.i..3l‘l£a.ii€ 51‘
.- .é..3§.€.il..i: .3
£137.52...

0.. 3....
.‘It0133s
ltzlx

7.51111... (1.1;.

.inlxi‘ldn
jatszialto

l1 . {.2 {lasxillziiat

I; )tz..i1...3..tl......

1133. 5‘13?!

(1-1.... (‘32:... :

 

THES‘

Willi/llillilll/lllfi/llI/lli/i/ll/ll/I

3 1293 00908 7739

This is to certify that the

dissertation entitled

Assessing the Competency of Teachers,
Curriculum Specialists, and Prospective

Teachers in Educational Measurement in
Bahrain.

presented by

Rashid Hammad Aldosary

has been accepted towards fulfillment
of the requirements for

Ph.D. mgmem Measurement,

Evaluation, and
Research Design.

    

ajor professor

Date la i ad [Eli]

MSU is an Affirmative Action/Equal Opparluniry Institution

0-12771

    
    

“BRARY
Michigan State
University

l3LACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.

 

MSU Is An Affirmative Action/Equal Opportunity Institution
c:\circ\dateduolnm3-D.1

ASSESSING THE COMPETENCY OF TEACHERS, CURRICULUM
SPECIALISTS, AND PROSPECTIVE TEACHERS IN
EDUCATIONAL MEASUREMENT IN BAHRAIN

BY

Rashid Hammad Aldosary

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Department of Counseling, Educational Psychology and

Special Education

1993

 

 

 

ABSTRACT
ASSESSING THE COMPETENCY OF TEACHERS, CURRICULUM
SPECIALISTS, AND PROSPECTIVE TEACHERS IN EDUCATIONAL
MEASUREMENT IN BAHRAIN
BV
Rashid Hammad Aldosary

Educators in general and teachers in particular use classroom tests to
make a variety of decisions.

This study was conducted to assess the competency of Bahrain teachers,
curriculum specialists, and prospective teachers in educational measurement, to
detect their strengths and weaknesses in specific areas in educational
measurement (such as item analysis and score interpretation) and to investigate
the relationship between competency in educational measurement and classroom
testing activities, measurement preparation, and background variables.

Two research instruments were used in this study: a measurement test (65
multiple-choice items) and a questionnaire (23 items). The two research
instruments were administered to 1300 in-service teachers. 146 curriculum
specialists, and 221 prospective teachers. The response rate was 82 percent for
the teachers, 67 percent for the curriculum specialists, and 52 percent for the

prospective teachers. The researcher was unable to generalize to the population of

prospective teachers.

The three groups in the study had a reasonable amount of measurement

and statistics coursework. with more measurement coursework than statistics.

 

 

 

 

About one third of the teachers had never been exposed to statistics. All three
groups need more training in measurement and statistics. The subjects, in general,
demonstrated about average performance on the educational measurement test,
except for the teachers who scored below average. The mean test score was
26.72 (41 percent) for the teachers. 37.49 (58 percent) for the curriculum
specialists, and 34.07 (52 percent) for the prospective teachers. The scores on the
measurement test ranged from 9 to 48 for the teachers, from 19 to 50 for the
curriculum specialists, and from 18 to 49 for the prospective teachers. The results
indicated that the three groups need training in measurement in general, and
especially the areas in which they were weak, such as "item analysis, “score
interpretation", and "standard error of measurement".

About half of the teachers spend more than 20 percent of their
professional time on classroom testing activities. The most preferred item type for
the teachers and prospective teachers was the essay, whereas the completion type
was the most preferred item type for the curriculum specialists.

Correlational results revealed that there was a significant relationship
(P<.05) between competency in educational measurement and several variables
such as purposes of using classroom tests, developing a test plan, conducting item
analysis, and self-assessment of the general competency in measurement. Yet the
proportion of variance in the measurement test accounted for by each of these
variables was small or negligible.

ANOVA results showed that there was a significant difference in

measurement competency among the three groups. Also, female teachers were
more competent in measurement than male teachers. Those who had
measurement and statistics coursework manifested better performance on the
measurement test than those with no measurement and statistics at all (except for

the prospective teachers with respect to statistics). These results also apply to the

 

 

test plan and item analysis self-ratings, but for the teachers only. The gender of
respondent had an effect on the competency in educational measurement from one
group to another. but age has no bearing on the competency of respondents from
one group to another.

