To Give Control to Learners or Not ? A Comparative Study of Two Ways of Teaching Listening

This paper reports a quasi-experimental study that was carried out in listening classes to address the following questions: What control do students expect in the listening classroom? What are the effects of the “zero class hour” way of teaching listening when more control is given to the students? The study was conducted at Huaiyin Normal University, where two non-English major classes taught by one teacher were selected. A questionnaire survey was conducted to get students’ opinion about the traditional way of teaching listening and what control they wanted to have. The results of the pretest and posttest were analyzed with SPSS. Both the experiment group and the control group made similar progress in the test of their listening comprehension, which indicates that the new way of teaching listening is as effective as the traditional one. From the amount of time that the experimental-group students spent on improving their listening comprehension, we can see that they made more efforts and became more active and more autonomous than before. The findings may give some support for the web-based distance education.


Introduction
General trends in education have been moving away from the teacher-directed regime towards a self-directed regime.In traditional listening classes, on one hand, there were many complaints from teachers in China about the heavy workload and about how students are not.On the other hand, many students complained about the boredom of the listening classes.Then how to solve the dilemma deserves our thinking.
According to the results of one questionnaire survey conducted by the researcher in two classes at Huaiyin Normal University, about 71% of students thought that they should be given more control of their own study.The percentage of students who want to control their learning times, their learning pace, the length of pause and the physical condition reaches 74%, 65%, 67% and 77% respectively.About 78.8% of the students expressed their expectation to have more control of their learning according to the open-ended question.
Can this be a way out?If students are given more control of the study, will they be more responsible for their own learning and their study become more efficient and more effective while at the same time teachers' heavy workload can be reduced?
In the traditional listening class, the teaching is usually carried out in the language lab or in a classroom.The teacher will have strict control of the class, for example, the teacher controls the time and the times that students listen to the listening material, the pace of learning process, etc.The teacher plays the tape and checks the answers with students.Then the teacher will give some explanation and play the tape again.Under this kind of teaching: 1) Students usually have little control of the learning process; 2) Students just passively do what teachers tell them; 3) Students don't have sufficient listening input; 4) Students' anxiety is very high in answering questions; 5) Students' efficiency of using learning time is low.
In order to solve those problems, many universities are carrying out reforms in the teaching model in the teaching of English listening.For example, Shanghai Foreign Language University and Sichuan University are teaching listening in the form of "zero-class hour" model.The former doesn't limit the learning hours and students only take the exams at the end of the semester.The latter has a strict control of students' total listening time that students must finish 400 hours' listening and pass four tests to get the credit.Some other universities are also carrying out similar reforms and the common thing of these reforms is that they give more control to students so as to make independent learners out of students.However, up to now there are few empirical studies on this kind of teaching model.This paper reports an experimental study which investigated the effects of listening teaching when more control was given to EFL learners.The study will shed light on the matter of learner control and give some implications to the current language teaching.

