Class Size, Class Ranking and Students’ Achievement

CLASS SIZE, CLASS RANKING AND STUDENTS' ACHIEVEMENT

INSTITUTE OF RESEARCH, DEVELOPMENT AND COMMERCIALISATION UNIVERSITITEKNOLOGI MARA 40450 SHAH ALAM, SELANGOR

MALAYSIA

BY:

NADZIRA YAHAYA MOHD. HALIM KADRI

NAIMAH AHMAD YAHYA

NOVEMBER 2003

i

COPYRIGHT @ UiTM

CLASS SIZE, CLASS RANKING AND STUDENTS' ACHIEVEMENT

BY:

NADZIRA YAHAYA MOHD. HALIM KADRI

NAIMAH AHMAD YAHYA

NOVEMBER 2003

ii

COPYRIGHT @ UiTM

Surat Kami: 100-CJ(AKA 9/10/7)Jld. 2 ( 5 )

Tarikh: 22 Julai 2002

UNIT PENYELIDIKAN DAN PERUNDINGAN I Universiti Teknologi MARA

Cawangan Johor Kampus Segamat

Beg Berkunci 527 55009 Segamat

Johor Darul Takzim

Tel: 07-9352085 Fax 07-9352288

tankuasa:

Dr. Hj Mohd Sahar awiran erusi

Luhaya bt Hj. Atan >alan Pengarah pus HEA >. Pengerusi

Madya Arun b. imed dinator

4ohd Yusof bin apa lahari

Chairul Anuar bin L. Rahim usaha

Jawatankuasa:

Madya Dr. Omar amat yarah Kewangan

Vhmad bin Che ob yarah PPI

Che Faridah bt Mahmood yarah Pengurusan

Puan Nadzira binti Yahaya Pensyarah Universiti Teknologi MARA Cawangan Johor Kampus Segamat

Puan

CLASS SIZE , CLASS RANKING AND STUDENTS' ACHIEVEMENT

Dengan hormatnya perkara tersebut di atas dirujuk.

Sukacita dimaklumkan bahawa Mesyuarat Jawatankuasa Penyelidikan dan Perundingan Bil. 2/2002 pada 8 Julai 2002 telah mengambil keputusan:-

i. Bersetuju meluluskan cadangan penyelidikan yang dikemukakan oleh puan dan Encik Mohd Halim bin Kadri serta Puan Naimah binti Ahmad Yahya

ii. Tempoh projek penyelidikan ini ialah 12 buian, iaitu mulai 1 Ogos 2002 hingga 31 Julai 2003.

iii. Kos yang diluluskan ialah sebanyak RM 3, 274.00 sahaja.

iv. Penggunaan geran yang diluluskan hanya akan diproses setelah perjanjian ditandatangani.

v. Kertaskerja boleh dibentangkan dalam seminar setelah 75% deraf awal laporan akhir projek dihantar ke UPP untuk semakan. Walau bagaimanapun puan perlu membuat permohonan kepada unit Penyelidikan dan Perundingan.

vi. Pihak puan dikehendaki mengemukakan Laporan Kemajuafi Projek Penyelidilcan bagi tempoh sehingga 1 Disember 2002 dan 1 April 2003. Laporan akhir perlu dihantar sebaik sahaja projek penyelidikan disiapkan. Format menulis laporan akhir boleh diperolehi di Unit Penyelidikan dan Perundingan.

iii

COPYRIGHT @ UiTM

Bersama-sama ini disertakan perjanjian untuk ditandatangani. Sila penunkan perjanjian berkenaan dengan menggunakan pen dakwat hitam dan kembalikan ke unit Penyeiidikan dan perundingan.

sekian, terima kasih.

"SELAMAT MENJALANKAN PENYELIDIKAN

Yang benar

cOF. Pengarah K Universiti Tikno/ogi MARA Cawangan Joho Kampus segam

II MOHD SAHAR BIN SAWIRAN

S.k 1) Koordinator unit Penyeiidikan dan Perundingan universiti Teknoiogi MARA cawangan Johor Kampus Segamat

2) Penolong Bendahari universiti Teknoiogi MARA Cawangan Johor Kampus segamat

s4CW^daf/iiruliitd«fLitl

iv

COPYRIGHT @ UiTM

Fakulti Perakaunan Universiti Teknologi MARA Cawangan Johor

Kampus Segamat

85000 Segamat, Johor.

Tarikh 3 November 2003

No. Fail Projek

Penolong Naib Canse\or (Penyelidikan)