Multiple regression analysis indicated that statistics coursework was
among those variables that best predict competency in measurement.

Further training in educational measurement for all three groups is highly

recommended.

 

 

 

 

 

If there is an opportunity to learn, take it.
If there is no opportunity to learn, make it.

— Rashid Aldosary

 

ACKNOWLEDGMENTS

First, I would like to express my thanks and gratitude to the Bahrain
Ministry of Education for granting me a scholarship to pursue my Ph.D. program in
Measurement, Evaluation, and Research Design at Michigan State University.

Thanks and deep appreciation are due to Dr. Irvin Lehmann, my adviser
and chairperson of my doctoral committee, for his cooperation, patience.
understanding. and valuable suggestions regarding the research instrument.

I am grateful to Dr. William Mehrens for being a cooperative member in my
doctoral committee, and for his suggestions during the preparation of the test
instrument. Thanks go to Dr. Peggy Riethmiller and Dr. Antonio Nunez for their
time and understanding, and for being cooperative members in my doctoral
committee.

My sincere thanks and appreciation to Dr. Malik Balla in the Department of
Arabic at Michigan State University, Ivan Filmer, Ahmad Y. Alghamdi, Naji Alarfaj,
Humood Almuqate, and Mohammad Al-Ameeri for their careful review of the
research instrument, and for their valuable suggestions.

I am really indebted to Yousif Alzayani, former Registrar of the University
of Bahrain, and Khalid Bumtea, Coordinator of Registration at the University of

Bahrain for their great cooperation before and throughout the data collection

process.

Thanks are due to Dr. Abbas Adebee, Dean of the College of Education,
and Dr. Hussain Alsadeh, Head of the Department of Education at the University of

Bahrain, for facilitating the data collection process.

vi

 

 

Thank you to Dr. Ibrahim Yousif Alabdallah, Director of Curricula in the
Bahrain Ministry of Education, for his cooperation during the data collection
process.

Great thanks and sincere appreciation go to all curriculum specialists,
teachers, prospective teachers, and professors of education at the University of
Bahrain for their cooperation and for giving generously of their time.

Finally, my deepest and sincerest love and appreciation are due to my wife

and best friend, Noora, for her understanding, sacrifice, and relentless support.

vii

 

 

 

TABLE OF CONTENTS

Page
LIST OF TABLES ............................................................ x
CHAPTER
I. THE STATEMENT OF THE PROBLEM ................................................................. 1
Introduction .......................................................................................................... 1
Purpose of the Study ........................................................................................... 6
Importance of the Study ...................................................................................... 7
Research Questions ......................................................................................... 12
Research Hypotheses ....................................................................................... 12
Limitations ......................................................................................................... 14
Definition of Terms ............................................................................................ 15
ll. REVIEW OF RELATED LITERATURE ............................................................. 16
Introduction ........................................................................................................ 16
Integrating Testing With Teaching ..................................................................... 16
Teachers' Classroom Testing Practices ............................................................ 19
Teacher Training in Educational Measurement ................................................. 23
Teachers' Competency in Educational Measurement ....................................... 30

Educational Measurement Needs of Teachers
Prospective Teachers, Curriculum Specialists, and

Other Supervisory Personnel ........................................................................ 34
Conclusions ....................................................................................................... 42
RESEARCH DESIGN AND PROCEDURE ........................................................ 45
Introduction ........................................................................................................ 45
Research Instruments ....................................................................................... 45
Subjects and Sampling Procedures ................................................................... 47

Characteristics of the Subjects ....................................................................... 48
Data Collection .................................................................................................. 56

Follow-up of Non-Respondents ...................................................................... 57
Data Analysis ..................................................................................................... 57

viii

 

 

 

 

