1
INFLUENCE OF STUDENT’S PERCEPTION ON TEACHERS’ BEHAVIOR ON
SELF-ESTEEM AND SELF-EFFICACY IN BETWEEN CLASS ABILITY
GROUPING
PRIHADI KUSUSANTO
A dissertation submitted
in partial fulfillment for the award
of the degree of Master of Education
(Educational Psychology)
Faculty of Education
Universiti Teknologi Malaysia
MAY 2009
3
To my beloved self
4
ACKNOWLEDGMENT
This thesis is not just meant to get the Master degree. This paper meant more
than that to me. It is the matter of dignity, self-actualization, and a brand new start of
an entity called life. I would like to express my sincere gratitude to God the
Almighty that always open a window when all the doors seemed to be closed.
I would like to thank my parents, Bambang & Santi Koestoyo for continuing
their responsibility as parents while they should have been retired for quite a while.
To my grandfather, Mr Prihadi, who reconfirmed to me that I was doing the right
thing. To my dearest family, Dian, Ditta, and Dion, whom I did this work for.
My special gratitude to the most important people behind the whole story:
My dearest supervisor Dr Azlina Kosnin, for giving more than a supervisor should.
To all members of the examiners panel: Dr Yeo Kee Jiar, Dr Noraini Othman, and
Dr Othman Johan, for an exciting viva. To Prof Dr Alias M. Jusuf (Fac of Science)
& Ass Prof Dr Noraini A.Rahman (Fac of Bioscience) for being my parents, eyes,
and ears in this very place, and to my local classmates who helped me a lot: Zac,
Man, Leha, Vijaya, Nafi, and others. May God pay all of their great deeds with
multiplied price, here and hereafter.
Finally yet importantly, a sincere gratitude to all my dearest friends in IDUTM08, especially my neighbours, Andre, Aulia, Salim, and Bobbie for keeping me
as ‘human’ as I can be. My Malaysian fellas , Mani Chow and Tan Meng Kuan for
helping me with the translation process; my office-mates Haikal, Emansa, Kean
Siong, Syaiful, Iwa, Indra, Ario, and others, who share times, rides, (and money)
whenever I was in need. Of course, to all international friends who shared with me
all of the wonderful ‘alien’ moments in this very country: Mazyiar & Shabnam,
Majid & Homa, Remaz & Azeeha, Sajjad, Shauqi, Salahudin, Saman, Fereshte,
Oldooz, Nora, Melvin, Chris, Alfred, Abdulazeez, Ayo, Obina, Alfred, and others.
All of them have taught me about life and share their examples on how a life should
be lived.
5
ABSTRACT
This is a quantitative study looking at the difference in self-efficacy, selfesteem, and perception on teachers’ behavior between the students from the high and
low-performers’ classes in the Between Class Ability Grouping school system. This
current study is also looking at the influence of students’ perceptions on teachers’
behavior on the students’ self-esteem and self-efficacy. A total of 302 form two
students (153 high-performers and 149 low-performers) from four secondary schools
in Pasir Gudang, Johor were selected via purposive sampling method to participate
in this study. Self-esteem was measured by the Culture-Free Self-Esteem Inventory
(Battle, 1992) while self-efficacy was measured by the General Self-Efficacy Scale
(Jerusalem & Schwarzer, 1981). The scale for perception on teacher behavior was
self-developed by the researcher, measuring two aspects of teacher behavior:
supporting and controlling behaviors. All scales were translated into the Malay
language and internal consistency reliability scores (Cronbach’s alpha) were found
to be above 0.6 for all scales. Results show that students from high performers’
classes scored significantly higher in self-esteem and self-efficacy compared to the
students from low performing classes. This study also discovered a significant
difference in perception on teachers’ behavior between the students in the high and
low performers’ classes. Students from high-performers’ classes perceived their
teachers to be supportive while students from low-performers’ classes perceived
their teachers to be more controlling. Amongst the students from high-performers
’classes, self-esteem were significantly influenced by perception on teachers’,
supporting behavior and self-efficacy was significantly influenced by perception on
teachers’ controlling behavior. Amongst students from low-performers ’classes, selfesteem were significantly influenced by perception on teachers’, controlling
behavior and self-efficacy was not significantly influenced by perception on
teachers’ behavior, either controlling or supporting.
6
ABSTRAK
Kajian ini berbentuk kuantitatif dan bertujuan untuk meninjau perbezaan
dalam konsep kendiri, efikasi kendiri dan persepsi terhadap tingkahlaku guru antara
pelajar-pelajar dari kelas cemerlang dengan pelajar-pelajar dari kelas lemah dalam
sistem persekolahan yang mengumpulkan pelajar ke dalam kelas-kelas berbeza
mengikut pencapaian akademik. Kajian ini turut melihat pengaruh persepsi pelajar
terhadap tingkahlaku guru ke atas konsep kendiri dan efikasi kendiri pelajar.
Seramai 302 pelajar tingkatan dua (153 pelajar cemerlang dan 149 pelajar lemah)
dari empat sekolah menengah di Pasir Gudang, Johor dipilih menggunakan kaedah
rawak bertujuan untuk terlibat dalam kajian ini. Penghargaan kendiri telah diukur
menggunakan Culture-Free Self-Esteem Inventory (Battle, 1992) manakala efikasi
kendiri diukur menggunakan General Self-Efficacy Scale (Jerusalem & Schwarzer,
1981). Alat ukur bagi persepsi terhadap tingkahlaku guru telah dibina oleh pengkaji,
dan ianya mengukur dua aspek tingkahlaku guru: tingkahlaku menyokong dan
mengawal. Semua alat ukur telah diterjemahkan ke dalam Bahasa Malaysia dan
markat bagi kebolehpercayaan ketekalan dalaman (alfa cronbach) bagi setiap
pembolehubah melebihi 0.6. Dapatan menunjukkan pelajar dari kelas cemerlang
mempunyai penghargaan kendiri dan efikasi kendiri yang secara signfikannya lebih
tinggi berbanding pelajar dari kelas lemah. Kajian ini juga mendapati perbezaan
yang signifikan dalam persepsi terhadap tingkahlaku antara pelajar kelas cemerlang
dan kelas lemah. Pelajar kelas cemerlang mempunyai persepsi bahawa guru-guru
mereka lebih menunjukkan tingkahlaku menyokong manakala pelajar-pelajar dari
kelas lemah berpendapat guru-guru mereka lebih menunjukkan tingkahlaku
mengawal. Dalam kalangan pelajar-pelajar kelas cemerlang, penghargaan kendiri
dipengaruhi oleh persepsi terhadap tingkahlaku menyokong dan efikasi kendiri
dipengaruhi oleh persepsi terhadap tingkahlaku mengawal. Dalam kalangan pelajarpelajar kelas lemah pula, penghargaan kendiri dipengaruhi oleh persepsi terhadap
tingkahlaku mengawal manakala efikasi kendiri tidak dipengaruhi oleh persepi
terhadap tingkahlaku guru, sama ada tingkahlaku menyokong atau mengawal.
7
TABLE OF CONTENTS
CHAPTER
1
TITLE
PAGE
DECLARATION
DEDICATION
ACKNOWLEDGMENTS
ABSTRACT
ABSTRAK
ii
iii
iv
v
vi
TABLE OF CONTENTS
vii
INTRODUCTION
1
1.1
Introduction
1
1.2
Background
3
1.3
Problem Statement
5
1.4
Research Objectives
6
1.5
Research Questions and Null Hypotheses
6
1.6
The Importance of the Study
8
1.7
Scope and Limitation of the Study
9
1.8
Conceptual Framework
10
1.9
Theoretical Framework
10
1.10
Definition of Key terms
12
1.10.1
Between-Class Ability Grouping
12
1.10.2
Self Esteem
12
1.10.3
Self-Efficacy
13
1.10.4
Students’ perception on teachers’ behavior
13
8
1.10.4.1
1.10.4.2
1.11
2
13
Students’ perception on teachers’ supporting
behavior
14
Conclusion
14
LITERATURE REVIEW
15
2.0
Introduction
15
2.1
Definitions and conceptions of the key variables
15
2.1.1 BCAG
15
2.1.2 Self-esteem
18
2.1.3 Self-efficacy
20
2.1.4
22
2.2
Students’ perception on teacher’s controlling and
supportive behavior
Previous related studies
24
2.2.1 BCAG and students’ self-esteem.
24
2.2.2 BCAG and self-efficacy
26
2.2.3 BCAG and students’ perception of teacher’s
behavior
28
2.2.4
Influence of perception of teachers’ behavior on
self-esteem and self-efficacy.
30
Conclusion
32
2.3
3
Students’ perception on teachers’ controlling
behavior
METHODOLOGY
33
3.0
Introduction
33
3.1
Design of Study
33
3.2
Population
34
9
4
3.3
Sample
35
3.4
Place and Time of Study
36
3.5
Instrumentation
36
3.5.1 Self-Esteem
37
3.5.2 Self-Efficacy
37
3.5.3 Perception on teachers’ controlling and supporting
behavior
38
3.5.3.1
Perception on teachers’ controlling behavior
3.5.3.2
Perception on teachers’ supporting behavior
3.5.3.3
Development of the Instrument
39
40
40
3.6
Pilot Study
43
3.7
Study Procedure
43
3.8
Data Analysis
44
3.9
Conclusion
45
RESULTS
46
4.0
Introduction
46
4.1
Reliability of Scales
47
4.2
Demographic Data Analyses
47
4.3
The difference of self-esteem between students from
high and low performers’ classes.
49
4.4
The difference of self-efficacy between students from
high and low performers’ classes.
50
4.5
The difference of perception on teachers’ controlling
behavior between students from high and low
performers’ classes.
4.6
The difference of perception on teachers’ supporting
behavior between students from high and low
performers’ classes.
51
52
10
4.7
4.8
4.9
The difference between perception on teachers’
supporting behavior and teachers’ controlling
behavior among students within the high-performers
classes.
The difference between perception on teachers’
supporting behavior and teachers’ controlling
behavior among students within the low-performers
classes.
The influence of students’ perception on teachers’
controlling and supporting behavior on students’ selfesteem.
4.9.1 The influence of students’ perception on teachers’
controlling and supporting behavior on students’ selfesteem in low-performers’ classes
4.9.2 The influence of students’ perception on teachers’
controlling and supporting behavior on students’ selfesteem in high-performers’ classes.
4.9. 3 The influence of students’ perception on teachers’
controlling and supporting behavior on students’ selfesteem in both groups.
4.10
The influence of students’ perception on teachers’
controlling and supporting behavior on students’ selfefficacy.
4.10.1 The influence of students’ perception on teachers’
controlling and supporting behavior on students’
self-efficacy in low-performers’ classes
4.10.2 The influence of students’ perception on teachers’
controlling and supporting behavior on students’
self-efficacy in high-performers’ classes
4.10.3 The influence of students’ perception on teachers’
controlling behavior on students’ self-efficacy in
both groups.
4.11
5
Conclusion
53
54
55
56
57
58
60
60
61
62
64
DISCUSSION
65
5.0
65
Introduction
11
5.1
Summary of the Findings
66
5.2
The difference of self-esteem between students from
high and low performers’ classes.
67
5.3
The difference of self-efficacy between students from
high and low performers’ classes.
70
5.4
The difference of students’ perception on teachers’
controlling behavior between the students in the high
and low performers’ classes
5.5
The difference of students’ perception on teachers’
supporting behavior between the students in the high
and low performers classes
5.6
The difference between perception on teachers’
supporting and controlling behaviors among students
within the high-performers classes.
5.7
The difference between perception on teachers’
supporting and controlling behaviors among students
within the low-performers classes.
5.8
The influence of students’ perception on teachers’
controlling and supporting behavior on students’ selfesteem
5.9
The influence of students’ perception on teachers’
controlling and supporting behavior on students’ selfefficacy
5.10
Practical Implications
83
5.11
Suggestion for Future Researches
85
5.12
Conclusion
86
73
74
75
77
79
80
REFERENCES
87
APPENDICES
91
LETTER OF APPROVAL
91
SCALES
92
VALIDATION LETTER
97
12
CHAPTER 1
INTRODUCTION
1.1
Introduction
It has been a challenge to deliver education as fair and as effectively as
possible to everybody in the country. The need of education might be the same in
every child, but children differ in many aspects, in term of knowledge, skills,
interest, motivation, ability, and many other aspects. This situation leads to the
challenge on how education should be effectively delivered, to fulfill the general
need of education of various children.
Providing the best education to every individual has been a difficult task for
educators in every country, since individual differences led to differences in
learning. Learning could be varied in terms of method, pace, preference, and many
others; hence, a suitable strategy is needed for a successful education system. One of
the known methods to be more effective in delivering education is to group students
based on their common attributes (Hallam, Ireson, & Davies, 2002).
13
There are several types of existing techniques to group students, and all of
them are being used by different education system in different countries. Most of
them are grouping students based on their abilities, because students with similar
abilities were believed to have similar ways of learning (Kulik J. , 1992)
Out of many grouping methods, one of them has been commonly practiced in
Malaysia. It is called Between-class Ability Grouping (BCAG). It is a practice of
grouping students in separate classes according to their level of ability, which refers
to their prior academic achievements (Slavin R. E., 2006). While other types of
grouping might group students in different classes for every subject, based on the
students’ ability in each subject, in BCAG, students are placed in classes based on a
test of their general ability, where they will remain in their streamed class for most
subjects.
BCAG is practiced based on the assumption that individuals have a certain
level of general intelligence that might predict their performance across all subjects,
and can be measured by objective tests (Hallam, Ireson, & Davies, 2002). According
to Kulik (2004), typical students in a non-grouped class might gain one year on a
grade-equivalent scale in a calendar year, whereas the typical students in BCAG
would gain 1.3 years; and the effects were positive for high, middle, and low groups
in cross grade program.
In BCAG, teachers face students from similar levels of ability at a time, and
it certainly would make it easier for the teacher to deliver the subject. School
authorities are seeing that BCAG is one of the methods to escalate the academic
achievement (Hallam, Ireson, & Davies, 2002). In other words, the aim of BCAG is
to enhance their academic achievement. As an instructional method, BCAG is
considered effective in order to gain the maximum result of academic achievement
out of the best students (Kulik, 2004).
While cognitive aspects that lead to academic achievement might be the
main positive factor of practicing BCAG, some studies noted its effects to noncognitive aspects of the students. This study would like to see the relevancy of
14
practicing the grouping system to the non-cognitive aspects of the students. Noncognitive aspects mentioned are self-esteem, self-efficacy, and the students’
perception on teachers’ behavior.
1.2
Background
Some research findings have noted that BCAG affects students’ socio
emotional domains, students feel stigmatized being assigned to low-achievers
classes, and such feelings affect their academic achievement (Slavin, 1987). It was
noted that teachers assigned in lower-achievers classes seem to have lower
expectations of the students than teachers assigned in higher-achievers classes
(Good, 1981), and it was reported that there is a significant relationship between
teachers’ expectation and students’ academic achievements; the lower the
expectation, the lower the academic achievement and vice versa (Rosenthal &
Jacobson, 1968).
Social cognitive learning theory supports the non-grouped class system,
where high-achievers might give a good model for the lower achievers. On the other
hand, BCAG limits the good model for the lower achievers because they are not put
together in a classroom. According to Slavin (1990, 2006), any educational system
should avoid BCAG, because there are no research evidences that the system would
significantly improve student academic achievement. Moreover, when labeled as
‘lower-achievers’, students are far more likely to become delinquent and truant and
drop out of school compared to the other students (Goodland, 1983; Oakes, 1985)
In Malaysia, BCAG is not a formal government policy, or in other words, the
Ministry of Education of Malaysia had never encouraged any schools to practice the
BCAG in the classrooms. Nevertheless, BCAG is a common practice and applied to
most of the schools in Malaysia. Formally the schools give different names to their
classes. The name given could be nominal terms like the name of flowers or national
15
heroes. However, it is almost common for Malaysians to call the high-performers’
class as Kelas Hadapan (Front-Class), and the low-performers’ class as Kelas
Belakang (Rear Class). Myers (2008) argued that students’ disposition of being
placed in high or low-performers’ classes might serve as prior information for
teachers which determine the level of teacher’s expectation; a phenomenon called
correspondence bias would lead the teachers to put higher expectation towards
students from the high-performers’ classes and lower expectation towards students
from the low-performers’ classes. This argument was based on the theory of
attribution (Heider, 1958; Ross, 1977).
A study by Oakes (1992) discovered that because of their expectations,
teachers assigned in low-performers’ classes are likely to focus on control the
students’ behavior in order to avoid disciplinary problems, while in high-performers
classes, teachers are likely to center on supporting students to get higher academic
achievements. This different kind of teachers’ behavior would be subjectively
perceived differently by students from different classes. A qualitative study by
Goods (1981) discovered that teachers assigned in the high-performers classes are
more likely to support their students in improving academic achievements, while
teachers assigned in the low performers’ classes are more likely to control the
students in order to reduce disciplinary problems.
According to the theory of symbolic interaction (Cooley, 1912; Mead, 1934),
students would subjectively interpret their teacher’s behavior as a main source of
data about themselves, without knowing that the teacher’s behavior was the product
of a correspondence bias. Eventually, these perceptions would affect the way the
students’ self-esteem and self-efficacy.
Self-efficacy is defined as an individual’s belief about his/her competence to
bring about a desired outcome in a particular situation (Bandura, 1997; Santrock,
2005; Von Der Haar, 2005). Compared to self-esteem, which is concerned with
judgments of self-worth, self-efficacy is concerned with judgments of personal
capabilities (Woolfolk, 2007). High self-efficacy in academic achievement will help
a student to believe that he/she has self-control of the outcome. It will help him/her
16
to study harder and avoid bad habit that might delay or distract them from having a
good achievement. Low self-efficacy in academic achievement, in the opposite, will
distract a student from trying harder to achieve high goals, because the student does
not believe that his/her effort might give them control of the outcome.
