a Thai university reported Thai learners' preference for autonomous learning. Again a case study by Intradat (as cited in Thang & Azarina, 2007) which investigatedboth Thai teachers' and learners'perceptions about learner autonomy in a CALLlComputer Assisted Language Learning) course also supported the frnding. The study showed Thai leanters' positive attitudes towards learrrer autonomy.
Another study by Guo and Zinarrg (2005) in another EFL context, China, with the undergraduate students at a Chinese university reported that teaching autonomous learrring strategies improved EFL outcomes of the students. The study compared the results after using traditional and autonomous teaching approaches with those students.
The study reported that the group receiving autonomous teaching became more confident in speaking English, more aware of learrring goals, strategies, and more involved in the learning process than the group receiving traditional teaching. Although this study revealed positive results for leamer autonomy, the researchers remarked that in the Chinese cultural context it might not always be easy to develop learrrer autonomy.
From the above studies we find that leamers' preference for autonomous and teacher-centered learning varies according to the learning contexts. In an E,SL context like Malaysia (See Thang 2003, 2005Thang & Azarina, 2007) learrrers were reported to be more teacher-centered. On the other hand, in EFL contexts like Thailand (see Vanijdee, 2003) and China (see Guo & Zhang, 2005) students preferred autonomous learrring. Besides learning contexts, the nature of the courses may also play an important role in students' preference for autonomous learning, as it is found that distance learrrers prefer to be autonomous while regular face to face learners express their preferences for teacher-centeredness.

Bangladeshi undergraduate English learners'perceptions on autonomous learning
In Bangladesh, English is taught and used as a second language though it is not officially recognized as a second language. For the pulposes of this study, however, English is used as E,SL and the leamers are considered ESL learners. In Bangladesh, English is taught as a compulsory subject from class one to undergraduate level and English is often used as a medium of instruction. At the pre-university level, teaching is heavily teacher-controlled and teacher-directed. This is also true for teaching at the university level as Choudhury (2006, p.88) mentions: "teachers prefer to conduct language classes in lecture mode." Students sit and listen to the lectures passively. Though this comment may not reflect the picture of all tertiary level institutions in Bangladesh, there is some truth in it that the Bangladeshi university students depend onteacher-centered leaming, which may be due to their previous schooling.
Students' dependence on teacher-centeredness is also found by Saha and Talukdar (2008) who investigated undergraduate ESP learrrers of a public university. They found none of their respondents as confident users of English and who preferred to depend more on their teachers and their classes. The researchers also reported that the students preferred to be taught by teachers over leaming English by themselves. Another study by Jamil (2010) investigated the possibilities of promoting learner autonomy for the EAP leamers of a private university. The study reported that making leamers aware of the learling process and giving them responsibilityfor their learring had both advantages and disadvantages for the leamers. While some learrrers made quick progress, others found it difficult to take charge of their learrring. They even refrained themselves from "interpersonal and social interactions like group work (and) peer review" (p. a8). The finding implies that some learners are not ready for autonomy.
From the above studies, we cannot clearly conclude whether Bangladeshi undergraduate students prefer autonomous learning or teacher-centered learting. We need more research, which is not available, to examine students' preferences or readiness for autonomous leaming if we want to promote learner autonomy in their classes. From this consideration, the present study was undertaken and so it was limited to the examination of learrrers' readiness for autonomous learrring.

Theoretical Framework
In this study 'autonomous leaming' is used synonymously for lear:rer autonomy.
Here'autonomy' does not refer to 'absolute independence' and autonomous learrring does not mean'leatrting without teachers' (Little, 1991). The reason for taking this view of autonomous learning is that students need to learn about learner autonomy from their teachers in the classroom situation as autonomy is not innate; "it must be acquired either by 'natural' means or by formal learning" (Holec, 1981, p.3). Holec's (1981) concept of learner autonomy has been used as the basis for the theoretical framework in this study. The reason for using his concept is that Holec has given a comprehensive idea of leamer autonomy and from his concept we find a clear guideline regarding the areas of learning where autonomy can be exercised by learrers. According to Holec(1981),autonomous learners have the ability to take charge of their learning by taking decisions regarding all aspects of learning, i.e. "deterrnining the objectives; defining the contents and progressions; selecting methods and techniques to be used; monitoring the procedure of acquisition properly speaking (rhythm, time, place, etc.) and evaluating what has been acquired" (p. 3). Based on this concept of leamer autonomy, the investigation in this study is limited to the five aspects of learning: decision making, learning styles and strategies, awareness about language and the self, assessment and feedback, and learner initiatives.