Summary ........................................................ 58
IV. ANALYSIS AND INTERPRETATION OF THE DATA ........................................ 59
Introduction ........................................................................................................ 59
Classroom Testing Practices and Measurement preparation ............................ 59
Performance of the Subjects on the TEST ........................................................ 68
Testing the Research Hypotheses .................................................................... 71
Multiple Regression Analysis ......................................................................... 89
Summary ........................................................................................................ 90
V. SUMMARY, DISCUSSION, CONCLUSIONS, AND
SUGGESTIONS. ............................................................................................... 93
Summary of Purposes and Procedures ............................................................. 93
Discussion and Conclusions .............................................................................. 94
Recommendation .............................................................................................. 99
Suggestions for Further Research .................................................................. 100
Reflections ........................................................................................................ 102
APPENDICES
A. THE ENGLISH VERSION OF THE TEST AND THE
QUESTIONNAIRE .............................................................................. 104
B. THE TABLE OF SPECIFICATIONS OF THE TEST ............................................ 122
C. THE TEST ANSWER KEY .................................................................................. 123
D. THE ARABIC VERSION OF THE TEST AND THE QUESTIONNAIRE ............... 124
E. CORRESPONDENCE ......................................................................................... 141
F. TABLES 4.14 TO 4.16 ....................................................................................... .154
BIBLIOGRAPHY ...................................................................................................... 157
ix

 

 

LIST OF TABLES

Table No.: Page
3.1 Information on the Sample of Teachers by Sex and

grade Level ........................................................................................................ 47
3.2 Information on the Participating Schools by Sex

and Grade Level ................................................................................................. 48
3.3 Information on the Population of Curriculum Specialists

and Prospective Teachers Used in the Study ..................................................... 48
3.4 Information on the Response Rate for the Total Sample

and for each Group ............................................................................................ 49
3.5 Distribution of the Respondents' Age in the Study ............................................... 49
3.6 Distribution of the Respondents' Sex in the Study ............................................... 49
3.7 Distribution of the Respondents by Highest Degree Held .................................... 50
3.8 Information on the Highest Degree Held Obtained from a

College of Education ...................................................................................... ‘ ...... 51
3.9 Distribution of the Respondents Whose Major was

Education in the Bachelors Program ................................................................... 51
3.10 Distribution of the Respondents by Major in College ........................................... 52
3.11 Distribution of the Respondents by Teaching/

Supervision Subject ............................................................................................ 54
3.12 Distribution of the Respondents by Grade Level .................................................. 55
3.13 Distribution of the Respondents by the Average

Number of Years in Teaching ............................................................................. 55
4.1 Distribution of the Respondents by Measurement

Coursework Completed ....................................................................................... 61
4.2 Distribution of the Respondents by Statistics Coursework

Completed .......................................................................................................... 61
4.3 Distribution of the Respondents by percentage of Time

Devoted to Classroom Testing ............................................................................ 62
4.4 Distribution of the Respondents by Purposes of

Classroom Test Use ............................................................................................ 63
4.5 Item Types Used in Classroom ( percentages) .................................................... 64
4.6 Item Types Ranked According to Preference ...................................................... 65
4.7 Distribution of the Respondents by Test Plan

Preparation .......................................................................................................... 65
4.8 Distribution of the Respondents by Item Analysis Practice .................................. 66

x

 

 

 

4.8 Distribution of the Respondents by Item Analysis Practice .................................. 66
4.9 Distribution of the Respondents' Self-Ratings Regarding Their

General Competency in Educational Measurement ............................................. 66
4.10 Distribution of the Respondents' Self—Ratings Concerning

Their Competency in Test Construction ............................................................... 67
4.11 Distribution of the Respondents' Practical Training Needs

in Educational Measurement (Percentages) ....................................................... 67
4.12 KR20 Reliability and SEM Estimates for the TEST .............................................. 68
4.13 Distribution of the Respondents' Performance on the

TEST by Content Area ........................................................................................ 70
4.14 Item Analysis Data for the TEST: Teachers ...................................................... 154
4.15 Item Analysis Data for the TEST: Curriculum Specialists ................................. 155
4.16 Item Analysis Data for the TEST: Prospective Teachers .................................. 156

4.17 Correlations Between the Scores on the TEST and
Classroom Testing Activities Variables, and Measurement

Preparation Variables ........................................................................................ 73
4.18 ANOVA for the TEST Scores With 'Purposes of Classroom Test Use“

Variable for the Teachers ................................................................................... 74
4.19 ANOVA for the TEST Scores with the 'Perceived Level of

General Competency in Measurement" for the Teachers .................................. 75
4.20a Distribution of the Respondents' Mean Performance on theTEST .................... 77
4.20b ANOVA for the TEST Scores with the 'Type of Respondent" ............................ 77
4.21a Distribution of the Respondents by Measurement Coursework