Self-efficacy can be manipulated, and a subtle manipulation of self-efficacy
can affect behavior (Levy, 1996). Manipulating self-efficacy to students could be
done by giving them information about themselves. An obvious placement in a
grouped class might manipulate the students’ self-efficacy. When they are often
exposed to a fact that they are part of a high-performers group, their self-efficacy in
academic achievement might be upgraded and they will gain more beliefs that they
are able to achieve high academic performance. The opposite situation might happen
to the students in the low-performers’ classes, where their self-efficacy in academic
achievement might be degraded and reach a point where they do not believe that
they can control their outcome by putting more effort; hence they will not even try
any harder.
1.3
Problem Statement
BCAG has been a common practice in Malaysia. Although numerous studies
have been done about BCAG, and discovered both positive and negative effects
from the system, not so many has been done in Malaysia. In BCAG, some classes
would be considered as high-performers’ classes and some classes would be
considered as low-performers’ classes. Students from both classes might perceive
their teachers’ behavior differently; hence, the influence of those perceptions might
be different from one type of class to another. This study is focusing on the
influence of students’ perception on teachers’ behavior on the students’ self-esteem
and self-efficacy and the difference of self-esteem and self-efficacy between
students from high and low-performers’ classes in BCAG.
17
1.4
Research Objectives
The primary objectives of the research are as follows:
1. To identify the difference in self-esteem between the students from
high and low performers classes
2. To identify the difference of self-efficacy between the students in
high and low performers classes
3. To identify the difference of students’ perception on teachers’
controlling behavior between the students in high and low
performers classes
4. To identify the difference of students’ perception on teachers’
supporting behavior between the students in the high and low
performers classes
5. To identify the difference between perceptions on teachers’
supporting behavior and teachers’ controlling behavior among
students within the high-performers classes
6. To identify the difference between perceptions on teachers’
supporting behavior and teachers’ controlling behavior among
students within the low-performers classes
7. To identify the influence of students’ perception on teachers’
controlling and supporting behavior on students’ self-esteem.
8. To identify the influence of students’ perception on teachers’
controlling and supporting behavior on students’ self-efficacy.
1.5
Research Questions and Null Hypotheses
18
The research questions and null hypotheses for each question are reported in
the table 1.1. Research questions number 7 and 8 needed 3 hypotheses for each of
the question due to the possibility of having difficulty during statistical analysis.
Table 1.1: Research Questions and Hypotheses
No
1
2
3
4
5
6
7
Research Questions
Is there any significant difference in
self-esteem level of the students in
the high and low performers’ classes
of BCAG?
Is there any significant difference in
self-efficacy level of the students in
the high and low performers’ classes
of BCAG?
Is there any significant difference in
perception on teachers’ controlling
behavior between the students in
the high and low performers’ classes
of BCAG?
Is there any significant difference in
perception on teachers’ supporting
behavior between the students in
the high and low performers’ classes
of BCAG?
Is there any significant difference
between perception on teachers’
supporting behavior and teachers’
controlling behavior among students
within the high-performers’ classes?
Is there any significant difference
between perception on teachers’
supporting behavior and teachers’
controlling behavior among students
within the low-performers’ classes?
Null Hypotheses
There is no significant difference in
self-esteem level of the students in the
high and low performers’ classes
There is no significant difference in
self-efficacy level of the students in the
high and low performers’ classes
There is no significant difference in
perception on teachers’ controlling
behavior between the students in the
high and low performers’ classes
There is no significant difference in
perception on teachers’ supporting
behavior between the students in the
high and low performers’ classes
There is no significant difference
between perception on teachers’
supporting behavior and teachers’
controlling behavior among students
within the high-performers’ classes.
There is no significant difference
between perception on teachers’
supporting behavior and teachers’
controlling behavior among students
within the low-performers’ classes.
(a) There is no significant influence of
Is there any significant influence of
students’ perception on teachers’
students’ perception on teachers’
controlling and supporting behavior on
controlling and supporting behavior
students’ from low-performers classes’
on students’ self-esteem?
self-esteem.
(b) There is no significant influence of
students’ perception on teachers’
19
controlling and supporting behavior on
students’
from
high-performers
classes’ self-esteem.
(c) There is no significant influence of
students’ perception on teachers’
controlling and supporting behavior on
students’ from both high and lowperformers classes’ self-esteem.
8
1.6
(a) There is no significant influence of
students’ perception on teachers’
controlling and supporting behavior on
students’ from low-performers classes’
self- efficacy.
(b) There is no significant influence of
Is there any significant influence of
students’ perception on teachers’
students’ perception on teachers’
controlling and supporting behavior on
controlling and supporting behavior
students’
from
high-performers
on students’ self-efficacy?
classes’ self- efficacy.
(c) There is no significant influence of
students’ perception on teachers’
controlling and supporting behavior on
students’ from both high and lowperformers classes’ self- efficacy.
The Importance of the Study
Information on the influence of students’ perception on teachers’ behavior on
students’ self-esteem and self-efficacy might enrich the literacy of BCAG and
students-teachers relationships. Moreover, the teachers, schools, and other
educational stakeholders, including the parents would be aware of the differences
that might come up as results of the influence of teachers’ behavior, especially under
the BCAG environment. Such knowledge might help the educators to have suitable
instructional methods to each type of the class for the benefit of every student.
20
1.7
Scope and Limitation of the Study
The study is focusing on the students of public secondary schools (Sekolah
Menengah Kebangsaan / SMK), which are located in the area of Permas Jaya,
district of Pasir Gudang, state of Johor, Malaysia, in term of:
1. Self-Esteem
2. Self-Efficacy
3. Perception in Teachers’ Controlling Behavior
4. Perception in Teachers’ Supporting Behavior
The study is also focusing on the difference within students in highperformers and low-performers’ classes in term of:
1. Perception of teachers’ controlling behavior
2. Perception of teachers’ supporting behavior
Another focus of this study is on the influence of students’ perception of
teachers’ controlling behavior and supporting behavior on self-esteem and selfefficacy.
This study does not control the extraneous variables that might involve, such
as students’ physical conditions, social economic status, gender, or any other
dispositional differences that might influence their levels in term of the variables
mentioned in this study.
21
1.8
Conceptual Framework
Perception on Teachers’
Controlling Behavior
Between
Class
Ability
Grouping
HighPerformers
Class
LowPerformers
Class
Perception on Teachers’
Supporting Behavior
Differences
Influence
Self-Esteem
Self-Efficacy
Figure 1.1 Conceptual Framework
This study looks at the difference between students from high and low
performers’ classes in self-esteem, self-efficacy, perception on teachers’ controlling
behavior, and perception on teachers’ supporting behavior. It is also looking at the
influence of students’ perception on teachers’ controlling and supporting behavior
on self-esteem and self-efficacy.
1.9
Theoretical Framework
The BCAG is merely an instructional strategy, aimed at creating conducive
learning environment for students of quite similar performance level. This study is
investigating the socio emotional aspects of the students by looking at the
differences between highest and lowest group in BCAG in term of self-esteem, selfefficacy, and perception on teachers’ behavior.
22
According to symbolic Interaction Theory by Cooley (1912), a person would
subjectively interpret others’ behavior as a main source of data about themselves,
without knowing that the others’ behavior was the product of a fundamental
attribution error. In other words, Cooley’s theory stated that it was not others’
behavior that determined one’s self-esteem or self-efficacy, it is one’s perception of
others’ behavior toward themselves that determined one’s self-esteem or selfefficacy.
Figure 1.2 illustrates the flow of how students from high and lowperformers’ classes of BCAG interpret the teachers’ behavior as the teachers’
expectation towards the students, which might influence their self-esteem and selfefficacy.
Perception on Teachers Behavior
of students under BCAG system:
Theory of
1. Controlling behavior
2. Supporting behavior
Symbolic
Interaction
(Cooley, 1912)
Self-esteem
student’s value of
characteristic, ability, and
behavior based on his /
her own evaluation.
Self-efficacy
student’s belief about his /
her personal general
competence.
Figure 1.2: Theoretical Framework
23
1.10
Definition of Key terms
Several key terms will be conceptually and operationally defined in this sub-
chapter. Those key terms are BCAG, self-esteem, self-efficacy, students’ perception
on teacher’s behavior, academic achievement, and students’ preference of ability
grouping system.
1.10.1 Between-Class Ability Grouping
The definition of BCAG is the practice of grouping students in separate
classes according to ability level (Slavin R. E., 2006). Some schools have their own
standards, but for most of the schools, ability is often measured by the academic
performance. (Kulik J. A., 2004).
In this study, BCAG is defined as a system of grouping students in
Malaysian Secondary Schools, based on students’ previous academic achievements.
1.10.2 Self Esteem
Self Esteem is an individual’s evaluation of his/her self worth. (Von Der
Haar, 2005) It is also defined as the value each individuals place on own
characteristics, abilities, and behavior. (Woolfolk, 2007). In all cases, self-esteem
results from an evaluation of oneself. (Larsen & Buss, 2008). According to Larsen &
Buss (2008), self-esteem measures of many areas are moderately correlated. A
person with high self-esteem in one area also tends to have high self-esteem in the
other areas as well.
In this study, self-esteem is defined as a student’s value of characteristic,
ability, and behavior based on his / her own evaluation.
24
1.10.3 Self-Efficacy
Self-Efficacy is defined as a person’s belief about one’s own personal
competence in a particular subject and situation (Von Der Haar, 2005; Woolfolk,
2007). Works of Bandura explained self-efficacy as beliefs in one’s capabilities to
organize and execute the courses of action required to produce given attainments
(Bandura, 1997).
In this study, self-efficacy is defined as a student’s belief about his / her
personal competence in academic performance and other areas related to academic
performance and his / her being placed at the front class or rear class in BCAG
system.
1.10.4 Students’ perception on teachers’ behavior
Students’ perception towards their teacher’s behavior defined as students’
assumption on what their teachers would expect them to be like. Perceptions are
selective and are often a result of the distortions engendered by motives, goals,
attitudes, and defense mechanisms (Bruner & Goodman, 1947); hence teacher’s
behavior might affect how students’ perception about the teacher.
1.10.4.1 Students’ perception on teachers’ controlling behavior
Students’ perception on their teachers’ controlling behavior defined a
perception from the students that their teachers are more likely to be focused on
control of student disruptions, hostility, and alienation, while their relationship with
students in high-performers classes are likely to center on supporting students to get
higher academic achievements (Oakes, 1985).
25
In this study, Students’ perception on teacher’s controlling behavior refers to
the students’ perception that their teachers are more likely to control the students’
behavior in order to maintain the discipline in the classroom.
1.10.4.2 Students’ perception on teachers’ supporting behavior
Students’ perception on their teachers’ supporting behavior defined a
perception from the students that their teachers are more likely to be focused on
improving the students’ academic achievements (Oakes, 1985).
In this study, Students’ perception on teacher’s controlling behavior refers to
the students’ perception that their teachers are more likely to support the students in
order to improve their academic achievement.
1.11
Conclusion
This chapter had discussed about the background, objectives,
questions,
hypotheses, the importance, scope and limitation, theoretical framework, conceptual
framework of the study and the definitions of variables involved. The next chapter
will discuss about the theories and literacy behind each variables.
26
CHAPTER 2
LITERATURE REVIEW
2.0
Introduction
This chapter is divided into 3 sections. The first section focuses on the
definitions and conceptions of BCAG, self-esteem, self-efficacy, and students’
perception on teacher’s behavior. The second section is focusing on some previous
studies and research done by other researchers related to the objectives of the current
study.
2.1
Definitions and conceptions of the key term
This section would explain more details about the terms used in this thesis in
relation to relevant theories. Terms used in this thesis are BCAG, self-esteem, selfefficacy, student’s perception on teachers’ supporting behavior, and student’s
perception on teachers’ controlling and supporting behavior
2.1.1 BCAG
BCAG was practiced as early as 1920, after Alfred Binet worked out a scale
by which intellectual capacity could be measured (Davies, 1975). It measured inborn
intelligence of a general kind, showing not only what a child had learned, but what
one could learn. Based on Binet’s scale, some schools were able to refine their
27
ability grouping of pupil into the super-normal based on their intelligent quotient
(IQ), and since then grouping system started its evolution.
Ability grouping is often mistakenly thought as the synonym of streaming.
According to Hallam, Ireson, & Davies (2002), streaming is a grouping system that
differentiates students based on their field of studies and what subject would be
emphasized in their studies, while ability grouping is based on the students’ ability
in a stream. For example, a student might be sent to a science stream, and he is
grouped into a lower track classes, which means that he is studying science subjects,
and compared to other students who also studying science, he is considered a lowperformer.
Studies showed that there are many kinds of ability grouping system
practiced in schools (Hallam, Ireson, & Davies, 2002). Table 2.1 provides a list of
the main types of ability grouping used in schools and their brief definitions:
Table 2. 1 Methods of Ability Grouping.
Ability Grouping
Banding
Pupils are placed in classes based on a test of their general
ability. They remain in their streamed class for most subjects
Pupils are placed in two, three, or four bands on the basis of a
test of their general ability. Each band contains a number of
classes and pupils may be regrouped within the band for some
subjects
ability There is no attempt to group together pupils of similar ability or
attainment
Mixed
(heterogeneous
grouping)
Setting (Regrouping)
Within-class grouping
Cross-grade grouping
Pupils are grouped according to their attainment in a particular
subject
Pupils are grouped within the class on the basis of ability or in
mixed-ability groups. They may be regrouped within the class
for different subjects
Pupils in two or more year groups are placed in the same class,
for all or part of the curriculum. They may be regrouped or
taught as a mixed-ability class
28
BCAG, which Hallam et al (2002) termed as banding, is the most rigid form
of ability grouping. Students were put into a classroom based on their overall ability,
and remain there for most subjects. It is based on the assumption that individuals
have a certain level of general intelligence, which might predict their performance
across all subjects and can be measured by objective tests (Ireson & Hallam, 2001).
Their study was also explored why and how schools adopt particular grouping
practices. They discovered a list of reasons to practice the system:
•
•
•
•
•
•
•
•
•
•
•
Raising standards;
Matching work to pupil needs;
The demands of different curriculum subjects;
Making the best use of teacher expertise;
The national literacy and numeracy strategies (in the UK);
Meeting the non-academic needs of pupils;
School and class size;
Resources;
Timetabling;
School ethos; and
Accountability to outside bodies (Ireson & Hallam, 2001)
Overall, Ireson and Hallam summarized that schools main consideration
when taking decisions about grouping practices is related to students’ academic
achievement and the need to match work to students’ need.
BCAG is believed to have positive effects on high-achiever, therefore
teachers and school managements who emphasis on the acquisition of knowledge
are likely to apply such grouping system, as quoted from a study:
Teachers who favor ability grouping tend to be ‘knowledge’
centered, with an emphasis on the acquisition of knowledge and
the attainment of a set of academic standards. They are
particularly interested in and concerned for the bright child,
concentrate on traditional lessons, give more emphasis to literacy
and numeracy, encourage competition and approve of selective
examinations. Teachers who favor non-ability grouping classes
tend to be more child-centered, with a greater concern for the allround
development,
learning
by discovery and
practical
29
experience, and dislike selective examinations. (Hallam, Ireson, &
Davies, 2002)
According to Slavin (2006), the definition of BCAG is the practice of
grouping students in separate classes according to ability level. The ability is often
measured by the academic performance. BCAG is also called tracking in the United
States education system; Slavin stated that the definition of tracks is curriculum
sequences to which students of specified achievement or ability level are assigned.
2.1.2 Self-esteem
Abraham Maslow and Carl Rogers stated that self-esteem is a person’s
overall evaluation of one’s self-worth or self-image (Maslow, 1954; Rogers, 1980).
According to Larsen & Buss (2008), self-esteem measures of many areas are
moderately correlated. A person with high self-esteem in one area also tends to have
high self-esteem in the other areas as well.
Albert Bandura stated that self-esteem is influenced by whether the culture
around an individual values one’s particular characteristics and capabilities
(Bandura, 1997). A study by Stanley Coopersmith (1967) was supporting Bandura’s
statement. Coopersmith looked at the conditions that enhance self-esteem. He
presented self-esteem as the judgment of personal worthiness that is conveyed to
others by each individual in what one says and what one does not say; by what one
does and what one does not. The self-evaluations that the individual makes and
customarily maintains are private and subjective. Nevertheless, Coopersmith argued
that it could be studied in their manifestations as overt behavior. The most general
statement about the antecedents of self-esteem can be given in terms of three
conditions: total or nearly total acceptance of the children by their parents, clearly
defined and enforced limits, and the respect and latitude for individual action that
exist within the defined limits (Coopersmith, 1967).
30
Horrocks and Jackson in 1972 studied developmental aspects of self-esteem,
and found that most children accept what they perceive as the evaluation of the
significant figures around them. An adult usually is more self-accepting and less
self-critical than an adolescent is; unfortunately, an adult is also typically less
idealistic. One’s self-expectations have been tempered by experience and some
perspective of society one has been received during the lifespan. Any individual in
society who faced discrimination or prejudice-based sanction would find difficulties
to exemplify oneself, because the individual’s self-concepts continually find
rejection in society’s non-acceptance of the roles, which one attempted to implement
one’s self-concepts; hence, the self-esteem would be heavily challenged. (Horrocks
& Jackson, 1972)
Individuals not only build up a concept of themselves (true self), which
including the way they are like and how they are seen by others at the present time,
they also construct a concept of what they would prefer to be like, and this concept
is called ideal self (Burns, 1982). In his book, Burns also stated that if one person’s
self-discrepancy (the gap between true and ideal self) is too much, the person would
likely have a low self-esteem.