Biro Penyelidikan dan Perundingan (BRC)

UITAA, Shah A lam

YBhg. Prof

LAPORAN AKHIR PENYELIDIKAN "CLASS SIZE, CLASS RANKIN© AND STUDENTS' ACHIEVEMENT"

Merujuk kepada perkara di atas, bersama-sama ini disertakan 3 (tiga) naskah

Laporan Akhir Penyelidikan bertajuk "Class size, class ranking and students'

achievement" oleh kumpulan Penyelidik UiTM Cawangan Johor, Kampus Segamat.

Sekian, terima kasih.

Vang benar,

NADZIKA-VAHAYA Ketua

Projek Penyelidikan

COPYRIGHT @ UiTM

KUMPULAN PENYELIDIK

NADZIRA YAHAYA

KETUA PROJEK

MOHD. HALIM KADRI Ahli

Tandatangan

N A I M A H AHMAD YAHYA Ahli

Tandatangan

vi

COPYRIGHT @ UiTM

Acknowledgement

In the Name of Allah Most Gracious Most Merciful

We wish to acknowledge the contribution of many people who have in one way or another

made this research a reality. Our greatest appreciation goes to Prof. Madya Dr. Mohd

Hashim Mohd. Kamil, Director of UiTM Cawangan Johor for his encouragement and giving us

the opportunity to conduct this research. We would like also to thank Prof. Dr. Rahim bin

Md. Sail and Prof. Dr. Shawaluddin bin Anis for their constructive criticisms, comments and

suggestions.

We are indebted to the University Teknologi MARA, particularly the Institute of Research,

Development and Commercialisation (IRDC) for giving us the opportunity and support to

conduct this research. Special mention and thanks to Cik Normah and Puan Jasbir for their

patience, time and effort in editing this manuscript. Last but not least, we are grateful to

our colleagues, students, friends and family for their support, understanding and

cooperation in helping us complete this research.

vii

COPYRIGHT @ UiTM

Abstract

This study investigates the effect of class size and class ranking on the

achievement of accounting students. T test and regression analyses were used to

analyze the data. The results of the t-test on class size indicated that students in

small classes performed better than students in large classes in the on-going

assessment whereas students in large classes performed better than students in

small classes in the final exam. The T test on class ranking indicated that 'ranking

1' class students had consistently outperformed their counterparts in 'ranking 2'

classes in the on-going assessment, final exam and overall achievement.

Regression analysis indicated that class size, class ranking and on-going

assessment contributed significantly to the prediction of students' overall

achievement.

viii

COPYRIGHT @ UiTM

LIST OF TABLES

Table Title Page

1 Respondents by size of class 20

2 Respondents by ranking of class 20

3 Cross-tabulation between class size and class ranking 21

4 T test of students achievements by class size 22

4a T test of students' achievements by class size (reading 24

courses)

4b T test of students' achievements by class size 25

(calculation courses)

5 T test of students' achievements by class ranking 26

6 The summary of multiple regression result for the 29

whole sample

ix

COPYRIGHT @ UiTM

Table of Contents

Title /

Letter of Approval Hi

Letter of Submission

Acknowledgement vii

Abstract viii

List of Tables ix

Table of Contents x

CHAPTER 1 : 1.0 INTRODUCTION 1

2.0 PROBLEM STATEMENT 3

3.0 OBJECTIVES OF THE STUDY 4

CHAPTER 2 : 4.0 LITERATURE REVIEW 5

4.1 Class Size and Class Ranking 5

4.2 The effect of class size and class

ranking on students' achievement 6

CHAPTER 3 : 5.0 RESEARCH METHODOLOGY 16

5.1 Population 16

5.2 Data collection 16

5.3 Variables 17

5.4 Data analysis 17

COPYRIGHT @ UiTM

CHAPTER 4 : 6.0 FINDINGS 20

6.1 Sample characteristics 20

6.2 Hypothesis testing 21

6.3 Multiple regression 28

CHAPTER 5 : 7.0 CONCLUSIONS AND

RECOMMENDATIONS 30

Bibliography 33

xi

COPYRIGHT @ UiTM

. Class Size. Class Ranking And Students' Achievement

CHAPTER ONE

1.0 INTRODUCTION

Class size is still an educational issue today. While many research studies indicate

that class size is unimportant, others seem to feel that it is. Studies on the effects

of class size have been conducted since the 1900s and have produced mixed

findings with some methods of instruction favouring small classes and other

methods being as or more effective in large classes.