Completed ............................................................................................................ 78
4.21b ANOVA for the TEST Scores with 'Measurement Coursework" ........................ 79
4.22a Distribution of the Respondents by Statistics Coursework
Completed ............................................................................................................ 80

4.22b ANOVA for the TEST Scores with 'Statistics Coursework" ................................ 81
4.23a Distribution of the Respondents by Test Plan Preparation ................................ . 83
4.23b ANOVA for the TEST Scores with 'Test Plan" ................................................... 82
4.24a Distribution of the Respondents by Item Analysis Practice ................................ 83
4.24b ANOVA for the TEST Scores with "Item Analysis" ............................................. 84
4.25 ANOVA for the Joint Effects of Gender and Type of

Respondent with the TEST Scores ................................................................... 85
4.26a Distribution of the Respondents' Age in the Study ............................................. 86
4.26b ANOVA for the Joint Effects of Age and Type of Respondent

with the TEST Scores ................................................................................... 86
4.27 t-test for the Difference Between Male and Female Teachers

in Educational Measurement ............................................................................ 87
4.28 ANOVA for the Grade Level of the Teacher with the TEST Scores ................... 88
4.29a Information on the Sample of Teachers by Sex and Grade Level ..................... . 89
4.29b ANOVA for the Joint Effects of Teacher's Gender and Grade Level

with the Competency in Educational Measurement .......................................... 88
4.30 Multiple Regression Analysis for the TEST Scores with

Measurement Preparation, and Background Variables For the Teachers ........ 90
4.31 Summary of the Major Findings in the Study ....................................................... 92

xi

 

 

CHAPTER I
THE STATEMENT OF THE PROBLEM

Introduction

It seems reasonable to believe that the relationship between
measurement and instruction is inextricable. Rudman et al. (1980, p.2)
stated that 'there is a growing sense of need today to link assessment with
teaching and to make instructional decisions more rational and less
intuitive".

Mehrens and Lehmann (1991, p.vii) reported that "educators have
always been concerned with measuring and evaluating the progress of their
students", and that “the role of measurement is to provide decision makers
with accurate and relevant information' (p.3.). Hence, the linkage between
testing and instruction is a fundamental concern in educational practice
(Stiggins & Bridgeford, 1986), and all educators know that testing is a regular
part of the school routine (Gullickson, 1984).

Rudman (1989, p.2) indicated that

Testing and teaching are not separate entities. Teaching

has always been a process of helping others to discover

new ideas and new ways of organizing that which they

learned, whether this process took place through

systematic teaching and testing, or whether it was

through a discovery approach, testing was, and remains,

an integral part of teaching.

 

 

 

Rudman, among others, says that testing can be used to help
teachers and administrators make promotion and retention decisions, and
can be a useful tool for measuring the effectiveness of teaching and
learning.

Stiggins and Conklin (1988, p.3) illustrated the linkage between
classroom instruction and teacher's assessment of student achievement and
reported that 'the quality of classroom instruction is a function of the
teacher's assessment of student achievement'. They state that sound. daily
assessment of student learning is important for several purposes, such as
diagnosing student needs, assigning grades, and evaluating the impact of
instruction. In addition, McMorris and Boothroyd (1992) reported that 69
percent of the teachers they interviewed (n=42) stated that their main reason
for classroom testing is to assess students' mastery and understanding of
the material taught.

Research on teachers' testing practices has consistently revealed
that tests are used extensively in the classroom (Green & Stager, 1986-
1987; Stiggins, 1991), and that 'assessing, grading, and evaluating students
has been identified as one of a teacher's six core job functions' (Schafer,
1991, p.3). In addition, Stiggins & Conklin (1988), Schafer (1991), & Stiggins
(1991) found that teachers do spend from a quarter to one third of their
instructional time on assessment-related activities. Based on their research

on testing practices, Green and Stager (1986-1987) found that teachers
spend 10 to 15 percent of their time in testing, and that 40 to 50 percent of

students' course grades are based on test scores. As Mehrens and
Lehmann (1984) have indicated, ”teacher-made tests are frequently the

major basis for evaluating students' progress in school. One would have

 

 

great difficulty in conceptualizing an educational system where the child is
not exposed to teacheramade tests' (p.56). A major part of the teacher's
preparation programs in educational measurement and classroom testing
should be devoted to training them to meet the classroom assessment
demands (Stiggins & Conklin. 1988).