Hagborg in 1993 postulated some question to measure self-esteem. The
questions were attitude scale (1 for strongly disagree, and 5 for strongly agree, as in
the likert scale) about (1) satisfaction on oneself, (2) the feeling of having a number
of good qualities, (3) an individual wish to respect oneself, (4) a feeling that one is
not good at all, (5) the feeling of individual uselessness at some times, and (6) taking
positive attitude towards oneself.
Self esteem, defined by Christine Von Der Haar (2005) is the value an
individual put on oneself. Anita Woolfolk (2007) defined self-esteem as the value
each individuals place on own characteristics, abilities, and behavior. In all cases,
self-esteem results from an evaluation of oneself (Larsen & Buss, 2008). In a
significant degree, self-esteem would give an impact on the students’ academic
performance.
31
Stanley Coopersmith (1967) looked at the conditions that enhance selfesteem. He presented self-esteem as the judgment of personal worthiness that is
conveyed to others by each individual in what one says and what one does not say;
by what one does and what one does not. The self-evaluations that the individual
makes and customarily maintains are private and subjective. The most general
statement about the antecedents of self-esteem can be given in terms of three
conditions: total or nearly total acceptance of the children by their parents, clearly
defined and enforced limits, and the respect and latitude for individual action that
exist within the defined limits (Coopersmith, 1967).
2.1.3 Self-efficacy
Psychologist Albert Bandura argued that people have intention and
forethought, they are reflective and can anticipate future events, they monitor their
behavior, and evaluate their own progress, and they learn by observing others.
(Bandura, 1997)
Bandura (1989) stated that the concept of self-efficacy refers to the belief
that one can execute a specific course of action to achieve a goal. For example, a
student who believes that he can do most of the exam is said to have a high selfefficacy, while other student who doubts his ability to do the exam is called to have
a low self-efficacy beliefs in this area. As it turns out, high self-efficacy beliefs often
lead to effort and persistence on tasks, and to setting higher goals, compared to
people with low self-efficacy beliefs. He introduced the term Reciprocal
Determinism to describe the way behavior, environment, and person/cognitive
factors interact to create personality. He discussed that being placed in certain
classes under the ability-grouping system is an environmental situation for a student,
which has reciprocal relationship with two other factors. Self-esteem, self-efficacy,
and their perception about teacher’s expectation play the role as personal/cognitive
32
process that takes place in the students head. The next factor was the students’
studying behavior, which will be measured with their academic achievement.
According to Bandura’s social cognitive theory of Reciprocal Determinism, these
three factors are inter-related to one another.
Self-efficacy and performance mutually influence one another. Self-efficacy
leads to better performance; then better performance leads to further increases in
self-efficacy. Self-efficacy can also be influenced by modeling, by seeing others
engage in the performance with positive results. Self-efficacy beliefs will also lead
to greater effort and persistence on relevant tasks, often resulting in better
performance. People with high self-efficacy beliefs approach their goals with the
more positive feelings associated with challenge, rather than the negative feelings
associated with threat (Larsen & Buss, 2008).
Self efficacy defined as beliefs in one’s capabilities to organize and execute
the courses of action required to produce given attainments (Bandura, 1997).
Bandura identified four sources of self-efficacy expectations:
1.
2.
3.
4.
Mastery Experiences. Refer to individual’s direct experiences. This
is the most powerful source of efficacy information. Successes raise
efficacy beliefs, while failures lower efficacy beliefs.
Level of Physiological & Emotional Arousal. Refer to individual’s
eager towards a behavior.
Vicarious Experiences. Refer to other people’s accomplishments;
the more closely the student identifies with the model, the greater the
impact on self-efficacy will be.
Social Persuasion. Refer to specific performance feedback from
society. This source alone could not create enduring increases in selfefficacy, but a persuasive boost in self-efficacy could lead a student
to make an effort, attempt new strategies or try harder to succeed.
Woolfolk (2007) was agree with Bandura’s view that defined self-efficacy
influences motivation through goal setting. If an individual has a high sense of
efficacy in a given area, he or she will set higher goals, be less afraid of failure, and
find new strategies when old ones failed.
Self-efficacy helps people in unsatisfactory situations by encouraging them
to believe that they can succeed (Santrock, 2005). For example, self-efficacy might
33
help heavy smokers or overweight individuals to believe that he has the self-control
to restrict their unhealthy habit. Without a belief that they are able to break their
habit, probably they won’t even try to quit, even though they know that their habits
are likely to cause poor health.
2.1.4 Students’ perception on teacher’s controlling and supportive behavior
It is not the real world that affects an individual; it is the world according to
the perception of an individual that will affect the individual. The statement was
advocated by a group of scholar called Symbolic Interactionists (Myers, 2008). One
of the symbolic Interactionists is C.H Cooley who stated that a symbolic interaction
involved three basic premises. First, humans respond to the environment on the basis
of the meanings that elements of the environment have for them as individuals.
Second, such meanings are a product of social interaction, and third, these societal /
cultural meanings are modified through individual interpretation within the ambit of
this shared interaction. (Cooley, 1912)
It can be demonstrated experimentally that a major perspective how an
individual put a value on oneself is how the individual thinks others think of
him/her. Supporting those symbolic Interactionists, psychologist Robert Burns
(1975) stated that many studies have shown that what an individual thinks about
oneself is similar to what an individual thinks about what others think about oneself.
Another name included as a symbolic interactionist is sociologist G.H Mead.
In agreement with Cooley, in 1934 Mead suggested that the self was a social process
within the individual involving two parties, the ‘I’ and the ‘me’. Mead saw the ‘me’
as a representative of the incorporated ‘others’ within the individual, while ‘I’ was
the impulsive tendency, the unorganized, undisciplined, undifferentiated activity of
the individual. Every behavior commences as an ‘I’, but develops and ends as a
‘me’, as it comes under the influence of societal constraints. In other words, self-
34
concept is not how people see themselves but the way they imagine others see them
(Mead, 1934).
Based on the theories by symbolic interactionists, students would likely to
perceive their teacher’s behavior to develop their perception about the teacher’s
behavior. Their perception will affect their self-esteem and self-efficacy. The
behavior shown by the teacher is also likely to be affected by the BCAG system.
A study by Oakes (1992) discovered that relationships between teachers and
students in low-performers classes are likely to center on control of student
disruptions, hostility, and alienation. In high-performers classes, the relationships
appeared to be supportive for teaching and learning activities. Hence, there might be
a significant difference between the perception of students from high-performance
and low performance classes about their teachers’ behavior.
Thomas L Good (1981) studied the relationship between students’ perception
and teacher’s expectancy. He found that teachers often treat students differently
based on their expectancy towards the students. Good found that towards the slow
(or low performing) students, teachers tend to:
•
•
•
•
•
•
•
•
•
Paying less attention by smiling and making eye contact less often
Calling less frequently to answer classroom questions
Waiting less time to answer questions
Failing to provide clues or ask follow-up questions in problem
situations
Criticizing more frequently for incorrect answers
Praising less often for correct or marginal responses
Giving less feedback and less detailed feedback
Demanding less effort and less work
Interrupting the performance more frequently
As a result, less performing students who receive such treatment become less
willing to take risks in the classroom by volunteering answers or seeking for
teacher’s help (Good, 1981).
35
2.2
Previous related studies
This subsection would discuss about the previous studies and research done
by other researchers about the variables used in this study or that are related to this
research objectives. The discussion would be started by some studies about BCAG,
self-esteem, self-efficacy, and the perception on teachers’ behavior on the students’
self-esteem and self-efficacy.
2.2.1 BCAG and students’ self-esteem
A quantitative study by Barbara M Byrne (1988) on 248 high-school
students from the low-performers classes and 582 students from the high-performers
classes in Ottawa illustrated the possible interplay of compensatory factors in the
formation of self-esteem. Her study was also indicated that students from lowperformers classes place more importance on their non-academic competencies,
rather than on their academic competencies, which they do not value very high. Her
results showed that there is a significant difference in self-esteem between students
from the high and low-performers classes. The finding about the importance the
students from low-performers classes placed showed that there might be some
relationship between their preference and the labeling atmosphere in BCAG.
In 1996, Sally Kemp and David Watkins studied self-esteem and the practice
of BCAG in Hong Kong. The sample consisted of 132 male and 148 female 1st-year
secondary school students from the same kind of area in Hong Kong to control the
socioeconomic factors. Each student was classified into one of five ability bands:
Band 1 includes the top 20% of the Hong Kong student cohort, Band 2 the next
highest 20%, and so forth. A Chinese translation of the Self-Description
Questionnaire 1 (SDQ 1; Marsh, 1989) based on a hierarchical, multifaceted model
of self was used to assess self-esteem in this study. The results indicate that the
effects of ability grouping on self-esteem may not be linear and may show gender
differences when a range of ability groups is considered.
Despite did not argue
36
clearly about how the gender differences might affected the self-esteem of students
in BCAG, Kemp and Watkins indicated some intervening variables that affected the
linearity of their result. In other words, they stated that BCAG has influence on the
students’ self-esteem.
Sociologist Anthony Glendinning from Aberdeen University examined
connections between family life, self-esteem, health, and lifestyles in a sample of
around 1700 young people aged 14-16 years old in 8 rural locations in Northern
Scotland. The educational system in Northern Scotland was applying BCAG, and
due to their academic performance, somehow the grouping was represented the
social economic status (SES) of the students’ family. Students from high SES
background often are put in the high-ability classes while students from low SES
often are put in the low-ability classes. The study concluded that the ability grouping
would affect the students’ self-esteem, and low self-esteem might be related to the
low SES. (Glendinning, 1998). In spite of the fact that Glendinning found that there
is a significant relationship between BCAG and students’ self-esteem, another fact
that the co-incidentally low-performers’ classes contains students’ from the low SES
might indicate another intervening variable between BCAG and the self-esteem,
which might be SES.
A research by Heather Collin (1995), a teacher at one of the school in
Sunderland, studied that a circle session involving high and low ability students in
one circle influences the self-esteem positively. Her subjects were 30 of her own
students. However, she argued
that circling students from the same level
(exclusively high-performers or exclusively low-performers) would make it easier
for the teacher to deliver the lesson. Thus BCAG made it easier for the teacher to
deliver the lesson, compared to having a mixed-grouped classes. Data of this
qualitative study were collected from the comments written by her students (Collin,
1995). From Collin’s findings, it could be concluded that it might take some good
instructional methods to make the mixed-ability grouping works. However,
regardless of the students’ self-esteem, some school management would prefer to
apply BCAG to make it easier for teachers to deliver the lesson.
37
Findings from these various related studies showed that BCAG and selfesteem of the students are related. Differences in self-esteem between high and low
performers were noted, as well as some indirect influence of BCAG towards the
self-esteem. It was also noted that if self-esteem of student is being considered,
mixed-ability grouping might be more appropriate than BCAG.
2.2.2 BCAG and self-efficacy
A qualitative study by Dina Salinitri (2006) from University of Windsor in
Canada found that ability grouping has been one of the most controversial issues in
secondary education. The purpose of her study was to explore the impact of ability
grouping on the self-efficacy at a composite secondary school in Southern Ontario.
Based on the statistical analysis of self-efficacy scales, the self-efficacy levels of
students in grouped classes were found to be lower than those who sit in nongrouped classes when considering attainment of high grades. The students in
grouped classes believed less to their own competence compared to the students in
non-grouped classes. Salinitri found that grouping students based on their abilities
would negatively affect their self-efficacy. Her findings showed that ‘not applying
BCAG’ might be an alternative way to develop positive self-efficacy among the
students.
Luszczynska, Scholz, & Schwarzer conducted a research in 2005 to study the
relationship between general self-efficacy and social cognitive variables (intention,
implementation intentions, outcome expectancies from others, and self-regulation),
behavior-specific self-efficacy, health behavior, well-being, and coping strategies
were examined among 1,935 respondents in three countries. These three countries
are Germany (n = 650), Poland (n = 344), and Korea (n = 941). Perceived selfefficacy was measured by means of the General Self-Efficacy Scale. Participants
were students in various classes and levels. Meta-analysis was used to determine
population effect sizes for four sets of variables. Across countries and samples, there
38
is consistent evidence for associations between perceived self-efficacy and the
variables under study, and one of them is the outcome expectancies from others,
which might stimulated the situation of BCAG.
A dissertation by Irene Kane for University of Pittsburgh in 2007 was using
match-mismatch paradigm to examine 3 psychological constructs that potentially
influence students’ beliefs about participation in school physical-activities. One of
those constructs was self-efficacy. 34 middle school students from both high and
low ability classes (18 high and 16 low) were tested using Physical-Activity
Enjoyment Scale. Self-efficacy of the exercise was greater (p<.001) pre- than post
exercise for both high-ability and low-ability students, hence there is no significant
difference of self-efficacy improvement between the two groups. However, Kane
stated that the students from high-ability classes shown higher average in selfefficacy compared to the students from low-ability classes (Kane, 2007). The
findings showed that there is a significant difference in self-efficacy between the
students from high and low-performers’ classes, and grouping students based on
their ability would not significantly improving their self-efficacy. In other words,
Kane argued that there is no significant influence between the self-efficacy and
BCAG.
One study that supports the implementation of BCAG is the one by Kristy
Beam Hendricks (2009).
The research questions investigated differences in
heterogeneously and homogenously grouped gifted student in the areas of academic
achievement, self-efficacy, and perceptions on teachers’ teaching methods. The
quantitative phase compared scores from the Iowa Test of Basic Skills (ITBS),
Classroom Environment Scale (CES), and Mathematics Self Efficacy Scale (MSES)
between the heterogeneously and homogenously grouped students. The results
showed significantly higher achievement gains and significantly greater self-efficacy
levels among the homogeneously grouped students. ANOVA analysis determined a
significantly positive relationship between the length of time in the homogenously
grouped class and achievement. During the qualitative phase, focus groups were
conducted. The results support ability grouping gifted students in the area of
mathematics by showing a positive impact on their achievement, attitude towards
39
math, and positive perception on their teachers’ behavior (Hendricks, 2009). The
findings showed that BCAG has positive impact towards the students’ self-efficacy.
It could be summarized that some researchers have different opinion about
BCAG in term of its relationship to self-efficacy. Most of them agreed that there
might be some difference in self-efficacy between the students from high and low
performers’ classes, but they did not share the same thouhgts on the importance or
the significance of applying BCAG in the classrooms; some said that BCAG
affected self-efficacy positively, some of them said the vice versa.
2.2.3 BCAG and students’ perception of teacher’s behavior
Previous studies about students’ perception of teachers’ behavior were barely
could be found. These several journal articles cited in this sub-section are those
regarding the teachers’ behaviors in BCAG.
David O. Trouilloud, Phillipe Martinek, and Thomas Guillet conducted a
study in 2002 to explore the relation between teacher behavior and student
achievement in physical education classes, in the light of three complementary
hypotheses. Student achievement may confirm teacher expectations because these
expectations create self-fulfilling prophecies, create perceptual biases, or accurately
predict, without influencing, student achievement. Another purpose was to examine
the mediating role played by students' perceived ability in the teacher expectancy
process. Study data were obtained from 173 students and 7 teachers. Path analysis
revealed that teacher expectations have weak self-fulfilling effects, strongly
predicted student achievement mainly because they are accurately translated into
their behavior in teaching process, and have no biasing effects on teacher judgments.
Results also show evidence concerning the role of partial mediator of perceived
student ability in the confirmation process of teacher expectations (Trouilloud, et.al
2002). From the results, it could be concluded that teachers’ behavior, which were
40
translated from their expectancy, might not have significant influence on the
students. Nevertheless, in this study, teachers’ behavior would not included as a
dependant variable, it is students’ perception on teachers’ behavior that may counted
as one factor that affected students.
In 2002, a dissertation from Hung Siu Tong for The University of Hongkong,
aimed at pointing out the important role of teacher in the ability grouping system.
The study explored the effect of ability grouping from the teachers’ perspective;
how teachers view their class of students made up of homogeneous ability, and
whether or not they have modified their method of instruction. After interviewing 18
teachers in a school that was applying BCAG, Tong concluded that the impact of
ability grouping on student may not be unidirectional; the perspectives of less
experienced teacher gradually take shape and then reflected in their instruction and
attitudes towards their class, then the students will perform as they were expected.
This qualitative study found an existence of the Pygmalion Effect (self-fulfillment
prophecy) in the classrooms under the BCAG. Tong pointed out that practicing
ability grouping should be really well prepared, and should have certain beneficiary
aims. Teachers should be told to modify their instructional method towards different
group of students (Tong, 2002). It was reported by Tong that teachers’ diversity
might play an important role. Teachers’ behavior, which would be interpreted by the
students might be different among teachers, while the method the students use to
interpret might work differently. Thus, students’ perception on their teachers’
behavior might be significant in influencing their non-academic traits, but it was not
based on the BCAG placement, it was more into the teachers’ attributes instead.
A study in 2006 conducted by Hussain Al-Fadhli from Jackson State
University, and Madhu Singh, from Tougaloo College, explored the relationship
between school’s achievement level and teachers’ expectations, locus of control, and
efficacy. 102 teachers from various middle schools were participating in the
qualitative and quantitative study held in Delta, Mississippi. They found that School
environment (climate/culture) explained variation in teachers’ expectancies and
locus of control, a prevailing culture of high expectations in high achieving schools
evidenced by perceptions of teachers and administrators, teachers in high achieving
schools based their expectation on students’ ability, while teachers in low achieving
41
schools based their expectations on personal characteristics (Al-Fadhli & Singh,
2006). Their results stated that differences in expectations between teachers in high
or low performers’ schools might be perceived by the students differently as well.