Decision making
In language teaching, decisions are usually taken by teachers who set the learning goals for learlers, select the materials and evaluate their performances. But in autonomous learning, it is the responsibility of learners to take and implement all the decisions related to their leaming (Holec, 1981;Dickinson, 1987). In reality it is hardly possible for learrrers to do so alone in a formal classroom setting where they have to depend on teachers for many of their tasks including the syllabus and cufficulum, assessment, etc. Therefore, it is the responsibility of teachers "to encourage learrrers to take active part in making decisions about their learning" (Scharle & Szabo, 2000, p 4) in order to develop learner autonomy (Balcikanli, 2010).
Learning stl,tles and strategies One of the main characteristics of autonomous learners is that they have "insights into their learning styles and strategies" (Omaggio, as cited in Wenden, 1998, p.41). They are consciously aware of and use both cognitive and metacognitive strategies which are again one of the conditions for autonomous learning to occur (Thanasoloulas, 2000). They take a greater degree of control not only over the contents but also over the methods of leaming than what they usually do in a classroom learrring environment (Norilda et a1.,2004). The autonomous learners constantly review and reflect on their learning processes. But it is noted that this is an ideal picture of an autonomous learner.
Awareness ubout the language and the self Autonomous learners are aware of themselves as leamers, their learning context, the language tobe learnt, and the learrring processes. They have a clear idea about the purpose of leaming the language, the areas of their problems and improvement in that language. They can assess their strengths and weaknesses in their learning (O'Connor, Farrar & Crome, 2009). Learners' conscious awareness of all these is important without which they cannot make informed decisions about their own learning (Sinclair, 2000) and review their progress and future directions of their learning (O' Mallay et a1., as cited in Sinclair, 1999).

Assessment and feedback
Traditionally, evaluation is thought to be the teacher's job, but in learner autonomy, leamers are encouraged to assess their performances as "autonomous leamers consciously monitor their own progress, and make aL effort to use available opportunities to their benefits, including classroom activities and homework" (Scharle & Szabo, 2000, p 3). Monitoring and evaluating own performances are important because if they do So, their leaming may be more successful (Little, 2000).
Autonomous learners also appreciate feedback from peers. D Learner initiatives Learners' own initiatives are important because "people who take the initiative in learning leam more things and learn better than those people who sit at the feet of the teachers, passively waiting to be taught" (Knowels, as cited in Guo & Zhang, 2005, p 13). Autonomous learners take initiatives and try out new things in their learning. Language is learned to be used to communicate with others at the time of need. So autonomy in language learning is essential for learners to become effective users (Littlewood, 1996;Nunan, 1997) because, according to Little (2004a, p. 19), "autonomy in language learning and autonomy in language use are two sides of the same coin: you simply cannot have one without the other." Finally, learners' own efforts are crucial because successful language learning depends not only on the teacher's contribution but also on the learner's participation (Scharle & Szabo, 2000).

Methodology Instruments
The study was based on the mixed method model. Two instruments, a questionnaire and semi-structured e-mail interviews, were used to collect quantitative and qualitative data.

Questionnaire
The questionnaire had two major sections: section one contained fifty items designed to elicit responses on a five-point Likert scale ranging from strongly agree to strongly disagree while section two contained four open-ended questions to get detailed information on specific items of section one.
The Likert scale items were based on Thang & Azarina (2007) and Cotterall (1995) because their questionnaires had good reliability scores and they were used in a similar context to Bangladesh. Nevertheless, the questionnaire items were modified considering Bangladeshi context and the competence level of the participants. In doing so, feedback from two participants in this study and expert opinions were taken into consideration before using the questionnaire for data collection. The items were categorized into two groups. Item 1 to 25 were designed to elicit students' opinions about autonomous learning while item 26 to 50 to elicit their responses about teachercentered learning.

Interview
Semi-structured interviews were conducted via e-mail with four participants. An interview through email was chosen because it is an easy way to get responses from participants. It supplies already transcribed data and it allows more time to the respondents. As a result, they can give their thoughtful opinions and detailed answers to the interview questions. The interview comprised seven questions which intended to solicit information in the relevant areas of learning which were used in the questionnaire.