The concept of autonomy
There is probably much confusion and disagreement about what is meant by the term learner autonomy.
Although it has taken many names and forms, the idea of learner autonomy really began to take hold in the theory and practice of pedagogy in the 1970s (Dickinson, 1987).Whether we are talking about self-directed learning (Knowles, 1975), student autonomy (Boud, 1988), learner-managed learning, independent learning or learner autonomy, we are for the most part talking about the same thing.As there have been a number of quite useful definitions put forward for the concept of autonomy in pedagogy (see Benson, 1997), it is beyond the scope and purview of this paper to re-cover this well-trodden ground.Instead, what is worth noting is the timing of the advent of contemporary learner autonomy theory.With both of them coming to prominence in the 1970s, it is not my aim to prove that post-modernity caused learner autonomy, or vice versa.Rather, it is the contention of this paper that theories of post-modernity and learner autonomy are inextricably linked, each informing and informed by the other, resulting in practice and applications in both fields that mirror each other in an endless cycle of reflection.
Based on the foundation document for the Council of Europe, the classic definition of "learner autonomy" is provided by Holec (1981:3), as the capacity or ability to take charge of one's learning which involves: Defining one's own learning objectives, determining the necessary means for attaining those objectives and the ways of using those means, while determining the means for assessing what has been learnt and acquired (Holec 1997:25).Sheerin (1997:54) emphasize that "autonomy" and "self-direction" are approaches related to helping learners develop/take control over learning, whereas, "self-access learning" and "self-instruction" (alongside "distance", "flexible" and "open" learning) are terms which refer to systems or organize learning.
In order to take responsibility for our own learning we must believe that we have control over learning success and failure, and consequently attribution theory has important implications for the promotion of autonomy.
According to Benson and Voller (1997:5) different roots create tensions between responsibility and freedom from constraint, and between the individual and the social, and have led to different interpretation of autonomy.Benson (1997:25) identifies three versions: 1) the act of learning on one's own and technical ability to do so, deriving from Positivist theories which view teaching as the means to equip learners for autonomy to manage learning outside the classroom or exercise independence within it; 2) the internal psychological capacity to self-direct one's own learning, deriving from Constructivist theories which support self-directed and self-access learning as a means of promoting learner autonomy; 3) the control over the content and process of one's own learning, deriving from Critical theory which emphasizes issues of power and control and the growth of autonomy as learners become more aware of the social context of their learning and the constraints it implies.
Although these versions are rarely clear-cut in practice, they affect methodological approaches for developing approaches for developing autonomy.Despite more recent exploration of the issues involved (Little 1991), Benson argues that "…so far, we have no theory of autonomous language learning" (1996:28) and warns against rapid implementation of autonomy without reference to theoretical bases (1997:2).Nunan (2000) explained his understanding of "autonomy": In its general application, autonomy implies a capacity to exercise control over one's own learning.Principally, autonomous learners are able to,

•
Self-determine the overall direction of their learning, • Become actively involved in the management of the learning process, • Exercise freedom of choice in relation to learning resources and activities.
Research on autonomy in language learning draws on two major sources.On the one hand, researchers within the sociology and psychology of education have argued persuasively that autonomy is beneficial to learning, irrespective of the subject matter to be learned (Candy, Brockett & Hiemastra, Boiud. cited in Nunan 2000).On the other hand, there is now a considerable body of research within the field of language education itself which supports the contention that autonomy and self-direction are beneficial to second language acquisition in particular.Although the importance of autonomy to effective second language acquisition is often argued on the basis of learning theory alone, research in the field of second language acquisition is of particular interest because it offers the possibility of grounding the theory of autonomy in language learning on evidence that is particular to the process of learning a second or foreign language.
In this study, learner autonomy refers to an attitude and capacity on the part of the learner towards taking control of the language learning process and assuming responsibility for the process.

Studies on autonomy and learner control
Some researchers advocated giving some controls to learners.In one paper, Brown (2003) compared teacher-centered and learner-centered classrooms and concluded that an instructional paradigm shift is needed to implement a learner-centered approach.Her research indicates that more control should be given to students.Benson (2002) explored the relationship of self-access and autonomy and thought that teachers should develop students' autonomy through self-access learning and one major characteristics of self-access learning is that learners have more control than ever.
However, some others scholars disagreed to give all the controls to learners.Chanock (2003) argues that although one major principle of autonomous learning is that students should take responsibility for their own learning, but in Confucian-heritage cultures where students are particularly resistant to that idea and are overly dependent on their teachers, it is not responsible to be dependent and depending on others can be a responsible way to learn.Methods of autonomous learning are not in fact incompatible with depending upon teachers.
Following his point, we can see that teachers should not give all the controls of learning to students in Confucian-heritage cultures.Garcia (1996) explored the effects of autonomy on motivation and performance in the college classroom and concluded that cultivating a sense of autonomy among college students need not mean a submission to anarchy or to complete student control.That is to say, teachers should give some control to students, for example, by allowing students to participate in course policy-making, college students reported greater levels of motivation at the end of the semester.
In Autonomy in language learning, Nunan (2000) explored four ways to encouraging autonomy in which teachers can begin to sensitize learners to the learning process, and thereby begin to encourage a greater degree of autonomy.There are: 1) integrating language content and learning process through learner strategy training.
2) incorporating reflective lessons into your teaching 3) drawing up learning contracts