In conclusion of this subsection, there are some researchers discovered
significant differences in teachers’ expectancy; teachers assigned to the lowperformers’ classes would expect different thing from teachers assigned to the highperformers’ classes. Some findings mentioned that these different teachers’
expectancy was reported to be perceived differently by the students.
2.2.4
Influence of students’ perception of teachers’ behavior on self-esteem
and self-efficacy.
Rohani (1998) studied about the perception of students and teachers on the
characteristics of quality teachers in six secondary schools in the district of Kulim.
The dependent variable, perception of quality teachers, was measured through two
aspects: behavioral and teaching methods. Included in the behavioral aspects was
teachers’ behavior. Parts of her findings showed that, students did not consider
teaching methods as significant elements to determine a quality teacher. They prefer
to consider the teachers’ behavior as more significant factor to see whether a teacher
has an adequate quality or not. According to this situation, it might be concluded
that students took consideration on their teachers’ behavior. Based on the theory of
symbolic interaction (Cooley, 1912), it was not really the teachers’ behavior, it was
the students’ perception on teachers’ behavior towards them that takes into account.
Calabrese, Goodvin, and Niles (2005) conducted a research to identify the
attitudes and traits of teachers with an at-risk student population in a multi-cultural
urban high school, and their effects on the students’ attitude on the teachers. The
research team’s findings indicate that those perceived as effective teachers were
42
culturally responsive, sought small successes, encouraged students, flexible, and
caring. They also formed meaningful relationships with students, had caring
attitudes, and viewed themselves as difference-makers. The research team also
found a number of non-supportive teacher attitudes and traits: blaming, racial
attitudes, frustration leading to inflexibility, co-dependency leading to encouraging
the neediness of students, and lack of respect for the contributions made by the
surrounding community and parents. In this previous study, at-risk students
represented students in the low-performers’ classes, culturally responsive teachers
represented teachers with supportive behavior, and teachers with non-supportive
attitudes represented teachers with controlling behavior. The results showed that
teachers with supportive behavior gains better attitude from the students, and in turn,
students felt they are supported; hence, their levels of self-efficacy might be
elevated.
In 2006, Christine M. Rubie-Davies from University of Auckland aimed to
track the self-perception outcomes of students (N = 256). These students were
divided into two groups, the first group contained those whose teachers had high
class-level expectations, while the other group contained those whose teachers had
low class-level expectations. At the beginning of the year, students completed the
Reading, Mathematics, Physical Abilities, and Peer Relations subscales of the Self
Description Questionnaire-1 (SDQ-1; Marsh, 1990), and asked to complete the same
subscales again at the end of the year. A subscale related to student perception of
how their teacher viewed their abilities was added. At the beginning of the year,
there were no statistically significant differences between the expectation groups in
any of the academic or teacher opinion scales. By the end of the year, statistically
significant differences were found in academic and teacher opinion areas due mainly
to a decline in the self-perceptions of students with low-expectation teachers. The
results found by Rubie-Davies showed that perceptions of the students would
significantly affect their academic performance. During the process, there might be
some slight change in their self-esteem or self-efficacy, which then leads to their
academic performance. If that happened, then it was confirmed that the perception
on teachers’ behavior has significant influence on students’ self-esteem and selfefficacy.
43
2.3
Conclusion
Some studies that have been reviewed in this chapter show that BCAG
applied for the beneficiary of the teachers to deliver instructions to the students.
Furthermore, BCAG might initiate different expectancy of teachers to each group of
students, which will influence their behavior towards the students in the different
classes of BCAG system. These different behaviors perceived differently by the
students. The difference of perception on teachers’ behavior between the students
from high-performer classes and low-performer classes triggers other differences in
term of self-esteem and self-efficacy of the students. Even though numerous
researches into BCAG have been done, a research on BCAG is still relevant in
Malaysia, because the practice of BCAG is common, while the research on this
subject is considered rare, especially regarding the influence of the system towards
the students.
The next chapter will discuss about the methodology used in this study; the
design, the sampling method, instrumentations, and study procedure.
44
CHAPTER 3
METHODOLOGY
3.0
Introduction
Methodology is defined as the activity of choosing, reflecting upon,
evaluating, and justifying the methods being used (Wellington, 2000). Research
methodology refers to process in applying the most effective methods to obtain
valuable data and achieve research aims with minimum cost. Failure in using
effective method to collect data will give inaccurate and bias information, thus
increase data load (Najib, 2003).
This chapter clarifies the overall design of study and methodological
approach used. The chapter is divided into the following sections: design of study,
population and sample, instrumentation, pilot study, data collecting procedures and
data analysis.
45
3.5
Design of Study
This is a descriptive study employing the survey method. The method was
selected because the study is going to see the differences between variables in an
existing phenomenon without manipulating any variable. Survey research employs
group administered pencil and paper questionnaires, face-to-face and telephone
interviews, mailed self-administered questionnaires, and/or some other techniques of
data collection that would produce quantitative. In a survey research, data is usually
collected only once involving the administration of a questionnaire or interview
schedule to a group of respondents who have been randomly selected. An important
point to note in the survey design is that the data is collected after the fact or ex post
facto. The task of the researcher who administrated a survey is to assess which
variables are more highly associated with the dependent variable, and to explain why
these variables are correlated (Leming, 1997).
3.6
Population
The population of the study consists of all form two students from Public
Middle Schools (SMK) in the area (Mukim) of Permas Jaya, district of Pasir
Gudang, Johor, from the highest-performers (first) classes in and the lowestperformers (last) classes in the school. This study does not look at other groups of
students in the population (e.g., students from average classes), because the
objective is to see the difference between the students who put at the extreme ends
of the BCAG System.
There are 22 Public Secondary School (SMK) in the area. Based on the
assumption that average classrooms in a normal SMK serves approximately 35
students, and this study only looks at the students from two classrooms per school
46
(the first and last classes), thus the approximate size of the population is 22 x 2 x 35
= 1540.
3.7
Sample
The sampling method used in this study is purposive sampling. Purposive
samples are sometimes called judgment samples, and are employed by the
researcher in order to approximate the cluster sample using a non-probability
sample. With probability sampling method, the sample taken might not represent the
subject population (e.g., not coming from either highest-performers classes or
lowest-performer classes). In this sampling method the researcher selects a "typical
group" of individuals who might represent the larger population and then collects
data on this group (Leming, 1997). In this study, ‘typical group’ selected is the
group of students from highest-performers class and lowest-performers class in each
school.
There are 22 Public Middle Schools (SMK) in the district, and each school is
assumed to have 1 lowest-performers class and 1 highest-performers class. Each
class approximately hosts 35 students, thus the population of students from highestperformers class and lowest-performers class in the district is approximately 1540.
Sample Size Table of Krejcie and Morgan showed that the sample size should be
302 for a population of 1500, and 310 for a population of 1600. Thus, the sample
size for the population in this study should be approximately between 302 and 310,
but the actual sample used in this study is 302 due to the questionnaire sheets
mismanagement. The sample size of 302 would still be significantly representative
due to its small gap to the approximate true sample size needed.
47
Samples of this study are students in the lowest-performers and highestperformers classes in 4 Public Secondary Schools (SMK) in the district of Pasir
Gudang, sub-district of Permas Jaya. The reason is to get the valid amount of
samples to represent the total population, which are only the students from lowestperformers and highest-performers classes.
3.8
Place and Time of Study
The study took place in 4 Public Middle Schools (SMK) in the district Pasir
Gudang, sub-district Permas Jaya, Johor DT, Malaysia in March 2009
3.5
Instrumentation
Scales were used to collect the data of students’ self-esteem, self-efficacy,
perception on controlling behavior, and perception on supporting behavior. Scales
for self-esteem and self-efficacy are taken from other researchers’ under their
permissions while scales for students’ perception on teachers’ behaviors were
developed by the author.
48
All of the instruments in this study were translated into Bahasa Malaysia by
some teachers of the respective subjects (English and Bahasa Malaysia) using back
translation technique in order to gain more effectiveness towards the students, and to
avoid comprehensive errors.
3.5.1 Self-Esteem
Self-esteem is measured by the Culture-Free Self-Esteem Inventory (Second
Edition) for Children and Adults (CFSEI-2), developed by Battle (1992). Reliability
tests provided by the Manual for the CFSEI include the followings: factor analysis
(alpha) of each sub-scale reported was .71, .67, .66, .76 and 70 respectively. Testretest reliability ranged from .79 to .92. (Azlina, 1995). The scale was used for this
study due to its nature to measure general self-esteem, which is suitable for the
objective of this study to see the general self-esteem.
Reliability of each scale has been tested again in this study. All items came
in a yes / no format and converted into 4 points scale as follows: 1 = Not at all true
2 = hardly true 3 = moderately true 4 = exactly true
3.5.2 Self-Efficacy
Self-efficacy is measured by the General Self-Efficacy Scale. It is a 10-item
psychometric scale that is designed to assess optimistic self-beliefs to cope with a
variety of difficult demands in life. The scale has been originally developed in
German by Matthias Jerusalem and Ralf Schwarzer in 1981 and has been used in
many studies with hundred thousands of participants. In addition, this scale has been
adapted to 30 languages including Indonesian and Chinese (Schwarzer & Jerusalem,
1995).
49
In 2005, the scale was used to study the relationship between general selfefficacy and social cognitive variables to 1,935 respondents in three countries:
Germany (n = 650), Poland (n = 344), and Korea (n = 941). Meta-analysis was used
to determine population effect sizes for four sets of variables. Across countries and
samples, there is consistent evidence for associations between perceived selfefficacy and the variables under study, confirming the validity of the psychometric
scale. In samples from 23 nations, Cronbach’s alphas ranged from .76 to .90, with
the majority in the high .80s (Luszczynska, Scholz, & Schwarzer, 2005).
Nevertheless, reliability of each scale have been tested again in this study.
3.5.3 Perception on teachers’ controlling and supporting behavior
Students’ perception on teacher’s expectation can be measured by looking at
their perception on teacher’s visible behavior. The reason is that the way the
teachers behave in the classroom interaction is often representing their expectations
toward the students (Goodland, 1983).
The instrument to measure students’ perception on teachers’ behavior is
developed by the researcher based on a study by Oakes (1992), which discovered the
difference between teachers’ behavior in the high-performers and low-performers
classes. It’s found that teachers tend to control the behavior of the student when
teaching low-performing students, and tend to support the teaching and learning
activities when teaching high-performing students. These behavioral differences,
according to symbolic interaction theory, were caused by the teachers’ perceptions
towards the students. Prior information about highest/lowest performance perceived
50
by the teachers might influence them to behave differently towards the students
(Cooley, 1912; Mead, 1934; Burns, 1975).
Based on the finding of Goods (1981) and Oakes (1992), teachers’ behavior
towards ability-grouped class can be divided into two characteristics; they are (i)
supportive to teaching and learning activities and (ii) controlling behavior of the
student. These two characteristics are used as constructs to develop the instrument.
3.5.3.1 Perception on teachers’ controlling behavior
Constructs and indices of the teachers’ controlling behavior are arranged
based on the study of Oakes (1992) and Goods (1981). These constructs and indices
would be used to develop the data collection instrument. The table 3.1 depicts the
constructs and indices of each construct.
Table 3.1. Constructs and Indices of Students’ Perception on Teachers’ Controlling
Behavior
Constructs
Controlling
behavior
of the
students
Indices for Positive Perception
No
Praising good behavior more than good marks
1
Paying attention on disciplinary matters more than academic
2
problems
Often asking for total silence, regardless academic topic
3
Concerned more about punishing misbehavior and concerned less
4
about giving special treatment to students who scored low in an
exam.
51
Disliking academically-active student instead of giving
5
opportunities to ask
3.5.3.2 Perception on teachers’ supporting behavior
Constructs and indices of the teachers’ supporting behavior are arranged
based on the study of Oakes (1992) and Goods (1981). These constructs and indices
would be used to develop the data collection instrument. The table 3.2 depicts the
constructs and indices of each construct.
Table 3.2 Constructs and Indices of Students’ Perception on Teachers’ Supporting
Behavior
Constructs Indices for Positive Perception
Supportive
No
Encouraging active academic discussion.
6
Often asking questions about the subject (oral quizzes)
7
Giving more learning advice, instead of disciplinary warning
8
Praising on good performance more than good behavior
9
to
teaching
and
learning
activities
Remembering students by academic performance, instead of 10
disciplinary matters
3.5.3.3 Development of the Instrument
52
In accordance to the literature reviews, teachers with high expectation
towards the students’ academic performance would likely to be supportive to
teaching and learning activities, while teachers with low expectation towards the
students’ academic performance tend to control the behavior of the students.
Students are usually have more than one teacher teaching in their classes,
thus their response to the scale are taken from the common behavior shown by their
teachers, and not taken from the behavior of specific teacher they might like or
dislike.
This scale is meant to measure student’s perception on teachers’ behavior
and the content validity is concerned with whether this scale is measuring the
mentioned variable. The content validity of the instrument was determined by an
expert who has adequate information and knowledge in the domain of classroom
interaction and social psychology from University Technology Malaysia. The expert
reviewed the items for content, clarity, and appropriateness.
The students’ perception on teachers’ controlling behavior and students’
perception on teachers’ controlling behavior questionnaire is a 4 points scale with
20 survey items. Each engagement domain is represented by a subscale. There are
10 items for each construct. The students’ response scale ranges from Strongly
Disagree (=1) to Strongly Agree (=4). Internal consistency of the questionnaire is
assessed using Cronbach’s alpha). A reliability study was conducted using the same
population as for other instruments. Figure 3.1 and Figure 3.2 is showing the
structure of the questionnaire used for the instrument in this study.
Ind no Item
1 2 3 4
Controlling behavior of the students
My teachers are…
1
1 paying attention to students with disciplinary problems
2
2 praising good behavior, regardless how good/bad the
53
3
4
5
1
2
3
4
5
academic performance is
3 often asking for total silence during class
4 frequently punishing misbehavior student
5 likely to prefer silent students than students who actively
asking academic questions
6 frequently giving disciplinary warning
7 more likely to spot disciplinary problems, no matter how
small they are
8 expecting students to be discipline, without focusing on
academic performance
9 more likely to scold students with discipline problem,
instead of students with low academic performance
10 seldom praising students with good exam result
Figure 3.1: Questionnaire of students’ perception on teachers’ controlling
behavior.
Ind no Item
1 2 3 4
Supportive to teaching and learning activities
My teachers are…
6
1 often encouraging active academic discussion among the
students
7
2 often asking questions about the subject to check the
comprehension
8
3 more likely to give learning advice, regardless the
students’ disciplinary problems
9
4 more likely to praise on good performance more than
good behavior
10 5 more likely to remember students by academic
performance, instead of by disciplinary matters
6
6 pay less attention to small disciplinary problem to
students with high academic achievement
7
7 often giving oral quizzes and letting students to answer
questions
8
8 often tell students that he/she has an expectation of good
marks, no matter how bad the behavior was
9
9 more likely to warn low-performers about the bad marks
10 10 pay more attention to students with high academic
achievements, regardless to her/his disciplinary records
54
Figure 3.2: Questionnaire of students’ perception on teachers’ supporting behavior.
3.6
Pilot Study
Pilot study is a preliminary and usually small-scale research study designed
to try out procedures, calibrate measures, and generally serve as dress rehearsal
before a major duty (Colman, 2003). Pilot study is important for the researcher to
know either any section of the questionnaire has any mistake, or not suitable items
that need to be changed before continuing to the real research (Baker, 1994).
In this research, a pilot study was carried out on 10 students from lowestperformers class and 10 students from highest-performers class in their schools. The
Cronbach’s alpha were .602 for the scales of self-esteem, .615 for the scales of selfefficacy, and .663 for the scales of perception on teachers’ behavior.
Before
carrying out pilot studies, the scale of students’ perception on teachers’ behavior
which developed by the researcher was validated by an expert in the field of
educational psychology.
3.7
Study Procedure
55
After the validity and reliability of the instruments is proved and the study
got the approval from EPRD department, the data collection had conducted in the
identified schools.
The data collection was administered by the researcher, starting with
introducing self as an overseas student from UTM, whereas this questionnaire is
merely for the sake of a research in finishing the study to obtain the master degree.
The highlighting of being overseas student is mentioned to help in gaining the
students’ confidence about the confidentiality of the survey result. The subjects then
be asked to response to the questionnaires sincerely, and to choose the most
convenient language version in the questionnaire (English, or Bahasa Malaysia).
Subjects had given 45 minutes time to complete all the questionnaires.
3.8
Data Analysis
Data collected from respondents would be analyzed to fulfill the objectives
and hypotheses of the study. As shown in table 3.3, several statistical methods were
used to analyze the data in this study.
56
Table 3. 3: Research questions and the data analysis methodologies.
Research Questions
Analysis
Is there any difference between self-esteem level of the students
in the high performer and low performer of BCAG?
Is there any difference between self-efficacy level of the students
in the high performer and low performer of BCAG?
Is there any significant difference in perception on teachers’ Mean & independent
controlling behavior between the students in the high and low t-test
performers’ classes of BCAG?
Is there any significant difference in perception on teachers’
supporting behavior between the students in the high and low
performers’ classes of BCAG?
Is there any significant difference between perception on
teachers’ supporting behavior and teachers’ controlling behavior
among students within the high-performers’ classes?