Newman and Stallings (1982) noted that over several decades,
educators have been aware of the need for teacher competency in
measurement, and that “much of the literature on this subject points to a lack
of emphasis on measurement training. Measurement educators almost
unanimously agree that this adversely affects classroom practices...but little
is known about teacher competency in classroom testing" (p.3).

Mayo (1964) surveyed what he referred to as "experts" (teachers.
principals and superintendents, college and university professors, and
research specialists) to judge what beginning teachers should know about
measurement. For example, he found that knowledge of general principles
of test construction, and the ability to interpret achievement test scores were
rated very important. In another study regarding students' first course in
measurement, Mayo (1970) found that measured competency in
measurement was mediocre not only among graduating seniors in teacher
training, but also among the same persons two years after graduation.
Mayo's (1967) landmark study, titled Preserv/ce Preparation of Teachers in
Educational Measurement, revealed that test scores on the measurement
competency test (MCT) were related to teaching field, and amount of testing
and measurement coursework taken. He concluded that, in general, there is
agreement on the importance of some measurement competencies for
teachers, and that beginning teachers do not demonstrate a very high level

of measurement competency. Based on the findings of his study, Mayo

 

 

recommended that some measurement coursework be mandatory, yet
meaningful, and that further research on teacher‘s competency In
measurement and testing be conducted. Green and Stager (1986-1987)
reported that there has been little change in teacher's competency in
measurement and testing since Mayo's (1967) landmark study, and that
there is a lack of competence in testing techniques in general among
teachers.

Farr and Griffin (1973, p.19) made a case in support of two
hypotheses: 'the first is that in general teachers have quite limited
knowledge of measurement concepts. The second is that teachers are not
being taught what they need to know about measurement in order to be
more effective teachers'. Newman and Stallings (1982, p. 12) also indicated
that 'measurement training practices do not seem to be any more effective
now than they were over a decade and a half ago'. Similarly, Gullickson
(1986, p.347) noted that “today's measurement and evaluation textbooks still
strongly reflect the stated emphasis' from Mayo's studies (1964, 1967).
Gullickson (1986) concludes that teachers would improve their instruction if
they were more knowledgeable about measurement and utilized better
measurement techniques. He adds that educators need to determine which
measurement tools and strategies are practical and beneficial for teachers.

Recent research on classroom assessment, teacher-made tests,
and teacher's knowledge and competency in measurement (Stiggins, 1985,
1988, 1991; Stiggins & Bridgeford, 1985; Marso, 1985; Marso & Pigge,
1987, 1989, 1991; and Carter, 1984) revealed several important findings that
have implications for research on, and teacher's preservice and in-service
training and competency in measurement and classroom testing. First,

public attention and a large and growing body of research have

     
 

in: m rum-“w“ ”
swig-s e.- .._... .. -

  

 

concentrated on standardized tests and testing issues, whereas research on
testing practices and teacher-made tests have received little emphasis or
have largely been neglected despite the fact that extensive literature exists
on classroom testing and evaluation practices. Second, the advice usually
given to preservice and in-service teachers about classroom test
development and test use relies heavily on professional judgment. Third,
the content of teacher preservice measurement and testing courses mainly
designed to facilitate the effective development and use of teacher-made
tests may not reflect the actual needs of classroom teachers. Fourth, a
thorough and comprehensive reexamination of preservice measurement
courses and in-service testing activities is needed. Finally, construction of
teacher-made tests, and specific test item writing skills need more emphasis

in in-service teacher training programs.

Supporting Mayo (1967), Newman (1981, p.2) states that 'if teacher
preparation institutions have acted upon Mayo's recommendations to
improve measurement, it would be expected that today's teachers,
especially young teachers and those with more recent degrees, would be
adequately prepared and competent in classroom testing principles and
techniques'. Newman continues that 'there is an urgent need to study and
document teacher competency in classroom testing' (p.23).

