An Experimental Study of Learner Autonomy in Language Learning

Chang Shih Chuan

Assistant Professor of the Department of Applied Foreign Languages

Cheng Shiu University

Abstract
Autonomous learning has greatly developed both on theory and practice ever since its
appearance in 1960s. It has become a hot issue in educational field over the past forty years and it
is important to cultivate learners’ ability of self-directed inquiry. The ultimate goal of quality-
oriented education is to promote learner autonomy. But its promotion through systematic learner
development is a relatively recent phenomenon, let alone the one proposed by Scharle and Szabo
(2000). After some modification, the program intends to make the students be aware of the
importance of responsible attitude and independent learning ability. The purpose is to develop the
students’ learning performance, give guidance on their learning strategies, and develop their
interest in language learning.
Two non-English major classes participated in the experiment. The learner development
program lasted for one semester. During the course, the SILL questionnaire and language
proficiency test were used and both the methods of interview and classroom observation were
employed. Besides, students were asked to write regular journals to keep track of their learning
experience. Comparisons were made in the following three aspects: the language performance in
the test; a change of learning strategies; stimulation of interest.
The results show that using the content of regular school classroom teaching as a meaningful
context for the development of responsibility can not only enhance the learning effect, but also
save time for the optional strategy training. According to the scores in the post-test are concerned,
the experimental group had a high mean score than the control group. The two sets of
questionnaire results show that the experimental group benefited from the program by using more
learning strategies. Besides, most of the students in the experimental group reported a higher
interest in language learning after the experiment. However, the control group showed no great
difference in all these aspects
Key words: learner autonomy, autonomous learning, strategy
Introduction
Since the 1970s, research interest in learner autonomy and independence has gained growing
attention in foreign language learning, and there has been a heated discussion about how teachers
should teach and how learners should learn in the context of language learning. The traditional way
of teaching foreign language lays emphasis on teaching instead of learning. The teacher, who
explains the text in a traditional grammar-translation way and tries to put everything into students’
head (as is called cramming method of teaching), is in complete charge of class activities. This has
severely prevented learners from thinking and learning independently. However, the present epoch

1
is expecting and demanding more from people than ever before. Just learning by what teachers
give is not enough, which means it’s important for people to be independent learners. On the other
hand, good teachers should be able to cultivate learners’ ability to do things individually and
independently. Thus, more and more teachers accept the learner-centered approach. According to
this approach, both teachers and students participate in the teaching and learning process,
accordingly, the students have more opportunities to practice the four basic language skills
efficiently. All this needs the ability of students’ autonomous learning as well as the change of
teachers’ role, and the former of which is just the ultimate goal of language learning. As a result,
growing interest in the promotion of autonomy was found since the early 1970s. In Taiwan, the
shift of research to learner-centeredness and individual learners took place mainly in the last ten
years, but the research is far behind that of the Western countries. However, autonomous language
learning has had some development in Taiwan since the communicative teaching method and some
other learner-centered approaches were introduced to English teaching during the 1980s.
Experiments have been made to foster learner autonomy. Though many teachers realized the
importance of learner responsibility and even tried to train their students to be autonomous both in
and out of class through various activities, it seems that little formal academic discussion about
learner autonomy has been held, let alone the carrying out of systematically practiced learner
development program that can be integrated into classroom teaching.
In view of the situation described above, this academic paper, on the basis of reviewing the
current theory and practices of fostering learner autonomy in foreign language, aims to explore the
feasibilities of one integrated learner development program in classroom environment. This paper,
by performing an experiment, also tries to justify the effectiveness of the program on the
development of learners’ language performance, their use of learning strategies and the stimulation
of their interest. As no consensus has ever been reached on the concept of learner autonomy, the
paper firstly reviews the definitions and studies on autonomy from different angles. It takes the
view that the understanding of learner autonomy requires a broader interpretation. It also explores
various reasons to foster learner autonomy and believes that the promotion of learner autonomy is
in urgent need and possible. In so doing, it argues that integrating learner development into regular
classroom setting is more practical in the Taiwanese context. Reasoning and arguing alone are not
enough. This paper then gives a brief introduction of the learner development program by Scharle
and Szabo (2000) and the modifications the author has made to it. An experiment was conducted to
testify the effectiveness of the modified program. Students from two classes participated in the
experiment, one as the experimental group that work with the program, while the other as the
control group under the traditional teaching method. The experiment is both qualitative and
quantitative and the results are satisfactory.
Based on the results of the experiment, this paper tries to explore the implications of the
program on foreign language teaching and puts forward some suggestions on both the program and
future research. It recommends that more experimental study is needed to examine this program
thoroughly in an effort to extend it and correct it.
I. Literature Review
1.1 Definition of learner autonomy
Learner autonomy has been a widespread phenomenon in the context of foreign language
teaching or second language teaching. It was introduced into the field of SL/FL pedagogy in the
1970s, and has drawn considerable attention of educational researchers ever since the 1980s,