Mean & Paired t-test
Is there any significant difference between perception on
teachers’ supporting behavior and teachers’ supporting behavior
among students within the high-performers’ classes?
Is there any significant influence of students’ perception on
teachers’ controlling and supporting behavior on students’ selfesteem?
Multiple Regression
Is there any significant influence of students’ perception on
teachers’ controlling and supporting behavior on students’ selfefficacy?
3.9
Conclusion
Data collection has been conducted to the appointed schools. The next
chapter will discuss the analysis of the data collected.
57
CHAPTER 4
RESULTS
4.0
Introduction
Data analyses were done according to the research objectives and research
questions of the study. They include:
i.
Reliability of scales
ii.
Analysis of subjects’ demographic factors
iii.
The difference in self-esteem between students from high and low
performers’ classes.
iv.
The difference in self-efficacy between students from high and low
performers’ classes.
v.
The difference in students’ perception on teachers’ controlling
behavior between the students in the high and low performers classes
vi.
The difference in students’ perception on teachers’ supporting
behavior between the students in the high and low performers classes
vii.
The difference between perception on teachers’ supporting behavior
and teachers’ controlling behavior within students from the highperformers classes.
viii.
The difference between perception on teachers’ supporting behavior
and teachers’ controlling behavior within students from the lowperformers classes.
ix.
The influence of students’ perception on teachers’ controlling and
supporting behavior on students’ self-esteem.
x.
The influence of students’ perception on teachers’ controlling and
supporting behavior on students’ self-efficacy.
58
4.1
Reliability of Scales
Instruments used in this study include sets of questionnaire developed to
measure the level of self-esteem, one set of questionnaire to measure the level of
self-efficacy, one set of questionnaire to measure the perception on teachers’
controlling behavior, and one set of questionnaire to measure the perception on
teachers’ supporting behavior. Reliability of the instruments were tested after the
data collection process, table 4.1 shows the reliability of each instrument:
Table 4. 1: Reliability of Scales
Instruments
Cronbach’s Alpha
Self-esteem
.64
Self-efficacy
.60
Perception on teachers’ controlling behavior
.77
Perception on teachers’ supporting
behavior
.76
The alpha values of the instruments used in this research are between .60 and
.077. According to Zaidatun and Mohammad Salleh (2003), the minimum
requirements of an alpha should be .60. Thus, all of the instruments used in this
research are considered reliable to collect the necessary data.
4.2
Demographic Data Analyses
59
Demographic data are one of the descriptive outcomes that are defined as
rates, means, and percentages of single variables. In this study, a demographic data
analysis includes mean or percent of the school and classes of the subjects. Data
were collected from 302 form two students at the sub-district of Permas Jaya, district
of Pasir Gudang, in the state of Johor, Malaysia. The demographic variables of this
study were shown in Table 4.2
Table 4.2 : Demographic Variables of the Study
Demographic Variables
Schools
SMK Permas
Jaya 1
SMK Permas
Jaya 2
SMK Permas
Jaya 3
SMK Sri
Rahmat
Total
Classes
High-Performers
Low-Performers
Total
Count
40
38
78
% within Class
26.8%
24.8%
25.8%
% of Total
13.2%
12.6%
25.8%
Count
35
38
73
% within Class
23.5%
24.8%
24.2%
% of Total
11.6%
12.6%
24.2%
Count
36
40
76
% within Class
24.2%
26.1%
25.2%
% of Total
11.9%
13.2%
25.2%
Count
38
37
75
% within Class
25.5%
24.2%
24.8%
% of Total
12.6%
12.3%
24.8%
Count
149
153
302
% within Class
100.0%
100.0%
100.0%
% of Total
49.3%
50.7%
100.0%
In the study, the number of students from high-performers and lowperformers classes were quite equivalent. There are 149 (49.3%) students from highperformers classes and 153 (50.7%) students from low-performers classes.
60
In the area of Permas Jaya, District of Pasir Gudang, State of Johor,
Malaysia, form 2 students from four schools participated in this study. From SMK
Permas Jaya 1, there were 40 (13.2%) students from high-performers class and 38
(12.6%) students from low-performers class. From SMK Permas Jaya 2, there were
35 (11.6%) students from high-performers class and 38 (12.6%) students from lowperformers class. From SMK Permas Jaya 3, there were 36 (11.9%) students from
high-performers class and 40 (13.2%) students from low-performers class. From
SMK Sri Rahmat, there were 38 (12.6%) students from high-performers class and 37
(12.3%) students from low-performers class.
4.3
The difference of self-esteem between students from high and low
performers’ classes.
Null Hypothesis 1: There is no significant difference of self-esteem level between
the students in the front class and rear class
The mean scores of self-esteem were compared between students from high
and low performers’ classes. As shown in table 4.3, it was found that the self-esteem
level of students from low-performers’ classes was lower (M=25.00) than students
from high-performers’ classes (M=25.97). The difference between two means was
tested using independent sample t-test, because the distribution of the scores is
normal in both groups. As shown in table 4.3, the result found that there is a
significant difference between self-esteem level of students from high and low
performers’ classes with t=1.984, df=299.49 and p value = .048.
Table 4. 3: Difference in Self-Esteem Levels
Level of
Class
N
Mean
sd
High-performers
149
25.97
4.32
Low-performers
153
25.00
4.25
selfesteem
independent sample t-test
t=1.984, df=299.49, p=.048
61
This study found out that there is a significant difference between students
from low-performers’ and high-performers’ classes in term of their self-esteem.
Thus, Null Hypothesis1 is rejected. In other words, it shows that students from lowperformers’ classes did not value themselves as high as students from highperformers’ classes in term of characteristic, ability, and behavior
4.4
The difference of self-efficacy between students from high and low
performers’ classes.
Null Hypothesis 2: There is no significant difference of perceptions on teachers’
behavior between students from high and low performers’ classes
The mean scores of self-efficacy were compared between students from high
and low performers’ classes. The distribution of the scores is normal in both groups;
therefore, the significance of the difference between two means was tested using
independent sample t-test. As shown in table 4.3, it was found that the self-efficacy
level of students from low-performers’ classes was lower (M=33.49) than students
from high-performers’ classes (M=34.44); t value was 2.80, df (299.96), p =.005.
Table 4. 4: Difference in Self-Efficacy Levels
Level of
Class
N
Mean
sd
High-performers
149
34.44
2.90
Low-performers
153
33.49
3.01
selfefficacy
independent sample t-test
t=2.80, df=299.96, p=.005
62
Null Hypothesis2 is rejected because there is a significant difference in selfesteem between students from low and high-performers’ classes. In other words, this
study found out that self-efficacy of students from low-performers’ and highperformers’ classes are significantly different. Students from the low-performers’
classes did not believe their own general competence as students from the highperformers’ classes.
4.5
The difference of perception on teachers’ controlling behavior between
students from high and low performers’ classes.
Null Hypothesis 3: There is no significant difference in perception on teachers’
controlling behavior between students in the high and low performers’ classes
Significance of the difference between two means was tested using
independent sample t-test, because the distribution of the scores is normal in both
groups. And the result found that there is a significant difference between perception
on teachers’ controlling behavior of students from high and low performers’ classes
with t=3.45, df=299.93 and Sig (2-tailed) = .001. As shown in table 4.5, the mean
score for high-performers’ classes’ students are 23.93, which is lower than the mean
score for low-performers’ classes’ which is 25.43.
Table 4.5: Difference in perception on teachers’ controlling behavior
Class
Perception on
High-
teachers’
performers
controlling
Low-
behavior
performers
N
Mean
sd
149
23.93
3.71
independent sample t-test
t=3.45, df=299.93, p= .001
153
25.43
3.87
63
there is a significant difference in perceptions on teachers’ controlling
behavior between students from the high and low-performers’ classes; thereby, Null
Hypothesis 3 is rejected. According to what the students from low-performers’
classes have perceived, teachers were behaving in such ways that are likely to
control the classroom and were more concerned about discipline matters in order to
void children disruption, instead of encouraging or supporting for academic
improvement. Students from high-performers classes scored lower than students
from low-performers’ classes in term of perceiving their teachers’ behavior as
controlling.
4.6
The difference of perception on teachers’ supporting behavior between
students from high and low performers’ classes.
Null Hypothesis 4: There is no significant difference in perception on teachers’
supporting behavior between the students in the high and low performers’ classes
Table 4.6 showed that the mean score for high-performers’ classes’ students
is 25.33, this is significantly lower than the mean score for low-performers’ classes’
which is 24.00. The distribution of the scores is normal in both groups; thereby the
significance of the difference between two means was tested using independent
sample t-test, and it found that there is a significant difference between perception
on teachers’ controlling behavior of students from high and low performers’ classes
with t=3.05, df=298.51 and Sig (2-tailed) = .003.
Table 4. 6: Difference in perception on teachers’ supporting behavior
Class
Perception on
High-
teachers’
performers
N
Mean
sd
independent sample t-test
149
25.33
3.87
t=3.05, df=298.51, p.003
64
supporting
Low-
behavior
performers
153
24.00
3.70
Since there is a significant difference in perception on teachers’ supporting
behavior between the students in the high and low performers’ classes of BCAG,
Null Hypothesis 4 is rejected. Compared to other students from the lowerperformers classes, students from the high-performers classes are more likely to
perceive their teachers behave in such a way to support students to improve the
academic achievements.
4.7
The difference between perception on teachers’ supporting behavior and
teachers’ controlling behavior among students within the highperformers classes.
Null Hypothesis 5: There is no significant difference between perception on
teachers’ supporting behavior and teachers’ controlling behavior within students
from the high-performers classes
This test was comparing the mean scores of perception on teachers’
supporting behavior and teachers’ controlling behavior within the students from the
high-performers’ classes. The mean value for perception on teachers’ controlling
behavior is lower (M=23.93) than perception on teachers’ supporting behavior
(25.33). Paired-sample t-test is used in this test because the distribution of the
scores is normal in both types of perceptions, and the data came from the same
group of sample. The result in table 4.7 shows that there is a significant difference
within the students from high-performers classes in term of their perception on
teachers’ behavior (Sig. (2-tailed) = 0.00), thereby, the null hypothesis 5 is rejected.
65
Table 4. 7: Difference of perception on teachers’ supporting behavior and teachers’
controlling behavior are compared within students from the high-performers classes.
N
Controlling
Behavior
Supporting
Behavior
Mean
Std. Deviation
paired sample ttest
t = -4.56
149
23.93
3.71
df = 148
p = .00
149
25.33
3.87
The difference within the students from the high-performers’ classes’
perception on teachers’ behavior is significance between controlling and supporting
behavior. In other words, students from the high-performers’ classes perceived their
teachers to be more supportive to improve their academic achievements than
controlling their behavior in order to maintain the discipline in the classroom, hence
the Null Hypothesis 5 is rejected.
4.8
The difference between perception on teachers’ supporting behavior and
teachers’ controlling behavior among students within the lowperformers classes.
Null Hypothesis 6: There is no significant difference between perception on
teachers’ supporting behavior and teachers’ controlling behavior within students
from the low-performers classes.
Perception of the students from low-performers classes on their teachers’
behavior was compared between the perception on controlling behavior and
supporting behavior. The mean scores were tested using paired sample t-test,
66
because the compared scores were normally distributed. As seen in table 4.8, the
result shows that the mean value for controlling behavior is higher (M = 25.43) than
the supporting behavior (M = 24.00) there is a significant difference within the
students from low-performers classes in term of their perception on teachers’
behavior (p = 0.00), therefore, null hypothesis 6 is rejected.
Table 4. 8: Difference of perception on teachers’ supporting behavior and teachers’
controlling behavior is compared within students from the low-performers’ classes.
Controlling
Behavior
Supporting
Behavior
N
Mean
Std. Deviation
153
25.43
3.87
paired sample ttest
t = 4.81
df = 152
Sig. (2-tailed) = .00
153
24.00
3.70
The difference within the students from the low-performers’ classes’
perception on teachers’ behavior is significant between controlling and supporting
behavior. In other words, they perceived their teachers are to be more controlling
their behavior in order to maintain the discipline in the classroom than supporting
them to improve their academic achievements. The difference between those two
kinds of perceptions on teachers’ behavior is significant (p= .00), that is the reason
why Null Hypothesis 6 is rejected.
4.9
The influence of students’ perception on teachers’ controlling and
supporting behavior on students’ self-esteem.
Multiple regression analysis with stepwise method is effective to investigate
the effects of more than one predictor variables on an outcome variable; thereby it
67
was used in testing these hypotheses. Compared to correlation, multiple regression
analysis is considered as a more powerful statistical tool for investigating causaleffect phenomenon. Based on the hypotheses, the section would be divided into
three sub-sections; the result of the test when it was applied to the data from (1) lowperformers’ classes, (2) high- performers’ classes, and (3) the pooled data from both
groups of students. Conclusions would be presented afterwards
4.9.1 The influence of students’ perception on teachers’ controlling and
supporting behavior on students’ self-esteem in low-performers’ classes
Null Hypothesis 7a: There is no significant influence of students’ perception on
teachers’ controlling and supporting behavior on students’ from low-performers
classes’ self-esteem.
In this section, the predictor variable is the students’ perception on teachers’
controlling and supporting behavior, while the outcome variable is students’ selfesteem of the students from the low-performers’ classes.
The stepwise method excluded the variables of perception on teachers’
supporting behavior, which showed a negative beta value (-.02) and insignificance
(p =.85). It indicated that level of self-esteem among the students’ from lowperformers’ classes were only significantly influenced by their perception about
teachers’ controlling behavior.
Results showed that the adjusted R2 = .017; df = (1,15); and F = 4.32. It
means that students’ perception on teachers’ controlling behavior can explain 1.7%
of the variance in students’ self-esteem, and the test is significant (p = .04). As
shown in table 4.9, the results also showed that the beta (β) was .17, and the p value
was .04. Although the beta value indicated that the influence was weak (lower than
.20), the p was lower than .05, which means that the influence was significant, hence
the null hypothesis 7a is rejected
68
Table 4.9: Influence of students’ perception on teachers’ controlling behavior on
students’ self-esteem in the low-performers’ classes.
Variable
Students’ perception on
teachers’ controlling
behavior
(in the low-performers’
classes)
Beta Value (β)
p
.17
.04
The way the students from the low-performers’ classes value themselves is
significantly influenced by their perception on teachers’ controlling behavior. In
other words, the more they perceive that their teachers are controlling their
discipline, the higher they value their own characteristics, abilities, and behaviors
would be. On the other hand, their perception that their teachers are supporting an
improvement on academic achievement was not found significant.
4.9.2 The influence of students’ perception on teachers’ controlling and
supporting behavior on students’ self-esteem in high-performers’ classes.
Null Hypothesis 7b: There is no significant influence of students’ perception on
teachers’ controlling and supporting behavior on students’ from high-performers
classes’ self-esteem.
Stepwise multiple regression test was also done to the students from the
high-performers’ classes. Results showed that the adjusted R2 = .38; df = (1,147); F
= 6.78; and p = .010. It was indicated that the test is significant and might explain
3.8% of the data variance. The test excluded the variable of perception on teachers’
controlling behavior due to its insignificance (β = -.09; p = .33). The beta value (β)
69
of the significant variable was .21, which indicated an influence, and the p value was
.01, which showed that the influence is significant, and the null hypothesis 7b is
rejected. The numbers are showed in the table 4.10.
Table 4.10: Influence of students’ perception on teachers’ controlling behavior on
students’ self-esteem in the high-performers’ classes.
Variable
Students’ perception on
teachers’ supporting
behavior
(in the high-performers’
classes)
Beta Value (β)
p
.21
.01
Hypothesis 7b is rejected due to the significant influence of students’
perception on teachers’ supporting behavior on the high-performers’ students’ selfesteem (p = .01). The stepwise method excluded the variable of students’ perception
on teachers’ controlling behavior; it showed that this group of students did not have
their self-esteem significantly influenced by their perception that the teachers
concerned more about disciplinary problems. However, their self-esteem level is
significantly affected by their own perception that their teachers were supporting
their academic achievements.
In other words, students’ from high-performers’ classes would value
themselves higher when they perceived that their teachers were giving them support
to improve their academic achievements.
4.9. 3 The influence of students’ perception on teachers’ controlling and
supporting behavior on students’ self-esteem in both groups.
70
Null Hypothesis 7c: There is no significant influence of students’ perception on
teachers’ controlling and supporting behavior on students’ from both high and lowperformers classes’ self-esteem.
Results showed that the adjusted R2 = .02; df = (1,300); F = 7.93; and the
significance of the test (p) was .005. The predictor variables can explain 2% of the
data variance, and the test is significant. As shown in table 4.11, the results also
showed that the beta (β) was .16, and the p value was .005. The beta value showed
that the influence was quite weak, but due to the p value, it was significant. With the
p value of .005, the null hypothesis is rejected.
Table 4.11: Influence of students’ perception on teachers’ controlling behavior on
students’ self-esteem in both groups (pooled data).
Variable
Students’ perception on
teachers’ supporting
behavior
(Both groups of students)
Beta Value (β)
p
.16
.005
When it was tested to the pooled data including students from the low and
high-performers’ classes, the influence of students’ perception on teachers’ behavior
(both controlling and supporting) on students’ self-esteem was significant, hence,
the null hypothesis 7c is rejected. The predictor variable of students’ perception on
teachers’ controlling behavior was excluded in the stepwise method. Its beta value
was as small as .006, while its significance was .921.
It can be concluded that students would value themselves higher when they
perceived that the teachers supports them to improve academic achievement.
Oppositely, their self-esteem level would not get any higher or lower when they
perceived that their teachers behaved in such way to control their behavior to avoid
disciplinary problems.