Samples
The sample population in this study comprised sixty nine (male:41, female:28) frrst year BA (Honors) students of the 20lO-\1 academic session majoring in English at a major public university in Bangladesh. All the participants attended a compulsory English language course (Eng101: Advancing English Skills) in their f,rrst year.
Purposive sampling method was used to select the samples in this study. The samples were selected from their regular class setting.
Procedures of datu collection snd dotct anulysis Likert Scale questionnaires were used to collect quantitative data while open-ended questions and semi-structured interviews were used to collect qualitative data. The questionnaires were administered during a class of the participants while the semistr-uctured interviews were conducted through emails. The interview questions were emailed to ten of the participants who showed their interest in participating in the interview. However, data were received frorrr four of them only.
The quantitative data were analyzed using Statistical Package for Social Sciences (SPSS version16.0). Descriptive statistics (frequencies, means and standard deviations) were calculated for each variable in order to describe the perceptions of the respondents about autonomous learning and teacher-centered learning. Again, a paired-samples / test was employed in order to identify any significant difference (p>0.05) in the respondents' preference for autonomous learning to teacher-centered learning. On the other hand, qualitative data were analyzed by identifying common themes. These data were used to verify the findings from quantitative data.

Results and Discussion of Findings
The results and frndings of the questionnaire data are discussed first in order to identiff the most commonly held perceptions about autonomous and teacher-centered learning among the participants.

Decision Making
The results in Table-1 below show that the highest percentage (97.1%) of the respondents agreed with item-1:"students should set their goals and objectives of learning English" while a huge majority (91.13'h) expected the teacher to allow them to choose the materials and 85.5o liked the opportunities to select classroom activities.
The responses of the similar items under teacher-centered learning show that the lowest percentage (39.1%) of the respondents liked "the teacher to choose (their) goals and objectives of leaming English" (item-26) whereas relatively a lower percentage (68.1%) of the respondents thought that the teacher should select materials and activities for students. These findings reflect the participants' willingness to be more autonomous in decision-making about their goals and objectives of learning, learning materials and class activities (see Holec, 1981). But they like to depend more on the teacher for evaluation of their performances as a large majority (87%o) of the respondents supported this (item-31). A possible reason for this may be that the students lack confidence to evaluate their performances as they had no experience of doing this before. Therefore, teachers should involve them in decision making processes and thus help students take decisions about their learning (see Scharle & Szabo, 2000). The results in Table-2 below reveal that a good majority (82.6yo) of the respondents agreed with item-7 that "teachers should give opportunities to students to learn in their own learning styles" and 79.7yo with item-9 that "teachers should allow students to choose their own methods and strategies to learn English effectively." But it is interesting to note that majority (69.60/o) of the respondents liked the teacher to tell them their learning styles (itern-32) while alarge majority (81.2%) liked to depend on the teacher for their methods and strategies to learn English successfully (item-34). The finding implies the participants' lack of insights about their learning styles and strategies although they have a desire to pursue their own styles to learn English. This frnding n;ray also be treated as a shorlcoming of these learrrers' autonomous characteristics because autonomous learners are aware of their learning styles and strategies (see Wenden, 1998). A possible reason for this may be that they are not used to thinking about their learning styles and strategies as they might not have been asked to do so before.
Again, agood majority (85.5%) of the respondents admitted the importance of reflection on their learning which may imply their awareness of metacognitive strategies (see Sinclair,2000). But at the same time,76.8Yo of the respondents liked the teacher to make them think and reflect on their learning (item-36) while a majority (79.7%) liked "the teacher to direct (them) on how to learn English" (item-35). This finding resonates with the idea that autonomous learning does not mean learning without the support of teachers (Little, 1991). We have a contradictory finding, however.Although 69.6oh of the respondents agreed with item-13, "I know which aspects/areas of my English I need to improve," in response to item-38, 85.5yo of the participants admitted that they liked the teacher to tell them the areas in English they should improve. It was also found that majority of the participants were not aware of the importance of learning English as the results show that 79.7yo of the participants liked the teacher to tell them the reasons for learning English (item-37).
All these findings reveal the participants' lack of a-wareness of their potentials as learners and knowledge of the target language which are important features of autonomous learners (see Sinclair, 2000).  (Holec, 1981;Little, 2000). This is confirmed by a vast majority (91.3%) of the respondents (see item-19). The results in Table-4 also reveal that a great majority (88.4'yo) of the respondents would like to get oppofunities to self-comect their mistakes in their work. This frnding contradicts the findings in Table-3 above (see responses for item-14 and item-39) and also with the responses for item-44 inTable-4 below that show 75.3Yo of the respondents liked the teacher to correct all their mistakes.
The results in Table-4 show that a large majority (81.2%) of the respondents admitted the importance of feedback in learning language successfully whlle 78.2oh of the respondents like to get feedback from their peers. This finding rnay indicate their autonomous nature as they like collaborative learring (see Dam, 1995). But it is noted that almost the same percentage (79.7%) of them liked feedback from teachers while a large majority (84%) opined that they needed the teacher to tell them about their progress in English. These findings here indicate the respondents' both autonomous and teacher-centered characteri stics. 5=strongly agree, 4-agree, 3not sure, 2-disagree, 1=strongly disagree, M= mean