2
especially in recent years.
Little (1991) defines learner autonomy as “essentially a matter of the learner’s psychological
relation to the process and content of learning. It is to be found in a wide variety of behaviors as a
capacity for detachment, critical reflection, decision-making and independent action”. In his
opinion, learner autonomy presupposes a positive attitude to the purpose, content and process of
learning and thus developing positive attitudes towards this is crucial to the success of the
development of learner autonomy and is an essential, long term aim of any learner training
program. Similarly, Benson (1996) suggests that a ‘psychological view’ of autonomy focuses on
the importance of the ‘psychological’ or ‘internal’ capacities of the learner, such as cognitive and
learning styles, motivation, attitudes, aptitude and so on. Its goal is ultimately the learners’
responsibility for their own successes and failures in learning. Gardner, in his book Establishing
Self-Access—From Theory to Practice, introduces the history of this concept and discusses the
wide range of terms related to it and used in the language field. He thinks that the concept of
learner autonomy “stemmed from debates about the development of life-long learning skills and
the development of independent thinkers, both of which originated in the 1960s” (Gardner, 2002).
Campbell defines learner autonomy as “a quality enabling a person to interact with text or
accomplish some other language task in a self-directed manner without significant or constant
assistance from others” (Campbell 2004).
Since 1990s, many educationalists shifted their focus to the social aspects of developing
autonomy in the field of foreign language teaching as a renewed interest in Vygotsky’s works took
place. Vygotsky’s sociocultural theory suggests that development depends on interaction with
people and the tools that the culture provides to help form their own view of the world. Therefore,
learner autonomy is not only concerned with the individual. It can only be developed through
social and collaborative learning. Benson (1996) suggests that “greater control over the learning
process, resources and language cannot be achieved by each individual acting alone according to
his or her own preferences” and Lee (1998) calls for “a supportive environment” too. He believes
that interaction, negotiation, collaboration, etc. are important factors in promoting learner
autonomy. However, individual autonomy is still the focus of learner autonomy in the West until
recent years. Social views of autonomy are believed to exist in what is called collectivist societies.
Perhaps the most often quoted definition is still that of Holec (1981), who defines autonomy
as “the ability to take charge of one’s own learning”. To take charge of one’s own learning is to
have, and to hold, the responsibility for all the decisions concerning all aspects of this learning, i.e.
1) determining the objectives; 2) defining the contents and progressions; 3) selecting methods and
techniques to be used; 4) monitoring the procedure of acquisition...; 5) evaluating what has been
acquired. This definition adequately covers the man areas of the learning process in which one
might expect the autonomous learner to exercise control. Besides, Holec believes that the ability to
take charge of one’s own learning is not inborn but must be acquired in a systematic, deliberate
way. His view that autonomy needed to be encouraged led to the development of a wide range of
techniques and procedures known most commonly as “learner training”, or “learning to learn”.
Complex as the concept is, there are at least three generally accepted senses of autonomy in
language teaching. They are as follows: (1) Students should take responsibility for their own
learning; (2) Teachers, courses and institutions influence the development of this responsibility; (3)
Learner autonomy is a goal of education that learners, teachers and institutions should work
together to achieve.

3
1.2 Description of learner autonomy
The different definitions of learner autonomy stress on different aspects of the term and lead
to a partial understanding of the concept and often confuse the language teachers with their stress
on different dimensions. Therefore, what is needed is a broader description. On a general note, the
following thirteen aspects of learner autonomy, adapted from Sinclair (1996), appear now to have
been recognized and broadly accepted: 1) Autonomy is a construct of capacity; 2) Autonomy
involves a willingness on the part of the learner to take responsibility for their own learning; 3)
The capacity and willingness of learners to take such responsibility is not necessarily innate; 4)
Complete autonomy is an idealistic goal; 5) There are degrees of autonomy; 6) The degrees of
autonomy are unstable and variable; 7) Autonomy is not simply a matter of placing learners in
situations where they have to be independent; 8) Developing autonomy requires conscious
awareness of the learning process; 9) Promoting autonomy is not simply a matter of teaching
strategies; 10) Autonomy can take place both inside and outside the classroom; 11) Autonomy has
a social as well as an individual dimension; 12) The promotions of learner autonomy have a
political as well as psychological dimension; 13) Autonomy is interpreted differently by different
cultures.
These aspects cover almost all the dimensions of the term and make it easy to understand.
Among the above 13 aspects, what attract our attention most are those dealing with the degrees of
autonomy. Oxford (1990) insists that autonomy is not an ‘all-or-nothing’ concept. Nunan (1988)
also says autonomy is not an absolute concept, and there are degrees of autonomy.
To conclude, whether learner autonomy should be thought of as capacity or behavior; whether it
is characterized by learner responsibility or learner control; whether it is a psychological
phenomenon with political implications or a political right with psychological implications; and
whether the development of learner autonomy depends on a complementary teacher autonomy, it is
clearly not something which learners achieve without having the opportunities to learn or be
taught, neither is it an all or nothing concept. It is something which is achieved over time with
practice, experience and support, and, possibly with some degree of maturity. Teachers should take
into account of all these aspects in facilitating students to be autonomous.
1.3 Significance of developing learner autonomy
1.3.1 The inadequacy of formal instruction
In essence, the purpose of developing learner autonomy is to cultivate an ability of learners’
for continuing learning and life-long education. Of course, we shouldn’t doubt about the role of
formal instruction in FLA as it improves learners’ second language proficiency and change the rate
of second language learning, and provides a chance for learners to learn systematically and
efficiently. However, formal instruction is not enough for learners to meet the constant change of
our society because of some major issues. For example, it cannot provide limitless resources for
learners, and neither can it provide for learners with adequate knowledge learners need in the
future; hence developing learner autonomy is undoubtedly the continuity of formal instruction to
make up this inadequacy. Therefore, both of providing knowledge and developing learner
independence, especially the latter, become the fundamental purpose of modern FLA.
1.3.2 The need of our changing society
Nowadays, our society is experiencing a series of changes. With the profound revolutions of
science and technology, our society has entered the information era in which the tendency of
globalization has been increasing. Education, supposed to meet the changes of the society and