71
4.10
The influence of students’ perception on teachers’ controlling and
supporting behavior on students’ self-efficacy.
Multiple regression analysis with stepwise method was used to investigate
the effects of a predictor variable on an outcome variable in this section. The method
was selected due to the number of predictor variables is more than one, and because
regression analysis is more powerful statistical tool for investigating causal-effect
phenomenon.
The section would be divided into three sub-sections, based on the
hypotheses; the result of the test when it was applied to the data from (1) lowperformers’ classes, (2) high- performers’ classes, and (3) the pooled data from both
groups of students. Conclusions would be presented after each analysis.
4.10.1 The influence of students’ perception on teachers’ controlling and
supporting behavior on students’ self-efficacy in low-performers’ classes
Null Hypothesis 8a: There is no significant influence of students’ perception on
teachers’ controlling and supporting behavior on students’ from low-performers
classes’ self- efficacy.
When the data was analyzed by using multiple regression with stepwise
method, both of the predictor variables were excluded, as they were not significant.
Non-significant results were found after the researcher tried to use multiple
regression with enter method. It was found that the adjusted R2 was -.002; df (2,
150); and the p value of the test was .441, which is insignificant. Both predictor
variables have negative beta values (-.10 and -.01) and high p values (.285 and .976).
It means that the results could not explain the influence at all. It can be concluded
that there is no significant influence of students’ perception on teachers’ controlling
72
and supporting behavior on students’ from low-performers classes’ self-efficacy.
Hence, Null Hypothesis 8a is accepted.
It was indicated that the self-efficacy levels of students from the lowperformers’ classes did not influenced by their perception on teachers’ behavior,
neither focusing on disciplinary matters nor supporting the improvement of
academic achievement. Their beliefs on their own competency remained the same
no matter how they perceived their teachers’ behaviors.
4.10.2 The influence of students’ perception on teachers’ controlling and
supporting behavior on students’ self-efficacy in high-performers’
classes
Null Hypothesis 8b: There is no significant influence of students’ perception on
teachers’ controlling and supporting behavior on students’ from high-performers
classes’ self- efficacy.
After being tested with stepwise multiple regression, the results showed that
the adjusted R2 =. 42; df = (1,147); F = 105.79; and the significance of the test (p)
was .000. It indicated the predictor variables can explain 4.2% of the variance in
students’ self-efficacy, and the test is significant. In this case, the excluded variable
was the students’ perception on teachers’ supporting behavior. The beta value was
.11 and the p level was .13, which was insignificant.
As shown in table 4.12, the results also showed that the beta (β) was .65,
which means that each time when the perception level increased by 1 point, the level
of self-efficacy would increase by 0.65. The p value was .000, which is significant.
There is a significant influence of students’ perception on teachers’ controlling and
supporting behavior on students’ from high-performers classes’ self- efficacy
(p=.000), thereby, the hypothesis 8b is rejected.
73
Table 4.12: Influence of students’ perception on teachers’ controlling and
supporting behavior on students’ self-efficacy in the high-performers’ classes.
Variable
Students’ perception on
teachers’ controlling
behavior
(in the high-performers’
classes)
Beta Value (β)
p
.65
.000
It was indicated that the self-efficacy levels of the students from highperformers’ classes are influenced by their perceptions towards the teachers’
behavior. Furthermore, in stepwise method, the insignificance predictor variable was
excluded.
The result showed that self-efficacy levels of the students from the highperformers’ classes were significantly influenced by their perception on teachers’
controlling behavior. In other words, the more they think that their teachers were
controlling their behavior, the more they would believe in their self-competence.
4.10.3 The influence of students’ perception on teachers’ controlling behavior
on students’ self-efficacy in both groups.
Null Hypothesis 8c: There is no significant influence of students’ perception on
teachers’ controlling and supporting behavior on students’ from both high and lowperformers classes’ self- efficacy.
After an analysis to the pooled data, results showed that the adjusted R2 =
.05; df = (2,299); F = 9.10; and the p value of the test was .000. It means that
students’ perception on teachers’ controlling behavior can explain 5% of the
variance in students’ self-efficacy, and the test is significant. As shown in table 4.13,
74
the results also showed that the beta (β) was .21, and the p value was .000.
The influence is considered weak, because the beta (β) value was between .20 and
.40, but it was significant. There is a significant influence of students’ perception on
teachers’ controlling and supporting behavior on students’ from both high and lowperformers classes’ self- efficacy. Null Hypothesis 8c is rejected.
Table 4.13: Influence of students’ perception on teachers’ controlling behavior on
students’ self-efficacy in both groups
Variable
Students’ perception on
teachers’ controlling
behavior
(Both groups of students)
Beta Value (β)
p
.21
.00
The stepwise method excluded one predictor variables, which was the
students’ perception on supporting behavior although it was significant (p = .049).
The other predictor variable (perception on teachers’ controlling behavior) was more
significant (p= .016). It means that the self-efficacy levels of the students from both
classes were significantly influenced by their perceptions on teachers’ behavior,
either controlling or supporting. Nevertheless, even both kinds of perceived
behaviors have significant influences, the perception that the teachers’ behavior was
meant to support their improvement on academic achievement played fewer roles on
increasing their beliefs in their personal competency compared to their perception
that the teachers were likely to concern about disciplinary problems.
It can be concluded that although perceptions from the low-performers’
classes on their teachers’ controlling behavior do not significantly influence their
levels of self-efficacy, there is a significant influence of the perceptions on teachers’
supporting behavior to the levels of self-efficacy of the students in general. Thus, the
null hypothesis 8a is rejected.
75
4.11
Conclusion
Analyses were done to the data, and the results have been presented to
answer all of the research questions asked in this study. The next chapter will
discuss about each of the research findings. The discussion will provide explanations
and comparisons of previous researches in relation to the aspects investigated in this
study, followed by the practical implications and the suggestions for future research.
76
CHAPTER 5
DISCUSSION
5.0
Introduction
This study is looking into the difference in self-efficacy and self-esteem
levels between the students from the high and low-performers’ classes. This current
study is also looking into the difference in perception on teachers controlling and
supporting behavior between those groups of students, as well as the influence of
those perceptions on teachers’ behavior on the students’ self-esteem and selfefficacy.
The study involved 302 form two students from four secondary schools in
the area of Permas Jaya, district of Pasir Gudang, state of Johor, Malaysia. Students
that are involved in this study are those from the highest and the lowest performers’
classes from each school.
In this chapter, discussion is done based on the results presented in Chapter
IV. The discussion provides explanation of the research in terms of these aspects:
i.
The difference in self-esteem between the students from the high and low
performers classes
ii.
The difference of self-efficacy between the students in the high and low
performers classes
77
iii.
The difference of students’ perception on teachers’ controlling behavior
between the students in the high and low performers classes
iv.
The difference of students’ perception on teachers’ supporting behavior
between the students in the high and low performers classes
v.
The difference between perception on teachers’ supporting and
controlling behaviors among students within the high-performers classes.
vi.
The difference between perception on teachers’ supporting and
controlling behaviors among students within the low-performers classes.
vii.
The influence of students’ perception on teachers’ supporting and
controlling behaviors on students’ self-esteem.
viii.
The influence of students’ perception on teachers’ controlling and
supporting behaviors on students’ self-efficacy.
5.1
Summary of the Findings
It was concluded that students from high and low-performers’ classes are
significantly different in term of their self-esteem, self-efficacy, and perception on
teachers’ controlling and supporting behaviors. In term of self-esteem and selfefficacy students from low-performers’ classes scored lower than the students from
high-performers’ classes. In term of perception on teachers’ behavior, students from
low-performers’ classes perceived that their teachers are more concerned about
disciplinary problems, while students from the other classes perceived that their
teachers are more concerned about academic achievements.
However, even though the t-test discovered that most of the influences of
dependent variables are significant, the percentages of the influence were low (R2
ranged from 0 to 4%). This situation showed that other variables might influence the
dependent variables. As stated in chapter one, this study would not cover some
variables such as SES, gender, personality, or other variables that might involved in
78
influencing self-esteem and self-efficacy of the students. The β level of each finding
would be discussed in each respective sub-section.
In each group, students have significantly different perception about their
teachers’ behavior. Students from high-performers’ classes perceived their teachers
to be significantly more supporting than controlling, while students from lowperformers’ classes perceived their teachers to be significantly more controlling than
supporting.
Based on pooled data analysis, self-esteem and self-efficacy levels of the
students from both classes are significantly affected by their perception on teachers’
supporting behavior. When the analyses have done to the groups individually, the
results showed that the self-esteem levels of the students in the high-performers’
classes were significantly influenced by their perception that their teachers behave in
such a way to support their academic achievements. Meanwhile the self-esteem
levels of the students from low-performers’ classes were significantly influenced by
their perception that their teachers behave in such a way to control the students
behavior in order to avoid disciplinary problems.
Students from high-performers’ classes have their self-efficacy significantly
influenced by their perception that their teachers are controlling them to avoid
disciplinary problems. Levels of self-efficacy of the students from low-performers’
classes were not influenced their perception on how their teachers’ behave, neither it
was controlling behavior nor supporting behavior. Pooled data analysis showed that
the self-efficacy levels of both groups were significantly influenced by their
perception on teachers’ controlling behavior.
5.2
The difference of self-esteem between students from high and low
performers’ classes.
79
Null hypothesis 1, there is no significant difference in self-esteem level
between the students from the high and low-performers’ classes in BCAG, was
rejected by this research’ findings. In other words, this study found that there is a
significant difference (t=1.984; p = .048) in self-esteem level between the students
from the high and low-performers’ classes in BCAG, and it was also found that
students from the high-performers’ classes have significantly higher level of selfesteem (M=25.97) than students from the low-performers’ classes (M=25). The
scale’s highest score is 40, thus these two groups of students scored highly
moderate.
Findings of this research support the statement of Slavin (1987). In his metaanalysis study, he found that some research findings have noted that BCAG affects
students’ socio emotional domains, students feel stigmatized being assigned to lowachievers classes, where they will value themselves lower, and such feelings will
eventually affect their academic achievement. Results of this study captured this
situation by finding that students in the low-performers’ classes scored significantly
low in self-esteem compared to students in the high-performers’ classes.
Results of this study are also supporting the quantitative study by Barbara M
Byrne (1988) which found out that students from low-performers’ classes do not
value themselves very high. This research found that students from low-performers’
classes scores significantly lower than the students from high-performers’ classes.
As defined by Burns (1982), self-esteem is also determined by how others
see an individual. Even though the names of the classes are camouflaged into neutral
terms, school society can still be aware of which classes belong to the lowperformers or high-performers, and put different value to the students from those
respective classes. This kind of situation would influence how students look at
themselves. Results of this study confirmed that the self-esteem levels are
significantly different between students from the high and low-performers’ classes.
This study also found something in line with the finding of Good (1981),
who studied that teachers assigned in lower-achievers classes seem to have lower
expectations of the students compared to the teachers in high-performers’ classes.
80
Low expectations from the teachers, would add some factors to lower the selfesteem level of the students in low-performers’ classes, while high expectations
from the teachers, would add some factors to raise the self-esteem level of the
students in high-performers’ classes.
Another statement that is in line with this study is a study by Horrocks and
Jackson (1972) about developmental aspects of self-esteem; most children accept
what they perceive as the evaluation of the significant figures around them. In a
school where BCAG system is applied, school officers attributed certain students as
‘high-performers’ and ‘low-performers’. These attributions were perceived as
evaluations by the students, which eventually accepted. The self-esteem levels of the
students who were attributed as low-performers were significantly lower than those
who attributed as high-performers. This evaluation of significant figures at school,
including teachers, principal, and any other officers have contributed to the
development of the students’ self-esteem.
In BCAG, being put in the highest or lowest classes might influence a
students’ academic self-esteem. According to Larsen & Buss (2008), an individual
with high self-esteem in one area also tends to have high self-esteem in the other
areas as well. Findings of this study confirmed the previous statement. It was found
that the students from high-performers’ classes scored high in general self-esteem
and students from low-performers’ classes scored low in general self-esteem.
General self-esteem is an overall self-esteem, which likely to be affected by
academic self-esteem developed by the BCAG. Thus, this research has proved that
Larsen & Buss (2008) were right by finding that students with high levels of selfesteem in one area (e.g., academic) tend to score significantly higher in general selfesteem.
There are also studies that have found different results, Kemp and Watkins
(1996) study about self-esteem and the practice of BCAG in Hong Kong. Their
results indicate that the effects of ability grouping on self-esteem may not be. Their
statement was denied by this study, by finding that the difference in self-esteem
levels between students from different extreme ends of BCAG system is
81
significantly linear; low-performers scored low, high-performers scored high, and
the difference between them is significant.
Research findings of Glendinning (1998), that studied the implementation of
BCAG in educational system in Northern Scotland, is both supported and not
supported by this research’ findings. He concluded that the BCAG would affect the
students’ self-esteem, and the difference in self-esteem between high-and low
performers’ classes’ students was significant. However, because somehow the
students from high social economic status (SES) background often are put in the
high-ability classes while students from low SES often are put in the low-ability
classes, Glendinning argued that the low self-esteem might be related to the low
SES. His conclusion of the difference in self-esteem between high-and low
performers’ classes’ students is supported by the results of this study, but his
argument that the low self-esteem might be related to the low SES could not be
supported by this research’ findings because this research did not involve SES as
one of the variables.
This respective finding of this research will be discussed further in the
discussion about the influence of students’ perception on teachers’ behaviors on
students’ self-esteem and self-efficacy levels.
5.3
The difference of self-efficacy between students from high and low
performers’ classes.
This research’ findings has rejected Null hypothesis 2, which stated that
there is no significant difference in self-efficacy level between the students from the
high and low-performers’ classes in BCAG. This study found that there is a
significant difference (t=2.80; p = .005) in self-efficacy level between the students
from the high and low-performers’ classes in BCAG, and it was also found that
students from the high-performers’ classes have significantly higher level of self-
82
efficacy (M=34.44) than students from the low-performers’ classes (M=33.49).
Despite being significantly different, both groups scored moderately high (the
scale’s higher score is 40).
Just like the result on self-esteem, what this research found about selfefficacy also supports the statement of Slavin’s meta-analysis study in 1987, where
he found that some research findings have noted that BCAG affects students’ socio
emotional domains, students feel stigmatized being assigned to low-achievers
classes, where they will not believe in their own competence. This phenomenon
reflected in the results of this study, which found that the self-efficacy levels of
students from the low-performers’ classes are significantly lower than the students
from the high-performers’ classes.
Results of this study are supporting Larsen and Buss (2008) statement that
self-efficacy and performance mutually influence one another. Self-efficacy leads to
better performance; then better performance leads to further increases in selfefficacy. Self-efficacy can also be influenced by modeling, and by seeing others. In
BCAG system, students are grouped based-on their previous academic performance.
Their statement explained the result of this study that students from low-performers’
classes scores significantly lower than the students from high-performers’ classes in
term of self-efficacy. Self-efficacy levels of a group is also maintained the same (to
be low or to be high), because those students would be sitting at the same classes
with other students who has the same self-efficacy levels, they tend to model their
peers and ended up by having the same self-efficacy levels. For instance, a student
who sits in a high-performers class would likely to model and to be influenced by
his peers in the class, which scores high in term of self-efficacy. The same case will
happen to a student who sits in a low-performers class, he might model the
delinquency, truancy, and other things that keep the level of his self-efficacy
significantly lower than a high-performer student.
A dissertation by Kane in 2007 is both supported and denied by the results of
this study. She stated that the self-efficacy of an exercise was greater (p<.001) prethan post exercise for both high-ability and low-ability students, hence there is no
significant difference of self-efficacy improvement between the two groups.
83
However, Kane also stated that the students from high-ability classes shown higher
average in self-efficacy compared to the students from low-ability classes. The first
part of Kane’s statement is not supported by this study. It might be because Kane
used a method that compared between pre-treatment and post-treatment to both
groups of students, where she found that the self-efficacy of both groups had the
same gap between pre and post treatment. Results of this study support the second
part of Kane’s statement. She stated that the students from high-ability classes
shown higher average in self-efficacy compared to the students from low-ability
classes.
Bandura’s statement in 1989 is supported by the results of this study. He
mentioned that a student who believes that he can do most of the exam is said to
have a high self-efficacy, while other student who doubts his ability to do the exam
is called to have a low self-efficacy beliefs in this area. Before being put into a highperformers’ class, a student might have nicely done series of exams. Her past
success would increase her present belief in her competency. Meanwhile, a student
who did not perform well in the same exams might have to be put in a lowperformers’ class. His past failure would decrease his present belief in his
competency. This different was detected in this research, which found that the selfefficacy levels are significantly different between students in high and lowperformers’ classes.
Self-efficacy and performance mutually influence one another (Larsen &
Buss, 2008). Self-efficacy leads to better performance; then better performance leads
to further increases in self-efficacy. Self-efficacy can also be influenced by
modeling, by seeing others engage in the performance with positive results. Selfefficacy beliefs will also lead to greater effort and persistence on relevant tasks,
often resulting in better performance. This statement explained this particular
finding of the research. Students from the high-performers’ classes were often get
their self-efficacy influenced by their own better performance and their peers
influences, and that is the explanation on how this study found that students from
high-performers’ classes scored higher in self-efficacy compared to the students
from the low-performers’ classes.
84
Due to its relationships to other findings of this research, this respective
finding will be discussed deeper in the discussion around the influence of students’
perception on teachers’ behaviors on students’ self-esteem and self-efficacy levels.