Class size is not the same thing as the pupil-teacher ratio or student-lecturer ratio.

Those ratios are global measures of the human resources brought to bear, directly

and indirectly, on the student's learning. Class size refers to the actual number of

pupils or students taught by a teacher or a lecturer at a particular time in a

particular class.

Determining class size seems simple enough, but it is actually more difficult and

more important than it appears. Class size involves organizing students for the

delivery of instruction, whereas pupil-teacher ratio is an administrative statistic

that helps account for the distribution of resources. The approximate difference

between pupil-teacher ratio and class size in U.S. school is 10. That is, if the

1

COPYRIGHT @ UiTM


school's pupil-teacher ratio is 17 to 1, then most teachers in the building will

have, on average of 27 students per class. (Achilles et. al. 1998)

There is no fixed number in a class size. Normally, the size of the class depends

on the discipline- a smaller number for engineering, science and medicine and a

larger number for the arts, humanities and social sciences. There is no agreed

definition of a large class in the literature, nor should there be. One person's large

class is what some others consider as 'regular', 'small' or 'normal'. Some

lecturers or teachers simply define 'large' as 'too many students whose names are

forgotten by the end of the term or semester'.

The number of students in the class can vary and has the potential to affect how

much is learned in a number of different ways. Class size could most probably

affect how students interact with each other. Besides that it could affect how

much time the lecturer is able to spend to focus on individual students and their

specific needs rather than on the group as a whole. It may also affect the

lecturer's allocation of time and teaching effectiveness.

Thus, changes to the class size are considered a potential means of changing how

much students can learn. However the amount of student learning is dependent on

many other factors such as school environment, students' background, attitude,

motivation, individual instruction and broader community influences.

2

COPYRIGHT @ UiTM


2.0 PROBLEM STATEMENT

Based on the previous studies, there are three contrasting school of thought on the

effect of class size on the students performance;

1. Glass et.al (1979), Nachman et.al (1958), Christine et.al (1994) and Mohd.

Halim

(2001) found that large class size has a negative effect on students

performance.

2. William et.al (1985) found that class size has inconsequential effect on

students performance.

3. Callahan (1998) found that students taught in large class performed better

than students taught in small class.

The effect of class size on students performance requires further examination,

because the prior research has found mixed results. Further, this issue is important

to the faculty and administrator because increasing class size is one of the most

obvious responses to tightened budgets. Another reason for further research is that

many educators believe that large class will decrease students performance and

small classes will have better performance.

3

COPYRIGHT @ UiTM


The purpose of this study is to further investigate the effect of class size and class

anking on the achievement of accounting students in UiTM Johor. This study

could contribute more to our understanding of the relationship between class

sizes, class ranking and students' achievement in several ways although there are

many other variables which are not taken into account in this study.

3.0 OBJECTIVES OF THE STUDY

The objectives of the study are as follows:

1. To identify whether size of the class affects students' achievement.

2. To identify whether class ranking affects students' achievement.

4

COPYRIGHT @ UiTM


CHAPTER TWO

4.0 LITERATURE REVIEW

4.1 Class Size and Class Ranking

Otto (1954) in his research selected samples of 50 small and 50 large

elementary school classes, where the median enrollment was 23 for small

classes and 37 for large classes. On the other hand, Nelson (1959) in his

studies used small classes ranging from 16 to 20 and large classes from 85

to 141. Whereas Frymier (1964-65) in his study on the effect of class size

on first grade reading achievement, chose class-rooms with over 36 pupils

and class-rooms under 30 pupils in 12 schools.

Balow (1969) in his study used 15 students for his experimental group

while the control group consisted of 30 students. Raimondo et al. (1990)

further investigated the influence of small (25 to 35 students) versus large

(200 to 350 students) class size on the students achievement for economic

courses.

Christine et al. (1994) defines a small class as a class, which consists of a

maximum of 38 students whereas a large class as a class which consists of

150 or more students. On the other hand, Krueger (1999) defines small

5

COPYRIGHT @ UiTM


class size as a class, which consists of 13 to 17 students, while normal

class size is a class, which consists of 22 to 25 students.