4) learner diaries
From the first way we can see that to let students control their learning content and learning process can encourage learners' autonomy.
Yang (1998) explored a new role for teachers to promoting learner autonomy by combining learning strategy instruction with the content course of second language acquisition.The new role of teachers is to develop students' learning strategy so as to promote their student's learning autonomy and one way is to let students have more control of their own study.Dickinson (1995) reviewed some studies on the relationship between autonomy and motivation and found that the common theme in justifications for autonomy, especially in general education but also in language learning, is that autonomous learners become more highly motivated and that the autonomy leads to better, more effective work.And one example is Knowles' claim: There is convincing evidence that people who take the initiative in learning (proactive learners) learn more things and learn better than do people who sit at the feet of teachers, passively waiting to be taught (reactive learners) …They enter into learning more purposefully and with greater motivation (1975: 14) From the concept of autonomy we can see that control is one of the key components of learner autonomy.In the broadest sense, learner control is the degree to which a learner can direct his/her own learning experience (Shyu & Brown, 1992).More specifically, learner control can be defined as the degree to which individuals control the path, pace, and/or contingencies of instruction (Hannafin, 1984).The meaning of learner control, however, has evolved over time to include the characteristics of new learning paradigms as well as new technologies such as a web-model.
To sum up, learner autonomy involves two characteristics: one is that learner should be responsible for their own studies and the second follows the first one that learners have the partial or full right to decide the learning matters that are traditionally controlled by teachers (Shu, 2004).That is the theoretical background for this experiment.

Studies on the listening teaching
Listening teaching is gaining more and more importance in the teaching of the skills of the English language.
That can be seen from the national English tests in China, in which the listening part occupies a good proportion of the whole test papers.
Rost (1990:76) mentioned that comprehension have been regarded as the most essential aspect of listening.Comprehension has been defined as "the process of relating language to concepts in one's memory and to the references in the real world" (Rost, 2002:59).That is, a listener realizes what he hears, and has the ability to connect the already-heard knowledge to the outside real world.Listening ability is of the great importance to communicate with others.
Rost (1990:87) mentioned that the teaching of could be divided into three aspects: selective listening, global listening, and intensive listening.Selective listening refers to providing to the students a task, which has adequate information, and students try to derive some specific knowledge from the information pool.
Essentialities of selective listening are a large information context with inference cues, a well-designed task to help students receive the crucial information, and pre-teaching activities to help students prepare for the following task.Global listening aims to help students to construct a whole sense of the given content and a gist of the texts.That is, through listening activities, students' ability to identify the topics, or connections and transitions between the topics, could be developed.The third type is intensive listening, which specially focuses on grammatical correctness.The traditional teaching classes mainly deals with the second and the third one.The new way of teaching listening copes with the three aspects all.
Research on the teaching of listening mainly focuses on the specific skills to improve listening comprehension.For instance, Field (2004) explored listeners' problems of using too much bottom-up and too much top-down.Some other researchers concerned about the instruments that can help improve listening proficiency, such as using multimedia computer, using DVDs, etc.Or researchers would like to do studies on how teachers' teaching methodologies affect the improvement of learners' listening proficiency, such as teaching language as communication (Widdowson, 1996) or task-based teaching and others etc.Other research was doing on listening difficulties.Knowing what difficulties students might encounter during the process of learning listening would give great help for teachers' course design and help them to create a low affective-filter learning environment.Few studies were conducted on the learners' part.
This study associates learner autonomy with the listening classes.Some questions would be addressed: what will happen if more control is given to students?Can they become more autonomous?Do learners make more progress in their listening proficiency?