Learner initiatives
The results in Table-5 below reveal that almost all the participants (91 .l%) admitted the importance of using English outside the classroom for developing their skills. But it is noted that a large majority (81.1%) of the respondents liked the teacher to tell them about their language learning activities outside the class (item-47). This finding reflects the respondents' dependence on teachers for taking initiatives to learn English even outside the classroom though a large majority (84%) of the participants admitted the importance of their own efforts for learning English. Interestingly, in response to item-50, a large majority (81.1%) of the respondents were of the opinion that the teacher's role was very impoftant for their success in language learning while 82.6% of the respondents liked the teacher to motivate them to learn English.
The findings here imply the participants' awareness of the importance of selfefforts and initiatives for their lear:ning. So, the participants may be considered to have a characteristic of autonomous leamers as self-efforts and initiatives are crucial for autonomous learning (see Scharle & Szabo, 2000). But it can be mentioned here that the participants also undefscored the importance of teachers' rolesin their language le arning, which agarn may refl e ct their te acher-dep endenc e. From the discussion of the descriptive statistics, we flnd mixed results.That is, the participants like some aspects of autonomous learning and some aspects of teachercentered learning. Therefore, it is diffrcult to determine whether they prefer autonomous learning to teacher-centered learning or vice versa. For this reason, a paired-samples / test was done in order to see any statistically significant difference in the participants' responses for autonomous learning and teacher-centered learning items.