Nitko (1991) indicates that much work still needs to be done in
developing teacher's assessment competencies, and makes it clear that
'reformers' call for increased relevance of course content will be ineffective if

preservice and in- service teachers do not enter testing and measurement

classrooms' (p.2).

 

Purpose of the Study

The study was conducted for the following purposes:
1. determining how well in-service teachers, curriculum specialists, and
prospective teachers in Bahrain understand measurement principles and the
techniques of test construction, and in which areas these three groups are
most and least proficient.
2. determining how in-service teachers, curriculum specialists, and
prospective teachers in Bahrain have been prepared in measurement, as
well as in basic statistics related to measurement.
3. collecting information from in-service teachers, curriculum specialists, and
prospective teachers in Bahrain regarding the percentage of their
professional time they devote to testing, the purpose for which they use
their self-constructed tests, the item types they write for their self-
constructed tests, and the item types they rank order in terms of their
preference.
4. collecting information on whether these three distinct groups develop a
test plan (e.g., table of specifications) prior to writing test items, and conduct
an item analysis. In addition, information was collected on how well the
subjects in each group assess their level of competence in measurement
and testing in general, and test construction in particular, and their practical
needs in measurement.
5. investigating the nature of the differences among the three groups
concerning their competency in measurement, and whether these
differences are related to sex, age, grade level, amount of measurement
coursework, recency of measurement coursework (the same applies for

basic statistics), highest degree held, and recency of highest degree held.

 

 

 

6. investigating the nature of the differences between males and females,
for each group separately, regarding their competency in measurement.

7. investigating the nature of the relationships between the three groups'
competency in measurement and such variables as: amount of
measurement coursework completed, amount of statistics coursework
completed, classroom testing practices, years of experience in
teaching/supervision, grade level of teaching/supervision, and subject of

teaching/supervision.

Importance of the Study

The significance of this study is that it provides current information
on an essential, important and often less emphasized and less researched
dimension of teaching; namely, knowledge and understanding of
measurement principles. Moreover, conducting such a study in Bahrain,
where there is no empirical information on the current level of knowledge
and competency in measurement among in-service teachers, curriculum
specialists, and prospective teachers, should provide valuable information
for educators and researchers in Bahrain, and valuable comparative

information for researchers outside Bahrain.

Educators in Bahrain are in need of such a study to help them make
less subjective decisions. As Alabdallah (1990) noted, most teachers in
Bahrain are deficient in diagnosing strengths and weaknesses in student
academic achievement and learning, which led them to incorrectly employ
classroom assessment. Moreover, more than 85 percent of teachers in

Bahrain have some deficiency in constructing and correctly employing

 

.9331"-
" ‘

w.

 

classroom test items that are valid, that require different levels of difficulty,
and require functioning at higher levels, such as analysis and synthesis.
Alabdallah continues that the implementation of observation techniques and
other measurement devices is limited among teachers and is not flawless.
More than 90 percent of in-service teachers, and even the new graduates
and those with a high diploma in education (one year of full-time study
beyond the bachelors degree), according to Alabdallah, are not proficient in
using basic statistical methods that deal with test reliability and validity in
particular, and other measurement topics in general.

Based on their research findings on the teacher-made tests they
analyzed, Marso and Pigge (1991) suggest more attention and emphasis on
test construction and item writing skills in teacher preservice training and in
measurement courses. They add that teachers should be encouraged to
develop the necessary skills in test construction, and use these skills to
construct test items that function at higher cognitive levels.

Research conducted by Marso and Pigge (1987, October) on
teacher-made tests and testing has revealed that about 80 percent (n=328)
of the teachers reported that they rarely or never calculate test means or
standard deviations, and that more than 50 percent of them reported never
having estimated test reliability or conducted item analysis.

Stiggins, Bridgeford, and Conklin (1988, pp. 14-18) concluded that

...teacher training only in paper and pencil measurement

methods face real difficulties in the classroom

assessment environment... There are fundamental, far
reaching differences between the science of testing and

the assessment demands of the classroom. We have

been aware of these differences for decades and have

 

. [no

 

failed to address them... The result of poor measurement
is poor decision making. Perhaps the most compelling of
our failure to address classroom assessment needs will
be the continued alienation of teachers from systematic

assessment and evaluation processes.