4
serve for its further development, faces unprecedented changes, with the dramatic changes of new
technology, and patterns of education are being changing greatly. Various kinds of educational
patterns such as electronic education, home-based education, community-based education, Internet
education are or will be available for people to meet their own needs, and, therefore, education will
meet its more objectives than ever before. In the meta-technological era, education assumes more
responsibilities, and the goals of education will be “education for employment, education for life,
education for the world, education for self-development and education for pleasure” (Schuller
1979). Even if the purposes mentioned above are proposed in the last century, they’re undoubtedly
still the main goals of current education, especially the purposes of education for life, for the
world, as well as for pleasure. Not only transferring and implanting knowledge, cultivating
creative thinking and life-long learning but also learning for a better life becomes the essential part
of education. In other words, in the new era, how fast one can find out new things, how much one
can enjoy learning and make his own time, especially his spare time, enjoyable seems more
essential and significant. Under this situation, the role of teachers is no longer dominant, the
relationship between teachers and students is no longer simple and the educational process
becomes more and more active and truly bilateral, in which students are the real centers. In a word,
life-long study becomes the need of the current and future society, so autonomous learning
becomes the demand of the society. Because the future society will be a ‘learning society’, in
which the human eagerness of learning and ability of autonomous learning will be substantial for
future life.
II. Learner Autonomy in Language Learning
2.1 Possibilities of learner autonomy in language learning
In most of regular Taiwanese foreign language classroom teaching, the component of learner
autonomy training is often missing or ignored because of the tight teaching schedule for
knowledge and skills. Teachers are busy explaining grammar, vocabulary and skills to pass the
exams instead of involving students taking on responsibility of their own learning and learning to
learn. It seems that learner autonomy is not indigenous to the classroom, it has no roots there, and
it has no chance of even surviving or thriving in that alien environment. However, no matter how
unfit the idea of autonomy may be in the whole-class environment, we can, if we look, find the
elements of ‘autonomy’ and ‘individualization’ even in that apparently uncongenial place.
Firstly, in recent years, teachers’ beliefs about learning and learners and about themselves as
functioning individuals within the role of a teacher changed a lot. Aiming to teach students enough
language items to pass the examination will obviously have different implications for the approach
to teaching compared with seeing learning a new language as a lifelong process. To view learners
as resisters, receptacles and raw materials is more teacher dominated than to view them as partners,
individual and democratic explorers. In current Taiwanese foreign language teaching context,
many more teachers tend to favor the learner-centered approach and pave the right way for the
promotion of learner autonomy.
Integrating autonomy in university classroom is possible because college students have
much more advantages in learning. Aside from the external factors, they are internally more
prepared for autonomous learning, both physically and psychologically. By “physically prepared”,
it means that they are mature enough to take responsibilities for whatever they intend to do and
they are competent enough to use metacognitive strategies to improve their autonomous learning.
Besides, the spreading use of CALL has been seen as an effective way to enhance language

5
learning. Studies seem to show that CALL aid language teaching and learning in terms of
increased motivation, interest and autonomy. The practice of incorporate multi-media and
computer technology into language instruction opens up a new horizon for EFL instructors to
improve the overall quality of language instruction.
In short, the impact of cultural context does not necessarily mean that Taiwanese foreign
language learners will not take on responsibilities for their own learning. With the stronger
advocacy for quality education in Taiwan, the components of learner autonomy will spread widely
in classroom teaching.
2.2 The learner development program by Scharle and Szabo
Scharle and Szabo (2000) choose to integrate responsibility development into the regular
classroom teaching because they think that students’ autonomy developed most when autonomy
training are integrated into regular classroom activities in an informal, natural way. Various
classroom activities are designed to help learners to realize the importance of their contribution
and develop their abilities to take charge of their own learning. This is a gradual process and
consists of mainly three phases:
1) Raising awareness is the starting point. This stage is designed to present new viewpoints
and new experiences to the learners and encourage them to bring the inner processes of their
learning to the conscious level of their thinking. Most of the activities at this stage are rather
tightly structured and controlled by the teacher because learners are assumed not yet very
responsible and need to be told what to do.
2) Changing attitudes is a slow process to practice what they discovered at the previous
stage. They need to be conscious of the strategies they apply in doing the tasks, how and why they
do things this way or that. Most of the activities at this stage are repeatable, and they tend to allow
more room for learner initiative.
3) Transferring roles to the learner requires a considerable change in classroom management
and so it may be the most demanding phase. Students are expected to take over some roles from
the teacher and enjoy the freedom of being more responsible. The activities are loosely structured,
giving a considerable amount of freedom to the students in accomplishing tasks.
In all three phases, various activities are designed to work on familiar targets of learner
development, such as motivation, learning skills, empathy, and cooperation. The novelty lies in the
systematic combination of such targets. However, if our students are directly and simply involved
in these program, even if the activities are aimed to raise their awareness of the of responsibility
and autonomy, they will surely feel uneasy, because as above discussed that Taiwanese students are
so accustomed to being listeners and receivers in EFL classroom that sudden active participation in
the teaching and learning process seems novel to them. Moreover, some of the activities are not
suitable or practical in Taiwanese teaching context, they should be carefully selected and then
integrated into main textbook. Therefore, modifications seem necessary to make it fit into the
Taiwanese foreign language teaching and learning context.
2.3 Modifications of the program
Before modifying the program, some difficulties faced by Taiwanese teachers of English
have to be taken into account. First, they are working under a rigid syllabus. Together with the
prescribed course contents, students bring into the same classroom their diverse interests, levels of
English, and goals in learning the target language. And it is difficult for the teacher to meet the
needs of all of them if all students are to learn effectively. The teachers sometimes have to slow

6
down for some students or, conversely, speed up to others. It is quite common to hear other
teachers speaking about ‘focusing on the average level of the students’ hoping that this will best
satisfy students’ needs. Second, they are in a fierce competition for the students’ time. When a
student’s time is consumed by work, outside interests and other subjects, it is difficult to consider
the requirements necessary for learning in a classroom even though a student may appear
interested in the subject content. In this context, how can the teachers manage to train their
students to be more autonomous? Nunan (1997) proposes five levels to do so: 1) Awareness:
learners are made aware of pedagogical goals, contents and strategies; 2) Involvement: learners are
actively involved in the learning; 3) Intervention: learners are encouraged to modify and adapt
their goals, learning style and strategies; 4) Creation: learners set up their own goals and plans for
self-directed learning; 5) Transcendence: learners move beyond classroom setting for independent
learning. To help the students understand the purpose of the program and learn more effectively, I
redivided the whole process of the program into two stages: the preliminary stage and the
dependent action stage.
2.3.1 The preliminary stage
The preliminary stage, also aiming to raise students’ awareness of the importance of learner
autonomy, is designed on the basis of the first phase of the original program. But it intends to
achieve more than that. It tries to get the students prepared for the oncoming changes they will
experience in their classroom, both psychologically and methodologically.
2.3.1.1 Psychological preparation
Holec (1980) writes of psychological preparation as a gradual ‘deconditioning process’ through
which the learner can free himself from many kinds of assumptions and prejudices about learning
languages, for example, that there is one ideal method and that teachers possess it; that he is not
capable of making any valid assessment of his performance, and so on. Psychological preparation
is concerned mainly with helping learners become willing to take responsibility for their learning.
Many researchers have already realized the importance of psychological willingness (Little, 1991).
Learners’ willingness to be autonomous is especially important to implement autonomous
language learning in most EFL classroom setting. It is a prerequisite for the actual occurrence of
autonomous language learning. So to promote autonomous language learning there, the first thing
to do is not to directly involve learners in actual autonomous language learning practice, but to
develop their psychological willingness to be autonomous.
2.3.1.2 Methodological preparation
Compared with the abstract arguments of psychological preparation, methodological
preparation impresses people with the practical value of particular activities, such as strategy
training. Language learning strategies were generally defined as specific actions, behaviors, steps
or techniques that learners employed to comprehend, store, remember new information, and to use
the second language (O’Malley & Chamot, 1990). Oxford (1990) expanded this definition by
saying that “language learning strategies were specific actions taken by the learner to make
learning easier, faster, more enjoyable, more self-directed, more effective and more transferable to
new situations.” In the classroom, the application of learner training often focuses on exclusively
learning strategies. Students are often induced to use strategies (usually cognitive strategies)
without any rationale as to why it might be helpful to them.
2.3.2 The action stage
Being aware of the values of autonomous language learning and equipped with strategy