Paired-samples / test results
The results in Table- Therefore, from the analysis of the paired-samples / test results we can say that the participants in this study prefer autonorious learning to teacher-centered learning. The possible factors which may contribute to this finding are the participants' maturity as learners; their understanding of the importance of using English outside the classroom (see Table-5, item-23 which has the highest mean, M:4.72 of all items), and university education system which requires a lot more self-study byuniversity students than the secondary students (the questionnaire was given to first year BA students too, not secondary students).  The participants were asked whether they had ever had any opportunity to choose learning materials, class activities and evaluate their own performances in the classroom. It was found that only about one-fourth of the respondents had such opportunities. They were also asked to specify when they got these opportunities. It was noted that only one mentioned that he had got an opporlunity of choosing class activities in school while the rest had the opporlunities of choosing materials and class activities in their first year English language classes. This finding indicates that students have started getting decision-making opportunities in the undergraduate classes to some extent.
In reply to a question of their willingness to have these opportunitiesto take decisions about their learning and evaluate their own performances, over three-fourths of the respondents replied positively. The responses were categorized on the basis of conlmon themes. The mainobjectivesof their willingnessare to achieve learning goals, to learn English effectively, to become active and consciouslearners, to improve language skills, to plan for future learning, to become aware of their mistakes, and to develop themselves as confident learners.
These responses may imply that the participants are willing to become autonomous in their learning of English as they are aware of the benefits of autonomous language learning.
Re sp o n s e s of s e mistru ctur e d interview s As mentioned in the methodology, semi-structured interviews were conducted with four of the participants. Their responses reveal their insights and points of view about autonomous learning. About choosing learning materials and classroom activities, the students gave practical opinions as they mentioned that if they were given such opportunities they would feel more comfortable and therefore would perform better in learning English. o was found that the respondents were also aware of the benefits of taking charge of their learring by selecting materials, activities and evaluating their own perforrnances.Regarding evaluation of their own performances, although the students acknowledged the traditional role of a teacher to evaluate students' perforrnances, they would like to experience it with the help of the teacher as one student mentioned "they (students) should give a try to evaluate their acts in the class, but finally the role is given to their mentors (teachers)." One student also explained in the following quote why they should depend on teachers but at the same time he also asserted his point of It is noted that the respondents were not quite confident in taking charge of their learning as they had neither expertise nor any experience to do so. In their preuniversity classes, no student had any opportunity to take decisions and choose activities on their own. In their undergraduate classes though they have some opportunities, but students consider them to be inadequate. Nevertheless they expressed a desire to experience it which is evident in one student's comment: "No, I'm not very confident about it, but if I am given a chance to choose materials or activities, I will get to know about my preferred way of learning." The findings from the open-ended data and interview data reveal the respondents' willingness to take decisions about their learrring of English though they would like to have some guidance from the teacher. These findings support and confirrn the findings of the quantitative data that the undergraduate English lear-ners in the university investigated are likely to prefer autonomous learrring. These findings contradict Saha and Talukdar (2008) who found their undergraduate learners more teacher-centered. A possible reason for this difference may be that the participants in their study were ESP learrrers whereas in this study they were learners majoring in English. But it is to be noted that the findings in this study are supported by those of Jamil's (2010) who repofted his participants' willingness to take part in learner autonomy although some of them were reluctant to accept autonomy.
The frndings again contradict those of Thang and Azarina (2007) who reported Malaysian undergraduate learners' preference for a teacher-centered approach to learrring although they possessed some characteristics of autonomous learning. But the findings resonate with that of Thang (2009) as they imply Bangladeshi undergraduate learners' willingness for autonomous learning within a predominantly teacher-centered teaching environment. Following Thang (2009),it can be stated that the findings in this study suggest that Bangladeshi undergraduate learners are not incapable of autonomous learning if they are given proper opportunities.
Implications of the Findings The findings in this study have important implications for teachers at the tertiary level in Bangladesh. They can attempt to implement learrrer autonomy in their English language classes as the findings reveal leamers' preference for autonomous learning.
But before implementing autonomy, they have to be cautious and be ready for some challenges. It may not be an easy task because before entering the university, students undergo twelve years of schooling which is predominantly teacher-centered (Choudhury, 2006) and this long term teacher-centeredness may havehelped develop a habit of teacher-dependence among the learners. So, teachers should proceed slowly in their attemptsat promoting autonomy in language classes. They need fuither research to identify learners' personality traits and preferred learning styles which influence the development of autonomy (Scharle & Szabo,2000, p. 5). Moreover, teachers also need some autonomy-oriented training; otherwise they may face difficulties in implementing learner autonomy in their classes (BalQikanli, 2010). They should be ready to accept their new roles of a facilitator of leaming, counselor and learning resource managff in the classrooms for autonomous learning to succeed (Little, 1995;Voller, 1997).
Overall, the study may help the teachers with a better understanding of students' insights to plan their lessons for English language teaching in a better way with a view to making the learners autonomous in their learning and using English. The study may encourage other teachers to survey their students before introducing any autonomous learning program.
The study may also have important implications for other students who will know about the undergraduate learners'perceptions about autonomous learning and teachercentered learning. They may be encouraged to appraise their own learning of English.

Conclusion
The study investigated Bangladeshi undergraduate English learners' readiness for autonomous learning. The findings of both quantitative and qualitative data suggest that the participants in the study prefer autonomous learning to teacher-centered learning which may imply their readiness for autonomous learning. Thus, the study reveals a potential for implementing autonomy in the undergraduate English classes. But it is hard to generalize the results and state that all the undergraduate learners of English in Bangladesh are ready for autonomy because the sample size in this study is small (N:69). Although the mixed method was followed in this study, data were collected from students only. If teachers were included in samples, we could have received better perspectives of students' readiness for autonomy. Again, qualitative data were collected through open-ended questions and semi-structured interviews via email. But face-toface interviews could have produced more detailed information. Furthermore, classroom observation and data triangulation would have helped to verify the findings and come to a better conclusion.
Despite these limitations, the present study is important because the findings imply a potential ground for promoting autonomous learning at the tertiary level in Bangladesh. But further research is needed with the participation of a larger sample from different universities to see whether the findings of this study remain the same. Teachers' perceptions of autonomous learning can also be researched in fufure before taking any attempt to promote autonomous learning as a goal of language learning program.