Not only is it important to assess the competency of teachers and
prospective teachers in measurement, it is equally important to assess
curriculum specialists in Bahrain because there is no empirical information
concerning their competency in measurement, and because they play a
leading role in Bahrain public schools. Curriculum specialists are the link
between the Department of Curricula at the Bahrain Ministry of Education
and the public schools. Their main tasks include participating in the in-
service training of teachers, contributing to curriculum development and
evaluation, and collaborating with public school teachers to improve
teachers‘ instruction. Moreover, they conduct research and prepare
workshops on instructional methods and assessment techniques and follow-
up these studies and workshops. Finally, they supervise what is referred to
in Bahrain as a 'subject team” of public school teachers who develop
instructional methods, design appropriate assessment techniques, construct
mid- semester and final examinations, design diagnostic materials for 'slow-
learning' students, analyze school syllabi, study and analyze initiatives
submitted by teachers to develop classroom assessment and instruction,
and construct national examinations in the fall and spring semesters for the
9th and 12th grades and construct 'model' tests for other grades to guide
instruction (Alabdallah, 1989). Given these responsibilities, it is essential to

assess the competency of curriculum specialists in measurement.

 

10

Research on measurement competency and needs (e.g., Marso
and Pigge, 1987, March, 1988, November, 1989, March) has concentrated
on curriculum specialists' perceptions concerning teachers' needs in
measurement, and how curriculum specialists can assist teachers in
identifying and alleviating the most common errors in teacher- made tests.
These studies, however, have never attempted to question the competency
and investigate the needs of curriculum specialists in measurement.

Schafer highlights this issue as he states that

Administrators and other supervisory personnel in

schools also use assessment, yet measurement

education in these programs has been found to be rarer

than it is in teacher education programs. If such training is

important for teachers it is also important for principals,

curriculum specialists, etc., who provide instructional
leadership. Moreover, supervisory personnel often have
responsibilities regarding evaluative judgments about
educational programs for which a broader methodology,

such as statistical decision-making, is pertinent. It would

be useful to expand the dialogue about teacher education

in measurement to include the specific assessment skills

needed by those in supervisory roles (Schafer, 1991,

p.6).

It is hoped that the information provided by this study will
1. shed light on the actual preparation of teachers, curriculum specialists,

and prospective teachers in Bahrain in measurement, and reveal each

 

 

 

11

group's strengths and weaknesses in these areas in order to plan
measurement programs for them as needed.

2. investigate whether incompetence in measurement is a major problem
within each group independently, or the major problem among the three
groups in general.

3. reveal the differences among teachers, curriculum specialists, and
prospective teachers in Bahrain in measurement in order to help
educational policy makers in Bahrain reexamine and reevaluate the quality
of preservice and in-service training programs in measurement.

4. guide the educational policy makers and administrators at the Bahrain
Ministry of Education to give serious consideration to the training of
curriculum specialists in measurement, and to reevaluate the criteria by
which these curriculum specialists are usually selected for their prospective
positions in the curriculum department.

5. establish a vital link between the Bahrain Ministry of Education and the
University of Bahrain to design a long-range plan. based on empirical
information, for preservice and in-service training in measurement for
teachers, curriculum specialists, and prospective teachers.

6. lead to more research on educational measurement needs in Bahrain,
and in the other Arabian gulf states that are currently on their way to unify
public school curricula in the whole region.

7. open a new horizon and encourage more research on the measurement

needs of curriculum specialists.

 

 

12

Research Questions

This study has attempted to answer the following questions.
1. What is the actual preparation of Bahrain teachers, curriculum specialists,
and prospective teachers in measurement?
2. What are the current testing practices of Bahrain teachers, curriculum
specialists, and prospective teachers?
3. In what areas of educational measurement do curriculum specialists,
teachers and prospective teachers need practical training?
4. What is the current level of knowledge and understanding of Bahrain
teachers, curriculum specialists, and prospective teachers in measurement?
a. How competent is each group in measurement?
b. In which measurement area(s) is each group the most

and the least competent?

Research Hypotheses

1. For each of Bahrain teachers, curriculum specialists, and prospective
teachers there is no significant relationship between their measurement
competency and

a. highest degree held.

b. recency of highest degree held.

c. major subject of specialization.

d. major subject of teaching/ supervision.

e. grade level of teaching/ supervision.

f. years of experience in teaching.

9. years of experience in supervision.