7
knowledge, learners come directly to the action stage of the program, which combine the last two
phases of the original program. That means learners begin to engage in autonomous language
learning activities. All the activities at this stage are classified into three types, the individual-
centered, the group-centered, and the project-centered activities, aiming to develop students’ sense
of responsibility in learning and establish a new way of teacher-student interaction.
1) The individual-centered activities focus on individual learners and their needs. Teachers,
co-learners and other resources for learning are enlisted to facilitate the attainment of the goals of
the individuals as defined by the individual.
2) The group-centered activities focus on the needs of a particular group of learners and a
strong commitment to group learning and group processes. Individuals pursue their own learning
needs within the context of the group, referring to others for support and feedback. Much learning
occurs from interactions between group members. However, equal attention must be given to a
consideration of what different individuals want to get from the class and bring to the class; and
equal opportunity must be given to different individuals to exercise their judgment or whatever
skills and resources they do bring to the class.
3) Learning through projects is one of the most common activities in courses in all
disciplines. It can be defined as an activity in which students develop an understanding of a topic
through some kind of involvement in an actual (or simulated) real-life problem in which they have
some degree of responsibility in designing their learning activities.
The key goal of the action stage is to help students move from relying on teachers to being
able to select their own achievable goals, make and exercise plans, monitor and evaluate their
progress step by step. However, in Taiwan, teachers are usually seen as an authoritarian figure and
the teacher-student relationship is usually maintained in a hierarchical way, in which stresses are
easily produced. To promote autonomous learning, the classroom phenomenon should be kept less
hierarchical and friendly, which will make learners feel relaxed and act more actively. Littlewood
(1999) notices that Asian learners have a high level of proactive autonomy in collaboration and
communication when they are amongst equals, but not when they are in a hierarchical educational
context, where they lose this proactive autonomy and become the ‘passive’ learners. In fact, most
Taiwanese learners just behave the same way. So maintaining a less hierarchical teacher-student
relationship in Taiwanese classroom setting is critical to successful application of autonomous
language learning.
III. Methodology
3.1 Research hypotheses
Generally speaking, the fundamental hypothesis of the present experiment is that the
modified program would have positive effect on the promotion of learners’ autonomous learning
ability. The following are concrete descriptions of all the hypotheses that are supposed to work for
the experiment.
Hypothesis 1: the program can improve students’ learning effectiveness;
Hypothesis 2: the program gives students guidance in their learning strategies;
Hypothesis 3: the program helps to enhance students’ motivation.
3.2 Participants
79 students whose ages ranged from 17 to 21 and with an average age of less than 19
participated in the experiment. All the participants were freshmen from the two classes, who were
enrolled in two different classes, covering majors in Biological Technology and Mechanical

8
Engineering. The experiment lasted for one semester. Both the two classes used the same textbook
under the guidance of the same teacher and followed the same syllabus. The only difference was
that one class was under the traditional grammar-translation teaching approach (Control Group),
and the other was under the learner development program (Experimental Group).
The background information about the participants was provided in Table 3.1, including
their age, gender, and major, etc.
Table 3.1 Background Information about the Participants
Age Gender Major
items Biological Mechanical
Max Min Mean Male Female Total
Technology Engineering
EG 20 17 18 29 11 40 0 40
CG 21 17 19 27 12 39 39 0
From table 3.1, it showed that the two classes did not differ significantly on any of the
background characteristics.
Their pre-test scores, as shown in Table 3.2, revealed that the two classes were of the same
level in language performance before the experiment. The average test score of experimental group
was 64.8750, with a range of 47.00; while that for control group was 64.8462, with a range of
47.00. And the spread of scores at different levels was kept almost balanced between the two
classes of students, with the standard deviation of 11.0504 for the experimental group and 11.5429
for the control group respectively.
Table 3.2 Comparison of Pre-test Scores between Experimental Group and Control Group
N Minimum Maximum Mean Std. D Range
EG 40 37.00 84.00 64.8750 11.0504 47.00
CG 39 37.00 84.00 68.8462 11.5429 47.00
3.3 Instruments
(1) Language proficiency test. Two language proficiency test papers (one before and another
after the experiment) were randomly selected from the test paper databank. To guarantee the
reliability of the tests, the objective parts of the tests were scored by computer and the subjective
parts of the tests were scored blindly by other teachers. The pre-test and post-test scores of the two
classes of students were collected then to test the effectiveness of the leaner development program
on their language performance, Analysis of data was done with SPSS 16.0.
(2) Questionnaire. The “Strategy Inventory for Language Learning (SILL, Oxford, 1990,
version 7.0)” questionnaire was delivered to the students during their self-study class twice—at the
beginning and the end of the experiment. The SILL questionnaire consists of 50 items in six
categories. They are Memory Strategies, A1-A9; Cognitive Strategies B10-B23; Compensation
Strategies C24-C29; Metacognitive Strategies D30-D38; Affective Strategies E39-E44 and Social
Strategies F45-F50. A five-point scale is employed to assess how true the statement is, from
1，which refers to “never or almost never true of me” to 5 which means “always or almost always
true of me”. To avoid students giving conforming answers, the questionnaire was distributed and
collected by a third person, without my personal attendance, which I believe to be able to help
students feel free to give their true opinion.
(3) Interviews and observations. During the experiment, I paid special attention to two
sample students, one high achiever and one low achiever from the experimental group. I observed