The "zero class hour" listening teaching
The new way of teaching listening may be called the "zero class hour teaching model", but it does not necessarily mean that we don't need teachers in classes.In the new way of teaching, the teacher mainly gives learning aims and assignments and provides learning methods.The students learn independently inside and outside the classroom and attend the final examination.There are mainly three ways of carrying out "zero class hour" listening teaching 1) Through web courses; 2) Through radio broadcasting; 3) Through language labs.The three ways of carrying out the "zero class hour" have their own advantages respectively.In the first and second ways of teaching listening, students still have little control of their learning.For example, the second way of teaching listening students still can't control the pace, the way, the speed, the material of listening.But with the third way, students have the greater control of learning.As a result I chose the third one in this study.
With the new way of teaching listening, there will also be a teacher who will be responsible for the listening class.In each class, the teacher will give a short lecture on listening skills and then give assignments to the students.Then the most learning time will be under the control of the students to finish the tasks given by the teachers.The students can decide to listen to the materials for enough time.They can decide by themselves to listen to other listening materials that interest them after they finished the given listening task.The teacher will check the students' learning results through questions, quizzes and tests.
Compared with the traditional way of teaching listening comprehension, students under the new way of teaching listening have much more control than before.The control is summed as following: 1) The control of the length of listening: students can decide on how long they will practice listening by themselves; 2) The control of pace of listening: students can decide the learning pace by themselves; 3) The control of speed of listening: through some digital devices students can change the delivering speed of the listening material; 4) The control of way of listening: students can decide to practice listening through tape-recorders, walkman or computers; 5) The control of place of listening: students can decide to practice listening in the dormitory, classroom, or language lab; 6) The control of material of listening: students can decide what to listen to that interest them after they finish teachers' tasks.
There are two major advantages with this kind of teaching: relieving the working load of teachers and strengthening students' learning autonomy.But what are the real effects of carrying out the new way of teaching?Should we give more control to students or not when we are teaching listening?What control should we give to students?What are the effects of giving more control to students?
In this paper the researcher compared two ways of teaching listening: the traditional one which teachers have strict control of the class and the new one which students have more control of the learning process to see whether the new way of teaching listening is as effective as the traditional one and other effects it brings about.
Inset Table 1 Here

Methodology
This study employed a mixed design: a survey study and a quasi-experimental study.The survey study was conducted to find out what control students expected to have, whether giving more control to students can foster their learner autonomy, whether learners make more effort when they receiving the "zero class hour" listening teaching.Here effort refers to the amount of time spent outside class improving listening comprehension within a week as reported by the subjects.
The quasi-experimental study was conducted to find out whether the new way of teaching listening "zero class hour" model characterized by giving more control to students in their learning process was effective in promoting learners' listening proficiency.Listening proficiency refers to the scores that learners get on the Listening part of CET-4.

Research questions
1) What control do students expect in the listening classroom?
2) What are the effects of the "zero class hour" model of teaching listening when more control is given to student? a. Do students receiving the new way of listening teaching make similar achievement to those receiving the traditional way? b.Do students receiving the new way of listening teaching make more efforts than those receiving the traditional way?

Subjects
Two intact classes of second-year non-English majors, one in Mathematics and the other in Chinese Linguistics and Literature at Huaiyin Normal University were chosen.In both classes there were 51 students respectively.Altogether there were 49 male students and 53 female students.The two classes are instructed by the same teacher, so from this we can say that the proficiency level of teachers didn't affect the learners' progress in their listening comprehension.The results of Independent-samples T-test of Pretest 1 and Pretest 2 also indicated that the two groups were at the same proficiency level.
The mean of the two tests is 11.14 and 11.59, which is very close to each other.We may say that the two groups are same at their listening proficiency.The null hypothesis here is that the two classes are different at the listening proficiency level.But the significance level is 0.456, which is much higher than 0.05.Thus we may say that the classes are at the same listening proficiency level and the null hypothesis is rejected.

Insert Table 2 Here
From the results we can concluded that the subjects of the two classes were at the same proficiency level.The two classes were treated as the control group and the experimental group respectively.

Instruments
Two listening tests, one questionnaire and an interview were used.The two listening tests, chosen from the listening part of two College English Test Papers (Band four) were used as the Pretest and the Posttest respectively.
The questionnaire mainly asks about students' opinion with the traditional way of teaching listening and what control they want to have about their own listening courses.For instance, the first statement "The current way of teaching listening can effectively promote my listening proficiency" asked about students' view of the traditional way of listening.Statement 17, 18 and 19 told us what specific control students want in their listening classes.The interview was mainly carried out to get students' opinions about the "zero class hour" model of listening.The questionnaire is given in the appendix one.
The interview was conducted mainly to find out students' reaction to the "zero class hour" teaching model.One main question was: "How have you being improving your listening comprehension recently?