5.4
The difference of students’ perception on teachers’ controlling behavior
between the students in the high and low performers’ classes
Null Hypothesis 3 is rejected, because this study found that students from
different groups perceive their teachers’ controlling behavior in significantly
different ways. Results of this study showed that students from low-performers’
classes were more likely to perceived their teachers’ behavior as controlling
(M=25.43) compared to the students from high-performers’ classes (M=23.93). The
difference between those two kinds of perceptions is significant (t=-3.45; p = .001).
This study was not looking at the teachers’ behavior or its effects to the students;
instead, it was looking at how students perceived the behavior of their teachers. This
section will discuss about the finding of this study around the students’ perception
that their teachers’ behavior is more likely to center on control of students’ behavior
in order to avoid disciplinary matters.
Results of the study are clearly supporting a statement by Oakes (1992),
which discovered that relationships between teachers and students in low-performers
classes are likely to center on control of student disruptions, hostility, and alienation.
It was found in this study that students in the low-performers’ classes perceived their
teachers’ as praising good behavior more than good marks, paying attention on
disciplinary matters more than academic problems, often asking for total silence,
regardless academic topic, punishing misbehavior more than giving special
treatment to low markers, and disliking academically-active student instead of
giving opportunities to ask. In term of perceiving their teachers were concerning
85
more on disciplinary problems, students’ from high-performers’ classes scored
significantly lower than the students from low-performers’ classes.
The phenomenon found in this study supports the findings of Goods (1981),
who argued that teachers often treat students differently based on their expectancy
towards the students. He found that as a result, students from low-performers’
classes became less willing to take risks in the classroom by volunteering answers or
seeking for teacher’s academic help. This situation, which is the opposite of what
happened to the students from high-performers’ classes, would drive students to
perceive that their teachers are more concerning into disciplinary problems
compared to improving academic achievements.
Because of its relationship with other findings, this respective finding of this
research will be discussed further in the following discussions about the influence of
students’ perception on teachers’ behaviors on students’ self-esteem and selfefficacy levels.
5.5
The difference of students’ perception on teachers’ supporting behavior
between the students in the high and low performers classes
Result of this study found that students from low-performers’ classes were
not likely to perceive that their teachers would support them in term of academic
achievements, vice versa, the students from high-performers’ classes perceived that
their teachers would. This difference was statistically shown by this study, students
from the low-performers’ classes scores lower (M=24.00) than the high-performers’
classes’ students (M=25.44), and the difference was significant (t=3.05; p=.003).
A study by Oakes (1992) discovered that relationships between teachers and
students in high-performers’- classes, the relationships appeared to be supportive for
teaching and learning activities. Result of this study supports the Oakes’ finding, by
86
confirming that the students from high-performers’ classes scored averagely high on
perceiving their teachers to behave in such a way to improve their academic
achievements. In the same study in 1992, Oakes also stated that teachers-students
relationships are different between high and low-performers’ classes’ students, this
respective finding also supported by the findings of this study.
A study by Ireson & Hallam (2001) was not supported by the finding of this
research. They discovered that one of the reasons for a school to practice BCAG is
to meet the students’ non-academic needs. One of the findings in this research was
stated that students from high-performers’ classes perceived that their teachers are
more likely to focus on supporting their study to achieve higher in term of academic
and less concern on their non-academic needs, in this case, it is self-esteem and selfefficacy.
This respective finding will be discussed further in the discussion about the
influence of students’ perception on teachers’ behaviors on students’ self-esteem and
self-efficacy levels.
5.6
The difference between perception on teachers’ supporting and
controlling behaviors among students within the high-performers
classes.
This section will discuss about the research question that look further into
students’ perception on teachers’ behavior. It has been discussed in previous
sections about the difference of perceptions between the high and low-performer
classers, and it was found that students from high and low performers’ classes are
significantly different in perceiving their teachers behavior. The following
discussion is based on the finding that there is a significant difference in perception
on teachers’ supporting and controlling behaviors among students within the highperformers classes, hence the rejection of the null hypothesis 5.
87
Students in the high-performers’ classes were asked to fill the questionnaires
about their perceptions on the teachers. It was found that they perceived that their
teachers are more likely to be supportive than controlling. According to their
perceptions, their teachers are more concerned about their academic achievements
(M = 25.33) than disciplinary problems (M = 23.93), and the difference was highly
significant (t = -4.56; p = .00).
This finding supports an argument by Oakes (1992) who discovered that
relationships between teachers and students in high-performers’ classes as
supportive for teaching and learning activities. It was found in this study that
students in the high-performers’ classes perceived their teachers’ as encouraging
active academic discussion, often asking questions about the subject (oral quizzes),
giving more learning advice, instead of disciplinary warning, praising on good
performance more than good behavior, remembering students by academic
performance instead of disciplinary matters. These behaviors would be perceived by
the students as supportive to academic improvements instead of controlling students’
behavior to avoid disciplinary problems.
This study supports the findings of Goods (1981), who argued that teachers
often treat students differently based on their expectancy towards the students.
Perception of the high-performers’ classers represented how their teachers behaved
in the classroom, and the behavior represented expectations. From a finding of this
study about students’ perception within the high-performers’ classes, it can be
concluded that the teachers were expecting the students to achieve higher in term of
academic achievement, and in turn, it drove the students to get such a perception
about their teachers were supportive instead of controlling.
As stated by Myers (2008), students’ disposition of being placed in highperformers or low-performers class might serve as prior information for teachers.
The prior information will determine the level of teacher’s expectation; they would
likely to put higher expectation towards students from the high-performers class and
lower expectation towards students from the low-performers class. These
expectation levels will influence the teacher’s behavior towards the students. For
88
example, the teacher would likely to give the high-performers more chance to reply
or to query compared to the low-performers. Through the teacher’s behavior,
students might be able to perceive about the teachers’ expectation towards them.
This study supports Myers’ previous statement by finding that students from highperformers’ classes are more likely to perceive their teachers’ behavior as supportive
towards academic achievements.
Further discussion about this respective finding will take place as this chapter
discuss around the influence of perceptions on teachers’ behavior on students’ selfesteem and self-efficacy.
5.7
The difference between perception on teachers’ supporting and
controlling behaviors among students within the low-performers classes.
It was asked in the research question 6, whether there is any significant
difference between perception on teachers’ supporting behavior and teachers’
controlling behavior among students within the low-performers’ classes or not. The
result of this study rejected the null hypothesis 6 by finding that the difference
within the students from the low-performers’ classes’ perception on teachers’
behavior is significance between controlling and supporting behavior.
From the way the students fill up the questionnaire about the perception on
teachers’ behavior, it seems like the students from the low-performers’ perceived
that their teachers to be more controlling than supporting. According to their
perceptions, their teachers were more likely concerning about disciplinary matters
(M = 25.43) than being supportive to improve their academic achievements (M =
24.00), and the difference was highly significant (t = 4.81; p =.00).
According to Goods (1981), teachers often treat students differently based on
their expectancy towards the students. Students from the low-performers’ classes’
89
perceptions represented how their teachers behaved in the classroom, and the
behavior represented expectations. It can be concluded that the teachers were not
expecting the students to achieve higher in term of academic achievement, and in
turn, it drove the students to get such a perception about their teachers were more
into controlling students behavior to avoid disciplinary problems.
Results of the study supports a statement by Oakes (1992): relationships
between teachers and students in low-performers classes are likely to center on
control of student disruptions, hostility, and alienation. Students in this group
perceived their teachers’ as praising good behavior more than good marks, paying
attention on disciplinary matters more than academic problems, often asking for
total silence, regardless academic topic, punishing misbehavior more than giving
special treatment to low markers, and disliking academically-active student instead
of giving opportunities to ask. These kinds of behavior can be assumed as the reason
why students perceived that their teachers are less likely to concern about academic
achievement.
Complimenting the previously discussed theories, Goodland (1983) and
Oakes (1985), argued that students from low-performers’ classes are far more likely
to have disciplinary problems compared to the other students. This study supports
their argument. Results of this study found that students in the low-performers
classes perceived their teachers behaved in such ways that shows more focus on
disciplinary problems instead of academic achievements.
This respective findings about the difference of perceptions within students
from low-performers’ classes will be discussed further in the section where this
paper discusses about the influence of students’ perceptions on teachers’ behavior on
their levels of self-esteem and self-efficacy.
90
5.8
The influence of students’ perception on teachers’ controlling and
supporting behavior on students’ self-esteem
The variable of students’ perception on teachers’ behavior were assumed to
have different effect to the different group of students and to a pooled group, where
two groups of students were combined. Therefore, the data analyses were done
separately. It was found that there is a significant influence of students’ perception
on teachers’ behavior on the self-esteem levels of the students from the highperformers’ classes (β = .21; p = .10, by teachers’ supporting behavior), lowperformers’ classes (β = .17; p = .039, by teachers’ controlling behavior), and a
pooled group of both types of students (β = .16; p = .005, by teachers’ supporting
behavior). It can be concluded into a single statement, that in general, students’
perceptions on teachers’ behaviors have significant influence on their self-esteem
levels.
Previous findings of this study stated that students from the low-performers
classes perceived that their teachers behave in such ways to control their behavior, in
order to minimize disciplinary problems. The way they value themselves (selfesteem) was significantly influenced by their perception on teachers’ controlling
behavior, the more they perceive that their teachers try to control their behavior, the
more they higher they will value themselves. However, the influence was considered
as weak; the b value was .21, which means that it takes 1 point of perception to
increase their self-esteem for 0.21 points. This might be due to a situation where
they did not perceive their teachers as significantly supportive; hence, the only
attention they can get from their teachers was in the form of controlling behavior.
The more the teachers show controls, the more the students of this kind will value
themselves. On the other hand, no matter how their teachers tried to support them
academically, their value upon themselves would not be affected.
Students from high-performers classes will value themselves higher when
they perceive their teachers support them in terms of academic achievements. This
might prove that being placed in the high-performers’ classes are not the only factor
to gain higher levels of self-esteem; it takes supportive teachers to develop their self-
91
esteem to a higher level. Compared to the students’ from low-performers classes,
their perception on teachers’ controlling behavior did not have any significant
influence towards their self-esteem levels. It indicated that when they perceived that
their teachers tried to control their behavior, they would not change their way to
value themselves.
As stated by Cooley (1912), Coopersmith (1967), Horrocks & Jackson
(1972), Burns (1975, 1982), and Bandura (1997), and Myers (2008), it is not the
evaluation of oneself alone that generated self-esteem; society, cultural values,
upbringing, and many other external factors that can be abbreviated as ‘others’ are
playing important roles in building one’s self-esteem. Thus, how a person thinks
others would value him/her would also generate his/her self-esteem. These theories
are all supported by this study by finding that perception on teachers’ behavior has a
significant influence on the self-esteem levels of the students.
Slavin (1987) and Good (1981) were noted that teachers assigned in lowerachievers classes seem to have lower expectations of the students than teachers
assigned in higher-achievers classes. These mentioned studies were supported by the
findings of this study. Teachers’ expectations would affect their behaviors towards
the students, who might perceived the teachers’ behaviors as supporting or
controlling, and in turns, the influence will get into their way of giving value to
themselves, which is called, self-esteem.
These particular findings will be discussed again thoroughly in the subsection when the paper discuss about the relationship between phenomenon in
BCAG and the theoretical framework.
5.9
The influence of students’ perception on teachers’ controlling and
supporting behavior on students’ self-efficacy
92
Data analyses were done separately towards the students from the highperformers’ classes, low-performers’ classes, and a pooled group of both types of
students in term of this variable. The separation has done based on an assumption
that this variable might have different effect to the different group of students and to
a pooled group, where two groups of students were combined. It was found that
perception on teachers’ controlling behaviors have significant influence on the selfefficacy levels in high-performers’ classes (β = .65; p = .000) and the pooled group
(β = .21; p = .000). Nevertheless, these perceptions on teachers’ behavior did not
have any significant influence on the self-efficacy of students from low-performers’
classes, the stepwise method excluded both predictor variables due to insignificance.
This study found that there is no significant influence of students’ perception
on teachers’ controlling and supporting behavior on self-efficacy levels of the
students from low-performers classes. Despite scored significantly lower than the
students from high-performers’ classes, these students scored moderately high
(M=33.49, where the maximum score is 40). It indicates that no matter how their
teachers behave, their level of self-efficacy would not significantly be influenced. It
seems like they do not care about what kind of behavior their teachers would
behave. It is true that they perceived that their teachers tried to control their
behavior, but it did not matter much to them in term of evaluating themselves. In
other words, their evaluations on themselves were not significantly affected by what
they perceived their teachers are like
According to Larsen and Buss (2008), self-efficacy and performance
mutually influence one another. Self-efficacy leads to better performance; then
better performance leads to further increases in self-efficacy. In BCAG, self-efficacy
levels of a group is also maintained the same (to be low or to be high), because those
students would be sitting at the same classes with other students who has the same
self-efficacy levels, they tend to model their peers and ended up by having the same
self-efficacy levels. Larsen and Buss (2008) added that a student who sits in a lowperformers class might model the delinquency, truancy, and other things that keep
the level of his/her self-efficacy significantly lower than the other group of students.
Their theory concluded that self-efficacy might be influenced by factors such as
93
performance and peers. Self-efficacy levels of the subjects of this study might not be
significantly influenced by their perception teachers’ behavior because other factors
have influenced it more significantly.
A study by Levy (1996) might explain the reason why students’ perception
on teachers’ behaviors did not significantly influence their self-efficacy. Levy
discovered that self-efficacy could be manipulated. Manipulating self-efficacy to
students could be done by giving them information about themselves. An obvious
placement in a grouped class might manipulate the students’ self-efficacy. When
they were often exposed to a fact that they are part of a low-performers group, their
self-efficacy in academic achievement might be degraded and reach a point where
they do not believe that they can control their outcome by putting more effort;
hence, they will not even try to try harder. The situation might reached to some point
where any behavior of the teacher would not significantly influence the students’
self-efficacy, because their self-efficacy was formed and maintained at such a level
by the daily manipulation they received by being put in the low-performers’ classes.
Another explanation about this particular result is that according to the
statistical findings, students from the low performers’ classes were perceiving that
their teachers’ behaved in such ways to focus more into disciplinary problems
instead of academic achievements. Meanwhile, as defined earlier, self-efficacy is
related to performance (Bandura, 1989, 1997; Woolfolk, 2007; Larsen and Buss,
2008). Thereby, it could be concluded that their perception on teachers’ behavior
did not influence their beliefs of their own competence, because they perceived that
their teachers only concern about their disciplinary problems instead of their
competence.
This study found that perception on teachers’ controlling behavior, have
significant influence (β = .65; p = .00) on the self-efficacy levels in high-performers’
classes and their perception that the teachers are supportive towards their academic
achievement did not have any significant effect (β = .11; p = .13) to their selfefficacy. Despite students from this group perceived that their teachers are
supportive towards their academic achievements, their beliefs in their competence
94
relied on how their teachers’ tried to discipline them. Based on the beta value, when
they increase such perception by one point, their level of belief in their selfcompetence would increase about 65%. Another study might be needed to look into
this fact deeper.
Levy
(1996)
discovered
that
self-efficacy
could
be
manipulated.
Manipulating self-efficacy to students could be done by giving them information
about themselves. Teachers assigned to high-performers’ classes would expect the
students to have good performances, these expectancies then showed by the
teachers’ behavior, which the students perceived as supportive. This supportive
behavior then influence the students’ beliefs in their own competencies; hence the
findings of this study supports the discovery of Levy (1996)
Bandura (1997) identified four sources of self-efficacy expectations, and one
of them is social persuasion, which refers to specific performance feedback from
society. To students from high-performers’ classes, teachers’ behavior would be
perceived as performance feedback that will influence their belief on what they can
achieve. Findings of this study supported Bandura’s statement in this term.
A study by Rosenthal & Jacobson (1968) was reported that there is a
significant relationship between teachers’ expectation and students’ academic
achievements. The lower the expectation, the lower the academic achievement
would be. This process occurred indirectly through the conversion of teachers’
expectation to behavior, which will be perceived by the students whose self-efficacy
levels are influenced. This study supports Rosenthal and Jacobson’s statement.
5.10
Practical Implications
This section would discuss the practical implication after the findings
discovered. This discussion would neither going to create a new educational system
95
nor trying to correct the existing one, instead this discussion would discuss about
what teachers and educators might able to do to gain the best results out of the
system and to reduce any negative influence that may occur.
Findings of this study showed that the influence of the students’ perception
on teachers’ behavior could only explain a small percentage of their self-esteem and
self-efficacy; however, the perceptions do have some contributions. It was shown by
the significance of the influences.
In a system where BCAG is kept being applied, it is good to have different
instructional strategies regarding the difference between high and low-performers’
classes. Nevertheless, teachers should at least try not to let themselves biased by the
label given to the students. With knowledge that students’ self-esteem and selfefficacy levels are significantly influenced by their perceptions on teachers’
behaviors, then it might be better to keep the balance between controlling students’
behavior and supporting academic achievement in both classes.
Teachers should be aware that the label given by BCAG system to the
students might represent their previous academic ability, but it does not represent
how good or how bad they would be in the future. Still, all of the students deserve
good expectations from the teachers. Teachers’ expectations might influence their
behavior, which then influenced the students’ self-esteem and self-efficacy.
In order to gain the positive results of academic achievement while not
sacrificing non-academic elements such as self-esteem and self-efficacy, a teacher
should show some balance between being supportive and discipline. In other words,
students from the high and low-performers’ classes might be different, but they
would have to perceive that their teachers are supportive, yet discipline towards their
behavior, students from high-performers’ classes should see their teachers in the
same way as students from low-performers’ classes. The instructional method might
be different between high and low-achievers’ students, but the perceptions students
have from their teachers should not be either extremely discipline or extremely
controlling.