9
the way they behave in class and chatted with them occasionally, either in their free time or during
the class breaks if time permitted. Besides, learner diaries and classroom observations were also
applied to find out whether there existed any changes in students’ attitude and learning behavior.
3.4 Data collection
The results of the questionnaire and the scores of pre-test and post-test were put into
computers, and SPSS 16.0 statistical analysis was used to analyze them, including the descriptive
statistics, means comparison, and paired sample T-test, when data allowed. The observations were
taken down and its subsequent themes and patterns were analyzed. The interpretation of different
information is contained within the following section.
Analysis of the two classes’ pre-test and post-test scores was conducted to test the
effectiveness of the program. Paired sample T-test was conducted between the pre-test and post-
test scores of the two classes respectively to test if there was significant difference in language
performance after the experiment between the two classes. In addition, analysis of the SILL
questionnaire was done by means of comparison and paired sample T-test, to find out what
learning strategies were used frequently by the students from the two classes, and if the two classes
held similar features in terms of learning strategy before and after the experiment.
3.5 Results and Analysis
The results of both the quantitative data and qualitative reports show that students in the
experimental group benefited a lot from this learner development program. Analyses focused on
comparisons in the following three aspects: the assessments of language achievement in the test; a
change of learning strategies; stimulation of interest.
3.5.1 The effect of the program on improving students’ language performance
To test the effectiveness of the learner development program, analysis of students’ post-test
scores has been conducted.
Table 3.3 Comparison of post-test Scores between EG and CG
N Minimum Maximum Mean Std. D Range
EG 40 46.00 88.00 71.5750 10.0534 42.00
CG 39 45.00 87.00 68.0256 11.3250 42.00
Table 3.3 showed that the students’ scores in the EG were higher than that of the CG after
the experiment. The mean score of the EG was 71.5750, with a range of 42.00; while the mean
score of the CG was 68.0256, with a range of 42.00. Their respective standard deviations were
10.0534 and 11.3250. As can be seen, students in the EG outperformed those in the CG after the
experiment.
A paired sample T-test was conducted between the pre-test and post-test scores in the two
classes to test further if there was significant difference in students’ language performance before
and after the experiment.
Table 3.4 Paired Sample T-test between Pre-test and post-test Scores in EG and CG
N Mean Std. D T df Sig.
Pre-test
EG -6.7000 11.9705 -3.560* 39 .001
&post-test
Pre-test
CG -3.1795 12.2214 -1.625 38 .112
&post-test

10
 From table 3.4, Compared with the CG, the EG achieved some improvement in language
performance after the experiment. The results of paired sample T-test showed that, there was a
mean difference of 6.7000 in the EG before and after the experiment. It suggested that, compared
with the pretest, the experimental group made significant progress in the posttest with a gain of
6.7000 in their mean score. What’s more, the t value of the EG was -3.560, exceeded the critical
value of 3.551, which was significant at the 0.001 level, reached the significance value for
difference (p< 0.05), showing that there was significant difference between the scores before and
after the experiment for the students in the EG. On the other hand, there was a mean difference of
3.1795 in the control group before and after the experiment, it showed that, the control group also
made some improvement with a gain of 3.1795 in their mean score, but the t value was -1.625,
which was significant at the 0.20 level, did not reach the significance value for difference (p=
0.112> 0.05), so the improvement demonstrated by the students of the CG was not significant. In
other words, the results seemed to claim that the EG performed better in the language performance
test than the CG after they experienced the experiment.
3.5.2 The effect of the program on the deployment of learning strategies
In formal educational context, students do not automatically accept responsibility for their
learning and not all students will spontaneously master the knowledge and personal qualities at the
outset of their learning career. Consequently, concern with the development of learner autonomy
has given rise to research into the means by which learners can be helped to acquire the insights
into language learning, the relevant attitudinal traits and, crucially, the learning strategies they need
in order to operate in an informed and self-directed manner.
In recent years, studies aimed at training learners to be better at the learning and use of
language has been growing. They all emphasize discussions about the use and values of strategies,
encourage conscious and purposeful strategy use and transfer of those strategies to other contexts,
and allow students to monitor their performance and evaluate the effectiveness of the strategies
they are using in an attempt to help learners consider the factors that affect their learning and
discover the learning strategies that suit them best. This tendency is predicted on the assumption
that if learners are conscious about and become responsible for the selection, use, and evaluation of
their learning strategies, they will become more successful language learners by improving their
use of classroom time, completing homework assignments and in-class language tasks more
efficiently, becoming more aware of their individual learning needs, taking more responsibility for
their language learning, and enhancing their use of the target language out of class. In other words,
the ultimate goal of strategy training is to empower students by allowing them to take control of
the language learning process. Therefore, to assess the effect of this program, I was not only
interested in the language learning outcomes but also interested in the process of the development
of learning skills and strategies. Comparisons were made based on the results of the two sets of
SILL questionnaires of the two classes before and after the experiment.
3.5.2.1 Comparison of the overall frequency of strategy use between EG and CG before the
experiment
Table 3.5 shows the descriptive statistics of the six types of learning strategies before the
experiment.
Table 3.5: Comparison of the overall frequency of strategy use in six SILL categories by the
students in EG and CG before the experiment