Treatment
Before the experiment, a questionnaire survey was given to the students in the experimental group to get their opinion about the current way of teaching listening and what control they want in the listening classroom.Then a pretest was given to both the experimental group and the control group to test their current listening proficiency.Next the control group was taught in the traditional way for two month while at the same time the experimental group was taught with the new way.
In the two months, mainly four listening skills were taught: Listening for details (Numbers, Places, Names, etc), Listening for the main idea, Prediction, Inference (Relationships of speakers, Occupation or professions of speakers).
In both classes, the teacher gave a short lecture about the listening skills each week.The control group learned with the teacher with the traditional method.The experimental group was given tasks to finish immediately after the lecture.The task was composed of two parts: one compulsory and the other optional.The compulsory part was directly related to the listening skills.The optional part was mainly composed of passages for students to listen to and write down the contents.The tasks given to both groups were the same.The experimental group students listened to the listening material by themselves and finished the tasks given by the teacher.In the next class, the teacher would give a lecture on a new listening skill and check the given assignments.
After two months, another listening test (Posttest) was given to both groups to compare their progress in listening proficiency.
The students in the experimental groups were asked to turn in a report about their listening practicing: they should report what they have listened to, how much time they spend on listening practice each week.The time students spent on listening practice before the experiment was marked as Time 1 and during the experiment as Time 2. They should also write down what difficulty they have met and what help they want to have from the teacher.
The following table summarized the differences between the treatments of the two groups: Insert Table 3 Here

Data collection and analysis
There were two kinds of data: qualitative data and quantitative data.Learners' responses to the statements in the questionnaire, learners' scores on the pretest and posttest, and the time that experimental group students spent on improving their listening comprehension outside classrooms before the experiment (Time 1) and during the experiment (Time 2) constitute the quantitative data.The quantitative data were analyzed with SPSS version 11.5.Mainly the Pair-samples T-test and Independent-samples T-test was conducted.The Independent-sample T-test was conducted to test whether the two groups made similar achievement in listening comprehension.The Pair-samples T-test was conducted to test whether the experimental group made progress in their listening comprehension, whether the experimental group made more effort on improving their listening comprehension.

Results and Discussion
Since the data for any T-test should be normally distributed, the marks of the four tests and the time which the experimental students spent on improving their listening comprehension were firstly checked.The results are that they are all normally distributed.But that is not focus of the study, so the results of normal distribution test were omitted here.

The control learners expected
From the students' responses to the open statement "The control I want to have in the listening classroom includes ______" we get much valuable information.We can see that students do want to more control by themselves.For example, one student wrote: "I hope I can control the length of pause between any passage and the next one I am asked to listen to." Most subjects hope that they could control the times for which they listen to the listening material, control the difficulty level, and control the content of the listening.Many expressed the hope of watching the scripts of listening material while practicing their listening.Even one student wished he could adjust his sitting posture.
All these responses reveal much about the current way of teaching listening.Teachers should dwell on the present teaching situation and work out new ways to meet the needs of language learners.
From the learners' responses to other questionnaire items we can see similar results.The five items which scores most were the following items: item 5, item 23, item19, item 17 and item 18.
The result reflects the main content of learner autonomy.Item 5 reflected that students do want to have more control of their own study.Other items reflected the specific control they want to have.From Item 23, 19, 17 and 18 we can see that students do want to control the emotional and physical state, the length of pause, the times of listening to the audio material and the speed of listening material.Those were traditionally controlled by teachers.
Some control that students want to have reflected that students want to have high level autonomy.For example, Item 23 indicated that learners wanted to "choose and shape their own learning context".Learner wanted to "make decisions in domains which have traditionally belonged to the teacher", as was shown in the fact that students wanted to control the listening material, wanted to see the scripts of the listening material etc.
From the following five items that scored least we can also get some information about the students' perception about the current way of listening teaching.From Item 1 we can infer that students are not very satisfied with the traditional way of listening teaching.They didn't think that the traditional way of teaching listening can effectively improve their listening comprehension.
From Item 7 and Item 13 we know that learners still want to have teachers' help in their studies.Few students thought that they had had good learning autonomy and they thought that they needed teachers' help to plan their study.
Item 11 tells us that few students thought that the class time for listening practice was sufficient.On the contrary students attach much importance to listening practice outside the classrooms.
Item 15 indicates that the teacher didn't play the tapes again and again.She explained many things during the lesson so most students thought that they can learn something from their teachers.From that statement we can see the teacher's role can't be ignored.Students needed teachers and teachers shouldn't give all the control to students.
From above we can see that students' responses reflected the general trend from teacher-centered regime towards learner-centered regime.Teachers should follow the trend and give more control to students to foster their learner autonomy.