In the case of UiTM Segamat Campus, a small class is defined as a class

consisting of not more than 29 students and a large class is a class

consisting of 30 or more students (Mohd. Halim, 2001). Class ranking is a

class that consists of students grouped according to their previous

semester's achievement based on Cumulative Grade Point Average

(CGPA).

4.2 The effect of class size and class ranking on students' achievement

The relationship between class size and student achievement has been, and

still is, a controversial educational issue. The effect of class size on student

achievement has been researched in a variety of disciplines. Class size has

also been studied extensively in elementary and secondary education and

has produced mixed findings.

The results of investigations conducted by Rice (1903) indicated that, in

general, unless elementary school classes exceed fifty there is no clear

evidence of less efficiency, and, he concluded that there were no

differences in achievement attributable to class size. In addition, a study

conducted by Spitzer (1954) in Iowa, concluded that there were no

6

COPYRIGHT @ UiTM


differences indicated on skills such as arithmetic, reading, language and

study skills between large classes and small classes (Hollingworth, 1992).

In 1954, a thorough study of class size factors was made by Otto and he

reported that the environment was better in small classes, but as far as

achievement was concerned there were no differences between small or

large classes. The study to examine the effect of class size on first grade

reading achievement by Frymier (1964-65), found that the achievement of

pupils in smaller classes was greater than the achievement of those in large

classes. McKeachie (1986) also noted that the larger the group the smaller

the proportion that can participate orally, and the less likely that an

individual will feel free to contribute. Because active thinking is so

important to learning and retention of learning, achievement may be less

in larger classes (Kennedy & Siegfried, 1997). Large class size is also

presumed to lower a student's sense of personal responsibility and activity

(McKeachie, 1990). Students may feel lost in a crowd or may lose interest

in the class because of the number of classmates within a particular course

(Scherk et.al, 1994).

The results of the Tennesse's Student/Teacher Ratio (STAR) project as

reported by Gursky and Daniel (1993) revealed that students in the smaller

classes did much better in reading and mathematics than students in

7

COPYRIGHT @ UiTM

Class Size, Class Ranking And Students' Achievement

regular classes. On the other hand, Woodson (1968) reported a study

involving fourth and sixth grade students taught in self-contained

classrooms in 95 school districts where test results from math and reading

subtests were used as the criterion. No significant differences were found

in all areas questioned by this study (Woodson, 1968).

In 1962, The Riverside Unified School District in California initiated a

study on two different groups of students, which encompassed seven

elementary schools with children in first through-third grades (Balow,

1969). Class size was reduced to 15 for the experimental group while the

control group remained at 30 students. The study concluded that students

in the experimental group for two or more years scored significantly

higher than pupils in the control group when first and second grade test

scores were analyzed. However, no significant differences were found

when third and fourth grade tests were analyzed (Hollingworth, 1992).

Weber (1971) in his study reported on four schools that were successful in

teaching inner-city school children to read. The class size of the four

schools was 22 to 29. He indicated that the class size was not the

determining factor but the extra teachers, specialists and teacher's aides

that reduced the student/teacher ratio was the important factor. However,

COPYRIGHT @ UiTM


Glass and Smith (1979) found that little gain in achievement could be

expected by reducing class size from 40 students to even as few as 25.

In the study conducted by Vanble and Gilman (1988) on first grade

classrooms in Indiana, ten sets of data collected from over 2,000 first

grade classrooms was examined. The results indicated that only three of

the ten studies reported a positive relationship between decreased class

size and increased achievement scores. The researchers concluded that

class size reduction is necessary, but not sufficient to increase

achievement scores.

In a meta-analysis of the relation between class size and achievement

conducted by McGiverin, Gilman and Tillitski (1989), it was concluded

that second grade pupils in smaller classes (19.1 mean class size) had

significantly higher achievement test scores than did students in larger

classes (26.4 mean class size). This analysis involved second grade

students in reading and math with an effect size of .34 standard deviations

for the smaller classes. This indicated that second grade children leam

more effectively in smaller classes (Hollingworth, 1992).