11
 EG CG
Item
Mean Std .D Mean Std .D
Memory Strategies 3.4079 .7182 3.0714 .6017
Cognitive Strategies 3.2104 .5942 3.1221 .5180
Compensation Strategies 3.3548 .5368 3.3051 .7111
Metacognitive Strategies 3.1737 .8245 3.2857 .5621
Affective Strategies 3.0240 .8316 2.9095 .4258
Social Strategies 3.3107 .7608 3.2714 .7093
Total 3.1864 3.1608
According to the pre-experiment questionnaire, the frequency of strategy use for both the
experimental group and control group was almost at the same level (Table 3.5). The total mean
were 3.1864 and 3.1608 respectively, which linked in the medium range according to Oxford’s
Profile of Results (High: mean from 3.5 to 5.0; Medium: mean from 2.5 to 3.4; Low: mean from
1.0 to 2.4). There were no great differences in any of the six categories of the frequency of strategy
use.
The frequencies analysis provided us with the strategies frequently used and less frequently
used by all the students. Before the experiment, students of the two classes reported use
compensations strategies most frequently. According to Oxford (1990), compensation strategies—
guessing when the meaning is not know, or using synonyms or gestures to express meaning of an
unknown word or expression—are the heart of strategic competence. Some researchers
(Mochizuki, 1999; Hashim & Sahil, 1994) also found that Asian students used compensation
strategies most frequently. Perhaps this is a characteristic of Asian students, trying to make up for
their lack of knowledge by other means such as paraphrasing or guessing when learning English as
a foreign language. Chamot et al. (1987) found that the powerful affective strategies are woefully
underused—reported by about 1 in every 20 language users. This was, perhaps, because few
studies have examined the frequency of use of affective strategies, and also because learners are
not familiar with paying attention to their own feelings as part of the L2 learning process. Memory
strategies, which involve the mental processes for entering new information into memory storage
and for retrieving it when needed, were the second least reported in the pre-experiment
questionnaire. Nyikos & Oxford (1987) report that although memory strategies can be powerful
contributors to language learning, university students report using memory strategies infrequently.
It might be that students simply do not use memory strategies very much, especially beyond
elementary levels of language learning. However, an alternative explanation might be that they are
unaware of how often they actually employ memory strategies.
3.5.2.2 Changing take place in employing the learning strategies between the two classes after
the experiment
After the experiment, some changes did occur in the deployment of learning strategies
between the two classes. (See table 3.6 & table 3.7)
Table 3.6 Paired Sample T-test on learning Strategy use before and after the Experiment in CG
Categories Sources Means Std. D df T Sig.
Memory Pre-test 3.0714 .6017
38 .147 .885
Strategies Post-test 3.0952 .5631

12
 Cognitive Pre-test 3.1221 .5180
38 -1.142 .186
Strategies Post-test 3.1057 .5711
Compensation Pre-test 3.3051 .7111
38 -1.359 .392
Strategies Post-test 3.1562 .4603
Metacognitive Pre-test 3.2857 .5621
38 -0.471 .645
Strategies Post-test 3.2143 .7523
Affective Pre-test 2.9095 .4258
38 1.031 .321
Strategies Post-test 2.9524 .4412
Social Pre-test 3.2714 .7093
38 -1.331 .219
Strategies Post-test 3.1135 .6838
Table 3.7 Paired Sample T-test on learning Strategy use before and after the Experiment in EG
Categories Sources Means Std. D df T Sig.
Memory Pre-test 3.0479 .6182
39 2.516 .026
Strategies Post-test 3.4524 .4051
Cognitive Pre-test 3.2104 .5942
39 1.009 .199
Strategies Post-test 3.2997 .6216
Compensation Pre-test 3.3548 .5368
39 .748 .662
Strategies Post-test 3.4232 .7396
Metacognitive Pre-test 3.1737 .8254
39 2.515 .034
Strategies Post-test 3.5817 .6112
Affective Pre-test 3.0240 .8316
39 2.766 .016
Strategies Post-test 3.5238 .4662
Social Pre-test 3.3107 .7608
39 3.010 .010
Strategies Post-test 3.6857 .3621
By examining the results of the Paired Sample t-test analysis for the change in strategy use
within the two classes in Table 3.6 and Table 3.7, it turned out that the students in the experimental
group tended to use four out of the six categories of strategies more frequently after the experiment
(Table 3.7) They were metacognitive strategies (P=0.034<0.05), affective strategies
(P=0.016<0.05), social strategies (P=0.010<0.05), and memory strategies (P=0.026<0.05). In sharp
contrast, the t-test result of the CG (Table 3.6) did not show clear changes. Below are detailed
discussions of each category.
Metacognitive strategies are extremely important in language learning. They involve
thinking about the learning process, planning for learning, monitoring of comprehension or
production while it is taking place, and self-evaluation of learning after the language activity is
completed. During the experiment, all students in the experimental group were asked to keep a
learning diary, making plans for their own English learning according to their ability, time
available, personal learning style, etc; they were also encouraged to notice and correct their own
difficulties and errors, to judge their work not only from teachers’ feedback but also from their
peer classmates’ and their own evaluation, hence the mean deployment frequency of the
metacognitive strategies in the experimental group was significantly higher than before the
program and higher than that of the control group (3.5817>3.1737; 3.5817>3.2143). Besides, the t
value of the experimental group in strategy deployment before and after the experiment, 2.515,

13
exceeded the critical value of 2.021 at the 0.05 level, showing that there was significant difference
between the strategy use before and after the experiment for this class. However, the t value of the
control group in strategy deployment before and after the experiment was -0.471, with
p=0.645>0.05, not reaching the significance value for difference. Therefore, no significant changes
in strategy use occurred in control group before and after the experiment.
The effect of anxiety in language learning is described as subtle and pervasive. However, the
affective strategies like lowering anxiety and self-encouragement are often forgotten by students.
They are so used to the teacher-centered classroom and rely on teachers’ encouragement that they
do not realize they can provide their own. During the experiment, one student once complained to
me that he was worried about his learning because he had not made any progress. I advised him
and the whole class that they had to firstly lower their anxiety level by thinking about any minor
progress as encouragement and never be discouraged by setbacks. Once they know the methods for
relaxing and lowering anxiety, they would apply these strategies to lessen their anxiety. However,
strategies directly targeted at anxiety reduction are not the only ones that help learners to calm
down. Self-encouragement and self-assessment can help learners realize when they are anxious
and indirectly reduce performance anxiety as well. To make the students more effective and
enjoyable, I not only offered students as many enjoyable classroom activities as possible, but also
tried to show them learn to make positive statements on their own learning, etc. Therefore, it was
not surprising that an increase in strategy deployment in the experimental group would occur after
the experiment. Paired Sample T-test in Table 3.7 showed that the mean frequency of affective
strategy deployment in the experimental group was 3.5238, and t value of 2.766, which was at the
significant level (p=0.016<0.05). Again, the t value of control group before and after the
experiment, 1.031，did not reach the significance value for difference (p= 0.321>0.05).
College students complain most frequently about the large amount of vocabulary they have
to deal with. It seemed that how to memorize the new words was a difficult problem for most of
them. The Paired t-test results showed that the experimental group did use the memory strategies
with a significantly higher frequency after the experiment (P=0.026< 0.05). One explanation might
be that some activities like “How do I learn word” provided students with opportunity to practice
the various means of memory strategies or mnemonics to enlarge their vocabulary. Another
explanation might be that by getting to be involved in those classroom activities, students
gradually reflected their own ways of learning and were aware of the strategies they already used.
Some of the memory strategies favored by the students in the experimental group were: connecting
the sounds of new words to an image or a picture, making a mental picture of a situation in which a
word might be used, using rhymes, physically acting out a word, and remembering new words or
phrases by remembering their location on pages, the board, etc. However, the students of the
control group, though also complain about the difficulty in memorizing new words, probably did
not know about these techniques, thus, they did not show any significant difference in using
memory strategies (p=0.885>0.05).
One major influence on learner autonomy arises from the work by the Soviet psychologist
Vygotsky. His emphasis on social relationships in the development of mental abilities and thus also
learning underlines the importance of peer support for any form of learning. Central to his theory is
the idea of “the zone of proximal development. It is the distance between the actual developmental
level as determined by independent problem solving and the level of potential development as
determined through problem solving under adult guidance or in collaboration with more capable