Students' achievement
This part reports the results of the experimental group students' scores on the pretest and posttest, which are chosen from the listening parts of two TOEFL papers.The full mark of the listening comprehension is 20 points.
From table 4.2.1, we can see that the mean of the posttest scores raised about two points.The significance level reached 0.000.This indicates that the learner made great progress in their listening proficiency.The new way of teaching listening did help students in improving their listening comprehension.

Insert Table 6 Here
The following table compares the two groups of students' scores they got on the posttest.Insert Table 7 Here From Table 4.2.2 we can see that the mean of posttest of the experimental group students is 13.43, which is only a bit higher than that of the control group students.Both groups improved their listening proficiency, as we can see that the mean score of both groups are about two points higher than that of Pretest.The P value as shown in Table 4.2.2 reached 0.686, which is much high than 0.05.Thus we reach the conclusion that students made the same improvement in developing their listening comprehension.

Learners' efforts
This part compares the amount of time that experimental group learners spent in improving their listening comprehension before the experiment and during the experiment, which gives the indication of learners' efforts.The Time 1 was got from the questionnaire and the Time 2 was got from the learner diaries.The writer used paired-samples T-test to compare the effort learners made on improving their listening comprehension.

The following table shows the results:
Insert Table 8, Table 9 Here From Table 4.3.1 we can see that before the experiment, the average time of the subjects of the experimental group spent on the listening practice is 1.83 hours per week and during the experiment 2.79 hours per week.That is to say, learner almost spent one hour more than before every week.From Table 4.3.2we can see that there is no zero between -1.4430 and -0.4688 of the 95% Confidence Interval of the Difference, which showed that there is significant difference between the two variables.That is also indicated by the Significance (2-tailed), which reaches 0.000.
We can draw the conclusion that the subjects in the experimental group made more efforts than they did when they were taught under the traditional way.

Major findings
According to the students' response to the questionnaire, they were not very much satisfied with the traditional way in which they have less control of their own studies.This also reflected the trend of changes in the education field.Most students expressed their hope of having more control in their studies.They hoped that they could control the learning pace, the learning material, the learning environment, etc.That was a good sign that teachers can make use of the learners' expectations to foster their learning autonomy.Also that indicates that nowadays college students are becoming more autonomous than before as we may that they are very courageous to speak their mind to ask for more control from teachers.
As to the second research question, it can be said that the new way of teaching listening is effective in improving students' listening comprehension, since the experimental groups made much progress in their listening proficiency.Compared with the control group, they made similar progress in improving their listening comprehension and their average scores were a bit higher than those of the control groups.
The experimental group students spent more time on listening practice outside the classroom every week, which shows that they had made more efforts than before.That may indicate that students are more motivated.And they spent the class time more effectively as is shown in the interviews.From that we may say that students have become more autonomous than they were under the traditional way of listening teaching.
From the interview we can see that the teacher approved the new way of teaching listening and the learners welcomed it, too.Students in the experimental group became more active and more autonomous than before.
The learner autonomy is a very complicated thing.It involves learners' attitude, learners' capacity and the learning environment.As we can see from the literature part there are several levels of learner autonomy.Learner control is only one important part of it, so the findings of this study can only show a bit of the great iceberg.
The findings of the study may lay some theoretical foundation for the web-based distance education.In the web-based distance education, learner takes great responsibility of their own study and they may become more autonomous too.Also the findings can give some support to the course reforms that are being carried out in colleges and universities since the main aim of the course reforms is to cultivate autonomous learners and solve the problems that professional teachers are becoming comparatively fewer when the enrollment are becoming higher and higher.

Limitations
There are some limitations of the study.Firstly, the experiment lasts only two months, so the external validity or the generalizability of the experiment cannot be too much.Secondly, some other English courses, for example, the Intensive Reading, may have contributed to the improvement in learners' listening comprehension, too.Thirdly, the reliability and validity of the two listening TOEFL tests cannot be obtained.Fourthly, the sample size is too small since they are only from two classes of one university.Fourthly, due to the reality of the university, the subjects of the study are not randomly selected.