On the other hand, Raimondo et.al (1990) concluded that students in a

large intoductory class performed as well as did students in a small

9

COPYRIGHT @ UiTM


introductory class. Furthermore, students who had studied in a large

introductory class performed as well in a subsequent intermediate class

that did not require higher level cognitive skills as did students who had

initially studied in a small introductory class. However, students from the

large introductory class did not perform well on a subsequent theoretical

intermediate course, which required higher-level cognitive skills. This

finding is consistent with that of Lewis and Dahl (1972), who found that

class size did not have a negative effect on student assessment

performance involving multiple choice questions but did have a negative

effect on performance in essay questions. (Gibbs and Lucas, 1996)

Large classes are also found to be as effective as small classes when the

goals involve learning and comprehending factual information. This could

be evidenced by research carried out by Callahan (1998) on class size and

students performance in introductory accounting courses. He found that

students in large classes outperformed students in smaller classes when

other explanatory variables are considered (attendance, Grade Point

Average).

On the other hand, large class size is found to have negative effect (Glass

and Smith, 1979) and inconsequential effect (William Cook, Quinn and

Jensen, 1985) on students' performance. Large class size had a significant

10

COPYRIGHT @ UiTM


negative effect on specific performance measures such as quizzes or final

exams and no effect on overall performance (Nachman and Opochinsky,

1958) and Christine et.al (1994). Charles Achilles argued that reducing

class size fosters better teaching (Achilles, 1999). His claims reflected the

conclusions of several earlier reviews. (Glass, & Smith, 1980; Glass,

Cahen, Smith, & Filby; Cooper, 1989)

Bowman (1979) and McKeachie (1980) debated the use of class size and

the effect on students' performance. Research investigating the

relationship between class size, which is defined as the ratio of students to

instructors (Glass and Smith, 1980) and students' performance produced

inconsistent results. The effect of class size on achievement is most likely

to occur if class size is linked to instruction (Barr, & Dreeben, 1983).

Normally, class size reductions may result in a different approach of

teaching as compared to larger classes, such as more frequent assessments,

more writings, more discussions and more help for individual students that

would raise the students' achievement. In this context, the direct cause of

this achievement increase would be instructional improvements and class

size would be the indirect cause. In fact, even teachers do not change

instructional practices; certain practices may work better in smaller

classes. For example, students may pay attention when there are fewer

students in the room.

11

COPYRIGHT @ UiTM

K g Class Size, Class Ranking And Students' Achievement

A study conducted by Williams et.al (1984) at Brigham Young University

of 27 courses on increasing class size from 20 to 40 students up to several

hundreds of students, indicated that class size has a much less important

influence on students' test performance. However, many lecturers of large

classes feel they spend too much time organising and managing class

activities and not enough time meeting the needs of the individual

learners. Large classes and overcrowded classrooms have negative effects

on students' behavior and learning.

A study by Gibbs and Lucas (1996) on student performance and class size

over a period of 10 years (1984-94) involving over 5,000 modules and

250,000 students' grades found a small but highly significant negative

relationship between module enrolment and grades. Students in larger

classes stood significantly lower chances of getting good grades. Smaller

classes seem to be related to higher achievement in the primary grades and

upper elementary grades as well as have positive effects on students'

behavior. (Hollingworth, 1992) The most beneficial effect of smaller

classes seems to be for pupils' achievement in reading and math. It was

found that it is more effective when instructional goals involve higher-

level cognitive skills including application, analysis, and synthesis.

12

COPYRIGHT @ UiTM


Besides, there are many other reasons why smaller classes might

contribute to higher achievement, which include greater opportunities

between students and lecturers for interaction regarding subject matter. In

addition it is most needed for students with low motivation and those with

little knowledge of the subject matter, or have difficulty in grasping

conceptual material.

So while the debate about size of class has often been in terms of reduced

size of class resulting in students' academic gains, it is also important

educationally to consider the effect of class ranking towards students'

performance. Betts and Shkolnik (2000) in their study on class ability ,

concluded that there were little or no differential effects of grouping for

high-achieving, average, or low-achieving students towards math

achievement growth. Slavin (1990) also found that ability grouping has

little or no overall effect on achievement.

On the other hand, Kerckhoff (1986) in his study pointed out that students

in the high ability class do better than the average students at an

ungrouped school, and students in a low ability class at a grouped school

do worse than the average student at an ungrouped school. In addition,

Ashbaugh et. al (1993) found that those students' high school class rank

and class size was to be statistically significant contributors to the

13

COPYRIGHT @ UiTM

Class Size, Class Ranking and Students’ Achievement

Documents

Transcript of Class Size, Class Ranking and Students’ Achievement