14
peers” (Vygotsky, 1978). The Vygotskian approach emphasizes the need for a collaborative
learning environment where learners are enabled and encouraged to interact and to give each other
support with their language learning, a public space characterized by interaction and scaffolding.
The group-centered and project-centered activities in this experimented learner development
program involve the cooperation among students. With more involvement in the activities in the
program aiming to stimulate cooperation and collaboration, students in the experimental group
were used to the new idea and appreciated value in group work. It’s obvious that students in the
experimental group tended to use the social strategies, especially the cooperative strategies, much
more frequently than before, while the control group did not show significant changes. The mean
frequency of the social strategies of the experimental group was 3.6857 after the experiment,
compared with 3.3107 before the experiment, and the t value before and after the experiment was
3.010, which exceeded the critical value of 2.704 at the 0.01 level, showing significant difference.
In contrast, the t value of the control group in the deployment of social strategies before and after
the experiment was -1.331, with p=0.219>0.05, showing no significant difference.
As for cognitive strategies and compensation strategies, Paired Sample t-test (Table 3.7)
shows that the change in the deployment frequency of the experimental group before and after the
experiment was not significant, with P=.199>0.05 for cognitive strategies, and P=.662>0.05 for
compensations strategies. The reason might be that before the experiment, the experimental group
had already used the strategy more often than the control group. After the experiment, though the
experimental group did make more progress compared with the control group, this difference was
not significant compared with the strategy deployment frequency of the experimental group before
the experiment.
3.5.3 The effect of the program on the enhancement of students’ interest for learning
Generally speaking, students in the experimental group showed great enjoyment in learning
during the experiment according to my observation. To them, learning was no longer a hard task.
Instead, it was enjoyable and interesting. Involved in a different teaching and learning setting,
students learnt with higher motivation and self-confidence, and some students were not merely
test-oriented any longer. They showed more interest in learning and participating in the activities,
thus their self-confidence in communication greatly improved. The students tended to grasp any
chances to learn and use English whereas in the past, they waited to be told about the knowledge.
They made full use of their faculties to hunt for information from different resources, including an
English encyclopedia and all kinds of English language magazines and newspapers. In addition,
they visited the library and internet, interviewed people and held discussion sessions regularly. It
seemed now that peers, teachers, other people and their life experiences were all rich learning
resources for them. One student reported in the interview that in the past he learnt English because
it was a compulsory course. He felt that he was forced to learn. Now, he thought he had learned
many things and skills such as how to do researches with others, how to present views freely in
public, how to assess both others and himself.
In short, the result of the experiment is quite satisfactory. It shows that using the content of
regular school classroom teaching as a meaningful context for the development of responsibility
and autonomy can not only enhance the learning effect, but also save time for the optional strategy
training and interest stimulation. Moreover, at the end of the experiment, when I asked the students
to vote the effect of the experiment, more than half of them voted for the conduction of this
program and some even asked for further experiments in the following semester.

15
 IV. Conclusion
4.1 Major findings of the research
Through the above quantitative and qualitative analysis, we can draw the following
conclusion about this experiment.
First, the research used six types of strategies to promote learner autonomy in the
experimental group. Specifically, they were memory strategies, cognitive strategies, compensation
strategies, metacognitive strategies, affective strategies and social strategies. After the experiment,
it turn out that the students in the experimental group tended to use four out of the six categories of
strategies more frequently, they were metacognitive strategies, affective strategies, social
strategies, and memory strategies, meanwhile, we can see that the experimental group benefited
from the program by using these learning strategies. The result also showed that the control group
didn’t make any changes in the deployment of learning strategies.
Second, before the experiment, we hypothesize that the program can improve students’
learning effectiveness, after experiencing one experimental semester, the results shows that the
learners in the experimental group have made a significantly greater achievement in language
learning than those in the control group.
Third, according to interviews and observations, most of the students in the experimental
group reported a higher interest in language learning after the experiment. However, the control
group showed no great difference in all these aspects.
4.2 Teaching implications
Findings of the present study indicate, despite some limitations, that a well-designed learner
development program does have positive effects on promoting learner autonomy. The results
provide implications for English teaching in Taiwan.
(1) Integrate learner autonomy into regular English classroom
Learner training can also be in the form of optional lectures, which introduce students to the
characteristics of autonomous learners step by step. However, as shown in this study, teachers can
explicitly and systematically integrate learner autonomy into regular classroom teaching through
various activities to promote them to be autonomous. During the course, the subjective preferences
felt by the learner are crucial for effective language learning, therefore, some kind of negotiation is
needed between the teacher and learner. Information has to be exchanged about roles and
expectations, both teachers’ and learners’ awareness of each other’s needs and resources has to be
raised and compromises have to be reached between what learners expect and want and what the
teacher feels he/she can and ought to provide.
(2) Procedures for implementing learner autonomy
Just as Brown (1994) suggests, not all learners are alike. The first step in implementing
learner autonomy is to assess the needs of the learners, including their past experience with
language learning, current language proficiency, etc. Teachers should adjust their plans and
teaching activities to the learners’ needs and wants within the domain of general teaching plans, if
it is necessary. Secondly, select or design carefully the program that is suitable for the students,
taking account of learner differences and time available. No single program fully addresses all the
needs of students. Yet if success can be measured by high academic performance and satisfaction
with oneself, then the specific program is clearly valid. Thirdly, carry out the program
systematically and explicitly. Teachers inform the students fully with the aims of the activities
included, the underlying rationale and some further suggestions on how to use them. Finally, it