96
5.11
Suggestion for Future Researches
This study has several limitations in terms of size of population, some
excluded variables, reliability of scales, and scope of the study. Further studies
would also be needed by some findings of this current study. This section will
shortly discuss about those limitations and suggest future researches to conduct.
Two findings of this study need to be studied further. One of them is the fact
that students’ self-efficacy levels in the low-performers’ classes were not
significantly influence by students’ perception on teachers behavior. The second one
is the fact that the students from high-performers’ classes have their self-efficacy
levels significantly influenced by their perception on teachers’ controlling behavior,
while they perceived their teachers to be more supporting than controlling. It might
take more samples or more variables included in a research to be able to discuss
about these particular findings better.
Result of this study might not represent the whole country of Malaysia;
hence, the practical implications suggested might not be appropriate before a
thorough research through the whole country or at least Peninsular Malaysia. Due to
the similar grouping system, same research might be conducted in other countries
like Singapore or Brunei. In this study, findings about the self-efficacy and selfesteem levels were significantly different, but both groups of students scored highly
moderate or moderate. With larger numbers of sample, the gap between scores
might be more obvious.
Some extraneous variables that could have influenced the results might have
been excluded in this study. They might be demographic factors such as gender,
socio economic status, place (urban or rural area), and school status (government or
private) might have possibly influenced the variables in this study. There might be
difference in self-efficacy between urban and rural students, or female and male
97
students, and soon. It is suggested to add more demographic factors to the future
research in order to get more accurate and representative results.
Scales used in this study, especially the one developed by the author, might
need to be tested somewhere else with more numbers of samples to get the more
accurate information on its reliability. Translations to more language should be
added to the scale to make it more effective during data collection.
BCAG, self-esteem, self-efficacy, and students’ perception on teachers’
behavior might not the only connected variables. There are other variables need to
be researched such as academic achievements, academic improvements, or
disciplinary problem decreasing, and so on. It might be important nowadays, to
study about the influence of students’ perception on teachers’ behavior towards
academic achievements or the influence of BCAG to general academic
improvements.
5.12
Conclusion
Findings have been discussed in this chapter. All of the findings supported
most of the main theories stated in the early chapters, and the most important thing
is that from this study, it confirmed the fact that there is a significant influence of
students’ perception on teachers’ behavior on the students’ self-efficacy and selfesteem.
98
REFERENCES
Bandura, A. (1989). Human agency in social cognitive theory. American Psychologist , 44,
1175-1184.
Bandura, A. (1997). Self-efficacy: The excercise of control. New York, NY: Freeman.
Burns, R. (1975). Attitudes to self and to three categories of others in a student group.
Educational studies (1), 181-189.
Burns, R. B. (1982). Self-Concept Development and Education. Dorchester: Holt, Rinehart,
and Winston.
Calabrase, R., Goodvin, S., & Niles, R. (2005). Identifying the attitudes and traits of
teachers with an at-risk student population in a multi-cultural urban high school.
International Journal of Educational Management , 19 (5), 437 - 449.
Cooley, C. (1912). Human Nature and the Social Order. New York: Scribners.
Coopersmith, S. (1967). The Antecedents of Self-Esteem. San Francisco: Freeman.
Davies, R. P. (1975). Mixed Ability Grouping. London: Maurice Temple Smith Ltd.
DeLisi, & Goldbeck, S. (1999). Implication of Piagetian theory for peer learning. In A.
O'Donnel, & A. Kings (Eds.), Cognitive Perspectives on Peer Learning (pp. 3-37).
Mahwah, NJ: Lawrence Erlbaum.
Festinger, L. (1954). A Theory of Social Comparison Process. Human Relations , 7, 117140.
Gamoran, A. (1992). Is Ability Grouping Equitable? Leadership , 50, pp. 11-13.
Good, T. L. (1981). Teacher's Expectancy and Students' Perception: A decade of research.
Educational Leadership , 38 (5), 415-422.
Goodland, J. I. (1983). A place called school. New York: McGraw-Hill.
99
Graham, S., & Weiner, B. (1996). Theories and Principles of Motivation. In D. Berliner, &
R. Calfee (Eds.), Handbook of Educational Psychology (pp. 63-84). New York, NY:
MacMillan.
Hagborg, W. (1993). Rosenberg Self-Esteem Scale and Harter's Self-Perception Profile for
Adolescents: A concurrent validity study. Psychology in Schools , 30, 132-136.
Hallam, S., Ireson, J., & Davies, J. (2002). Effective Pupil Grouping in the Primary School.
London: David Fulton Publishers Ltd.
Hallam, S., Ireson, J., & Davies, J. (2002). Effective Pupil Grouping in the Primary School:
A Practical Guide. London: David Fulton Publishers Ltd.
Hendricks, K. B. (2009). The impact of ability grouping on the achievement, self-efficacy,
and classroom perceptions of gifted elementary students. Walden University Journal .
Higgins, T. E. (1999). Persons and situations: Unique explanator principles or variability in
general principles? In D. Cervone, & Y. Shoda (Eds.), The Coherence of Personality (pp.
61-93). New York, NY: Guilford Press.
Horrocks, J. E., & Jackson, D. W. (1972). Self and Role: A Theory of Self-Process and Role
Behavior. Boston: Houghton Mifflin Company.
Ireson, J., & Hallam, S. (2001). Ability Grouping in Education . London: Sage.
Ireson, J., & Hallam, S. (2001). Ability Grouping in Education. London: Sage.
Katz, P., Zigler, E., & Zelk, S. (1975). Children's self-image disparity. Developmental
Psychology , 11, 546-50.
Kulik, C., & Kulik, J. (1982). Effects of Ability Grouping on Secondary School Students: A
Meta-Analysis of Evaluation Findings. American Educational Research , 19, 415-428.
Kulik, J. A. (2004). Grouping, Tracking, and De-Tracking. In H. J. Walberg, A. J. Reynolds,
& M. C. Wang (Eds.), Can Unlike Students Learn Together? Grade Retention, Tracking,
and Grouping (pp. 157-182). Greenwich: Information Age Publishing Inc.
Kulik, J. (1992). An Analysis of the Research on Ability Grouping: Historical and
contemporary perspectives. Research-Based decision making series. Storrs, USA: National
Research Center on the Gifted and Talented, University of Connecticut.
Larsen, R. J., & Buss, D. M. (2008). Personality Psychology: Domains of Knowledge abou
Human Nature. New York, NY: McGraw-Hill.
100
Loveless, T. (1998). The Tracking and Ability Grouping Debate. Washington DC: The
Thomas B.Fordham Foundation.
Malle, B. (2007). The Actor - Observer Assymmetry in Attribution: A (surprising) meta
analysis. Psychological Bulletin .
Markus, H., & Nurius, P. (1987). Possible Selves: The interface between motivation and the
self concept. In K. Yardley, & T. Honnes (Eds.), Self and Identity: Psychosocial
Perspectives (pp. 157-172). Chichester,, England: John Wiley & Sons.
Marlowe, B. A., & Canestrari, A. S. (2006). Educational Psychology in Context. Thousands
Oaks, CA: Sage Publicationns.
Marsh, H., & Parker, J. (1984). Determinants of student self-concept: Is it better to be a
relatively large fish in a small pond even if you don't learn to swim as well? Journal of
Personality and Social Psychology , 47, 213-231.
Maslow, A. (1954). Motivation and Personality. New York: Harper & Row.
Mead, G. (1934). Mind, Self, and Society. Chicago: University of Chicago Press.
Myers, D. G. (2008). Social Psychology. Holland, MI: McGraw Hill.
Oakes, J. (1985). Keeping Track: How Schools Structure Inequality. New Haven, CT: Yale
University.
O'Donnell, A. M., Reeve, J., & Smith, J. K. (2007). Educational Psychology: Reflection for
Action. Hoboken, NJ: John Wiley & Sons.
Oyserman, C. D., & Markus, H. (1990). Possible selves in balance: Implications for
delinquency. Journal of Social Issues , 46, 141 - 157.
Rogers, C. R. (1980). A way of being. Boston: Houghton Mifflin.
Rohani, A. H. (1998). Persepsi Guru yang Berkualiti: Satu Tinjauan di Daerah Kulim.
Sintok, Kedah DA: Sekolah Siswazah Universiti Utara Malaysia.
Rosenthal, R. (2002). Covert communication in classrooms, clinics, courtrooms, and
cubiclres. American Psychologist , 57, 839-849.
Rosenthal, R., & Jacobson, L. (1968). Pygmalion in the classroom: Teacher expectation and
pupils' intellectual development. New York: Holt, Rinehart, & Winston.
101
Santrock, J. W. (2005). Psychology, Updated Seventh Edition. New York, NY: McGrawHill.
Schunk, D. H. (1996). Learning Theories 2nd edition. New Jersey: Prentice Hall.
Slavin, R. E. (1987). Ability Grouping and Student Achievement in Elementary Schools.
Review of Educational Research , 57, 293-336.
Slavin, R. E. (1990). Achievement effects of ability grouping in secondary schools: A best
evidence synthesis. Review of Educational Research , 60, pp. 471-499.
Slavin, R. E. (2006). Educational Psychology; Theory & Practice 8th edition. Boston: Allyn
& Bacon.
Stryker, S. (2002). Symbolic Interactionism: A Social Structural Version. Caldwell, NJ:
Blackburn Press.
Tam, C. L. (2003). EFFECTS OF FAMILY FUNCTIONING AND FAMILY HARDINESS
ON SELF-EFFICACY AMONG COLLEGE STUDENTS. Sunway Academic Journal (4),
99-107.
Von Der Haar, C. M. (2005). Social Psychology: A Sociological Perspective. Upper Saddle
River, NJ: Pearson Education.
Vygotsky, L. (1978). Mind in society: The development of higher psychological processes.
Cambridge, MA: Harvard University Press.
Walberg, H. J. (2004). Recommendations and a Personal View. In H. J. Walberg, A.
J.Reynolds, & M. C. Wang (Eds.), Can Unlike Students Learn Together? Grade Retention,
Tracking, and Grouping (pp. 203-207). Greenwich, USA: INformation Age Publishing Inc.
Woolfolk, A. (2007). Educational Psychology 10th Editio. Boston, MA: Pearson Allyn &
Bacon.
ZaidatunTasir, & MohdSallehAbu. (2003). Analisis Data Berkomputer SPSS11.5 for
Windows (1st edition). Kuala Lumpur: Venton Publishing (M) Sdn. Bhd.
102
APPENDIX A
LETTER OF APPROVAL
103
APPENDIX B
SCALES
104
Self-Esteem Scale
1.
I feel that I am a person of worth, at least on an equal plane
with others.
Saya seorang yang berguna, sekurang-kurangnya setanding
dengan orang lain.
1
2
3
4
2.
I feel that I have a number of good qualities.
Saya memiliki nilai-nilai kualiti yang baik.
1
2
3
4
3.
All in all, I am inclined to feel that I am a failure.
Secara amnya, saya cenderung untuk merasakan saya
seorang yang gagal.
1
2
3
4
4.
I am able to do things as well as most other people.
Saya mampu melakukan sesuatu sebaik orang lain
1
2
3
4
5.
I feel I do not have much to be proud of.
Saya rasa saya tiada apa yang boleh dibanggakan.
1
2
3
4
6.
I take a positive attitude toward myself.
Saya mempunyai sikap yang positif terhadap diri saya.
1
2
3
4
7.
On the whole, I am satisfied with myself.
Secara keseluruhan, saya berpuas hati dengan diri saya
sendiri.
1
2
3
4
8.
I wish I could have more respect for myself.
Saya berharap supaya saya diberi lebih penghormatan
1
2
3
4
9.
I certainly feel useless at times.
Kadang-kadang saya berasa tidak berguna
1
2
3
4
10.
At times I think I am no good at all.
Kadang kala saya berfikir saya tidak bagus sama sekali
1
2
3
4
105
General Self-Efficacy Scale (GSE)
1
2
3
4
5
6
7
8
9
10
I can always manage to solve difficult problems if I try hard
enough.
Saya sentiasa mampu mengatasi masalah-masalah yang
sukar jika saya bersungguh-sungguh
If someone opposes me, I can find the means and ways to
get what I want.
Jika sesiapa cuba menghalang saya, saya mampu mencari
jalan untuk mendapat apa yang saya inginkan.
It is easy for me to stick to my aims and accomplish my
goals.
Senang bagi saya untuk berpegang pada matlamat dan
mencapai cita-cita
I am confident that I could deal efficiently with unexpected
events.
Saya yakin yang saya mampu mengatasi sesuatu yang
tidak dijangka.
Thanks to my resourcefulness, I know how to handle
unforeseen situations.
Disebabkan saya berkebolehan, saya tahu bagaimana
untuk mengendalikan situasi yang belum pernah berlaku.
I can solve most problems if I invest the necessary effort.
Saya mampu menyelesaikan masalah saya jika saya
berusaha bersungguh-sungguh.
I can remain calm when facing difficulties because I can
rely on my coping abilities.
Saya sentiasa tenang walaupun dilanda kesusahan kerana
saya yakin dengan keupayaan saya.
When I am confronted with a problem, I can usually find
several solutions.
Apabila berhadapan dengan masalah, saya mampu
mencari beberapa jalan penyelesaian.
If I am in trouble, I can usually think of a solution.
Apabila dilanda masalah, saya mampu memikirkan jalan
penyelesaian.
I can usually handle whatever comes my way.
Saya mampu menguruskan apa-apa masalah yang
menghalang saya.
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
106
Scale of Perception on Teachers’
Behavior
1
2
3
4
5
6
7
8
9
10
B
1
2
3
4
5
6
7
8
9
10
My teachers are…
Guru-guru saya…
praising good behavior, regardless how good/bad the academic performance is
sentiasa memuji pelajar atas tingkah laku baik mereka, tanpa mengira pencapaian
akademik mereka.
paying attention to students with disciplinary problems
memberi perhatian terhadap pelajar yang mempunyai masalah disiplin
often asking for total silence during class
sering meminta supaya murid-murid diam di dalam kelas
frequently punishing misbehavior student
sering mengambil tindakan terhadap kesalahan tingkah laku pelajar
likely to prefer silent students than students who actively asking academic questions
lebih menggemari pelajar yang diam daripada pelajar yang suka bertanya akan soalan
mengenai akademik.
frequently giving disciplinary warning
sering memberi amaran disiplin
more likely to spot disciplinary problems, no matter how small they are
seringlebih cenderung untuk mengenalpasti masalah disiplin walaupun kecil.
expecting students to be discipline, without focusing on academic performance
mengharap murid lebih bertingkah laku disiplin tanpa menumpu kepada pencapaian
akademik
more likely to scold students with discipline problem, instead of students with low
academic performance
lebih gemar memarahi pelajar yang bermasalah disiplin berbanding pelajar yang
mempunyai pencapaian akademik yang rendah
seldom praising students with good exam result
jarang memuji pelajar yang mempunyai prestasi akademik yang baik.
My teachers are…
Guru-guru saya…
often encouraging active academic discussion among the students
sering menggalakkan pelajar supaya aktif dalam perbincangan akademik
often asking questions about the subject to check the comprehension
sering bertanya soalan tentang mata pelajaran untuk menyemak pemahaman pelajar
more likely to give learning advice, regardless the students’ disciplinary problems
memberi nasihat mengenai pembelajaran, tidak kira masalah disiplin yang dipunyai
pelajar
more likely to praise on good performance more than good behavior
lebih gemar memuji pencapaian akademik yang baik berbanding tingkah laku yang
baik
more likely to remember students by academic performance, instead of by disciplinary
matters
lebih mengingati murid berdasarkan prestasi akademik berbanding berdasarkan
masalah disiplin
pay less attention to small disciplinary problem to students with high academic
achievement
tidak mengambil kira masalah disiplin yang ringan terhadap pelajar yang mempunyai
prestasi akademik yang baik
often giving oral quizzes and letting students to answer questions
sering memberi kuiz secara lisan untuk pelajar menjawab
giving enough time for a student who’s trying to answer question
memberi masa yang cukup kepada pelajar yang cuba menjawab soalan
more likely to warn low-performers about the bad marks
kerap memberi peringatan kepada pelajar yang mendapat markah rendah
pay more attention to students with high academic achievements, regardless of her/his
disciplinary records
memberi lebih perhatian terhadap pelajar yang menunjukkan pencapaian akademik
yang baik, tanpa mengambil kira rekod disiplin mereka
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
107
Questionnaire Set
Set Soal Selidik
English
Thank you for your participation on this study. This study is investigating your level of selfesteem, self-efficacy, and the perception on teachers’ behavior. As you see in the
Respondent Personal Information blank above, your identity and any information you
provide in the questionnaire set are confidential. To enter your reply, simply give X sign on
any box that representing your opinion based on scale given:
1 = Not at all true
2 = Hardly true
3 = moderately true
4 = exactly true
Respondent Personal Information:
School:
Class:
Bahasa Malaysia
Terima kasih di atas penglibatan anda dalam kajian ini. Matlamat kajian ini adalah untuk
mengetahui tahap keyakinan diri (self-esteem), kebolehan diri (self-efficacy), serta persepsi
pelajar terhadap tingkah laku guru. Identiti dan sebarang maklumat yang anda berikan
dalam borang ini adalah sulit.
Sila pangkah (X) pada kotak yang sesuai dengan pendapat anda mengikut skala seperti
diterangkan:
1 = tidak setuju
2 = agak setuju
Maklumat Peribadi Responden:
Sekolah:
Kelas:
3 = setuju
4 = sangat setuju
108
APPENDIX C
VALIDATION LETTER