Recommendations for future research
Following are some recommendations for future research: Firstly, since the current experiment is a quasi-experimental design, the true experimental design would be better for this kind of research.Longitudinal study would be better.Secondly, this experiment is carried out only in the listening classroom, thus the external validity cannot be too much.Similar experiments can also be conducted in other classes, such as intensive reading class, extensive reading class, etc. Thirdly, as the other side of the coin, teacher autonomy deserves our studies.For example, should teachers choose textbooks by themselves?Should teachers compile or adapt textbooks?Many more studies could be carried out in this area.
The qualitative data consists of the students' responses to the open statement in the questionnaire and their responses in the interviews.The interview was recorded and transcribed.Students' responses to the open statement 23 revealed what control they expected to have.Interviews reflected their opinion about the new way of listening teaching.
6.I think many students have no ability to arrange their own study. 1 2 3 4 5 7.I can have good learning autonomy without teachers.1 2 3 4 5 8. Two hours are too long for listening classes each time and the effects are very low in the second hour.can plan the time of my listening practice.1 2 3 4 5 10.The break times are too few for a two-hour listening class. 1 2 3 4 5 11.I think the in-classroom practice time for listening is sufficient. 1 2 3 4 5 12.I need teacher's advice on listening practice out of classroom.1 2 3 4 5 13.I have the ability to plan my listening study well outside the classroom.1 2 3 4 5 14.The listening material in the text is too easy for me. 1 2 3 4 5 15.I have understood the listening material but the teacher still played the tapes again and again. 1 2 3 4 5 16.Sometimes in the listening class I couldn't understand the material but the teacher won't play the material again. 1 2 3 4 5 17.I hope I can control the times of listening to audio material so as to fully understand it.1 2 3 4 5 18.I hope I can control the speed of listening material.1 2 3 4 5 19．I hope I can control the length of pause between any passage and the next one I am asked to listen to. 1 2 3 4 5 20.I think that listening tests can improve my listening proficiency. 1 2 3 4 5 21.I couldn't understand the listening material because the speed of it is too quick.1 2 3 4 5 22.Some personal factors like fatigue can decrease the learning outcome. 1 2 3 4 5 23.I hope I can practice my listening when I am physically and emotionally fit. 1 2 3 4 5 24.In my opinion I think I need to have the control of ___________.25.Approximately I spend ________ hours on listening practice.Table 1.A brief comparison of the traditional way of teaching listening and the "zero class hour" model The traditional way The "zero class hour" model the control of the length of listening Usually the teacher controls the length of listening The students controls the length of listening the control of pace of listening Usually controlled by the teacher Controlled by the students the control of speed of listening Students can't change the speed of the listening material Students can change the speed of the listening material the control of way of listening Usually students practice listening through tape-recorder Students can decide to practice listening through tape-recorders, walkman or computers etc the control of place of listening usually in the classroom or in the language lab Students can decide to practice listening in the dormitory, classroom, or language lab etc. the control of material of listening usually the text Students can decide what to listen that interest them after they finished teachers' tasks

Table 2 .
Results of Independent-Samples T-Test of Pretest 1 and Pretest 2

Table 3 .
A comparison of the treatment of the two groups

Table 4 .
Students' responses to questionnaire items (selected) : 1= This statement is never or almost never true of me.2= This statement is usually not true of me.3= This statement is somewhat true of me.4= This statement is usually true of me 5= This statement is completely or almost completely true of me. Notes

Table 5 .
Students' responses to questionnaire items (selected) the listening material but the teacher still played the tapes again and again.2.23 Notes: 1= This statement is never or almost never true of me.2= This statement is usually not true of me.3= This statement is somewhat true of me.4= This statement is usually true of me 5= This statement is completely or almost completely true of me.

Table 6 .
The progress of the experimental group

Table 7 .
The progress of the experimental group and the control group

Table 8 .
Results of Paired-samples Statistics of Time 1 and Time 2 Time 1= average time per week on listening practice before the experiment Time 2= average time per week on listening practice during the experiment Table 9. Results of Paired Samples Test of Time 1 and Time 2 Time 1= average time per week on listening practice before the experiment Time 2= average time per week on listening practice during the experiment