16
should be noted that ongoing evaluation and revision of the program is necessary to ensure its
success. Teachers can get insightful feedback either from the learners or from their own
observations.
(3) Adopt new roles for teachers
Promoting learner autonomy does not necessarily mean a complete rejection of teacher
authority. A complete rejection of teacher authority might be dangerous. In fact, the learners need a
great deal of guidance and feedback from the teacher in order to learn to organize their work on
their own. The specific help, support and encouragement from teachers are a central element in the
process. In such case, a “teacher” means more than the traditional teacher who teaches knowledge.
As put by Cohen (2000), teachers should act as change agents in the classroom—shifting the
responsibility for learning more onto the shoulders of the students themselves, and taking on a
series of roles as diagnosticians, learner trainers, coaches, coordinators, language learners and
researchers.
4.3 Limitations of the research and expectations
It is clear that students should have a major role in their own learning from the professional
literature. Their involvement is critical to producing learning that is both lasting and important.
Consequently, teachers should facilitate them to be autonomous and independent learners. This
paper has given strong support for the claim that integrating learner autonomy promotion into
regular classroom in EFL teaching is feasible in Taiwanese foreign language teaching context. But,
it is not without its own problems and faces some obstacles.
First, this study is an attempt to investigate the impact of a learner development program on
promoting students’ autonomous leaning ability. Since both learner autonomy and learner training
are extremely complex, and all the participants in the study were non-English majors from the
same university, the findings from the research can be only certain to be applied to learners as
such. Whether it can be applied to other English learners has to be left for further research.
Second, owing to the time limit, the conduction of the program is relatively short, and the
notion of autonomy embodied in the activities is still limited. So teachers should consciously
explore and modify materials for future use. Furthermore, carrying out such kind of learner
development program makes extra demands on the teacher’s time and energy, makes advance
planning more difficult, and, as a result of the developmental nature of course structure, can add
stress.
To conclude, the research is only an attempt in fostering learner autonomy in Taiwan. Some
problems do appear, various constraints do exist, but the author has noticed changes in students,
which is an indication that students can take responsibility of their learning. Future researches can
replicate this program cross-sectionally and longitudinally with larger samples in collaborative and
comparative studies.
REFERENCES
Benson, P. 1996. “Concepts of autonomy in language learning”. In Pemberton, R., Li, E. S. L.,
Or W. W. F., Pierson, H. D. (eds.). Taking control: Autonomy in Language Learning.
Hong Kong: HKU Press: 27-34.
Brown, H. D. 1994. Principles of Language Learning and Teaching.
Englewood Cliffs: Prentice Hall Regents.
Campbell, C. 2004. Teaching Second Language Writing: Interacting with Text.
Beijing: Foreign Language Teaching and Research Press.

17
Cohen, A. D. 2000. Strategies in Learning and Using a Second Language.
Beijing: Foreign Language Teaching and Research Press.
Charnot, A. U. 1987. “The learning strategies of ESL students”. In Wenden, A.& Rubin,
J. (eds.). Learner Strategies in Language Learning. Cambridge: Prentice Hall.
Gardner, D. & Miller, L. 2002. Establishing Self-Access: from Theory to Practice.
Shanghai: Shanghai Foreign Language Education Press.
Hashim, R. A. & Sahil, S. A. S. 1994. “Examining learners’ language learning strategies”.
RELC Journal 25/2: 1-20.
Holec, H. 1980. “Learner training: meeting needs in self-directed learning”.
In Altman, H. B. & James, C.V (eds.) Foreign Language Learning: meeting
individual needs. Oxford: Pergamon
Holec, H. 1981. Autonomy and Foreign Language Learning. Oxford: Pergamon.
Lee, I. 1998. Supporting greater autonomy in language learning. ELT Journal 52/4: 282-89.
Little, D. 1991. Learner Autonomy. Definitions, Issues and Problems. Dublin: Authentik.
Littlewood, W. 1999. “Defining and developing autonomy in East Asian contexts”.
System. 20 (1):71-74
Mochizuki, A. 1999. “Language learning strategies used by Japanese university students”.
RELC Journal 30/2: 101-13.
Nunan, D. 1988. The Learner-centered Curriculum: A Study in Second Language Teaching.
Cambridge: Cambridge University Press.
Nunan, D. 1997. “Designing and adapting materials to encourage learner autonomy”.
In Benson, P. & Voller, P. (eds.). Autonomy and Independence in Language Learning.
New York: Addison Wesley Longman.
Nyikos, M. & Oxford, R 1987. “Strategies for foreign language learning and second
language acquisition”. Paper presented at the conference on SLA and FLL,
Illinois, Champaign-Urbana.
O’Malley, J. M., Chamot, A. U. 1990. Learning Strategies in Second Language Acquisition.
Cambridge: Cambridge University Press.
Oxford, R. L. 1990. Language Learning Strategies: what every teacher should know.
New York: Newbury House Publishers.
Scharle, A. & Szabo, A. 2000. Learner Autonomy. Cambridge: Cambridge University Press.
Schuller, T. 1979. Recurrent Education And Lifelong Learning.
New York: Kogan Page, London / Nichols Company.
Sinclair, B. 1996. “Materials design for the promotion of learner autonomy: how
explicit is “explicit”?” In Pemberton, R., Li, E. S. L., Or W. W. F., Pierson, H. D.
(eds.). Taking control: Autonomy in Language Learning, Hong Kong: HKU Press
Vygotsky, L. S. 1978. Mind in society. USA. Harvard

18