 

 

 

13

h. amount of measurement coursework completed.

i. recency of measurement coursework completed.

j. amount of statistics coursework completed.

k. recency of statistics coursework completed.

I. percentage of professional time devoted to test construction.

m. purposes of using teacher-made classroom tests.
2. For each of Bahrain teachers, curriculum specialists, and prospective
teachers there is no significant relationship between their measurement
competency and

a. developing a test plan prior to writing test items.

b. conducting item analysis on self-constructed classroom tests.

c. self-assessment of the level of competency in measurement in

general.
d. self-assessment of the level of competency in test construction
principles.

3. There is no significant difference in measurement competency among
Bahrain teachers, curriculum specialists, and prospective teachers.
4. There is no significant difference in measurement competency (for each
group separately) between those who had measurement coursework and
those who did not.
5. There is no significant difference in measurement competency (for each
group separately) between those who had coursework in statistics and those
who did not.
6. There is no significant difference in measurement competency (for each
group separately) between those who develop a test plan and those who do

not.

 

 

I,

14

7. There is no significant difference in measurement competency (for each
group separately) between those who conduct item analysis and those who
do not.

8. There is no gender by type of respondent (e.g., a teacher) interaction
regarding their competency in educational measurement.

9. There is no type of respondent (e.g., a teacher) by respondents' age
interaction regarding their competency in educational measurement.
10. There is no significant difference between male and female teachers in
their measurement competency.

11. There is no significant difference among teachers at different grade

levels in measurement competency.

12. There is no gender by grade level interaction for the teachers

concerning their measurement competency.

Limitations

This study is limited to the populations of in-service teachers,
curriculum specialists, and prospective teachers in Bahrain. No
generalizations regarding the results of this study are expected to be made
beyond these three populations, nor are any inferences to be drawn
concerning any other educational populations related to these three groups

(such as public school principals).

 

15

Definition of Terms

The following terms are defined in the context in which they were
used throughout this study.
ASSESSMENT: "The process by which one person attempts to find out
about the knowledge, attitudes, or skills possessed by another“ (Rowntree,
1981, p.14).
COMPETENCY: 'Any single knowledge, skill,... the possession of which is
believed to be relevant to the successful practice of teaching" (Mehrens,
1987, p.197).
TEACHERS: All in-service public school teachers in Bahrain.
CURRICULUM SPECIALISTS: All in—service subject-area curriculum
specialists in the Department of Curricula; Bahrain Ministry of Education.
PROSPECTIVE TEACHERS: All preservice teachers in Bahrain who are
currently in the final year of their undergraduate program at the University of

Bahrain.

gu‘fl :N'i “cm-31d! Him

 

.' lite.” '_ 'I'. «data-am min-i .-'.I - -- " -- "

 

 

CHAPTER II
REVIEW OF RELATED LITERATURE

Introduction

This chapter is divided into five major sections. The first section
deals with the relationship between testing and teaching. and shows how
they are integrated. The second section emphasizes teachers' classroom

testing practices and highlights the research findings about these practices.
Teacher education and training in educational measurement constitutes the
central theme of the third section along with the related shortcomings in that
education and training. The fourth section discusses the past and present
research findings on teachers' competency in educational measurement.
The fifth section discusses what teachers, prospective teachers, curriculum
specialists, and supervisory personnel need in educational measurement,
either for their classroom assessment practices or for administrative
purposes, based on research findings and the suggestions of measurement

experts in the field. Some overall conclusions will follow.

Integrating Testing with Teaching
There is no doubt that teaching is a decision-making process.
Mehrens and Lehmann (1991) state that '... educational decisions are

continually being made. These decisions should be based on information

16

 

 

 

17

that is accurate. The responsibility of gathering, using, and imparting that
information belongs to educators' (p.vii). Therefore, “good schooling
depends on good decisions, and good decisions depend on good
information" (Mehrens & Lehmann, 1987, p.36). Hence, one would view
testing as systematic procedures for purposes of observing and classifying
students' performance in order to obtain instructionalIy-relevant information
for sound decision making (Nitko, 1989).

By and large, testing and teaching can be viewed as related
entities. Teaching can help us discover new ideas and organize them
through several ways. Thus, testing can be a useful tool by which teachers
can gain a general idea of what their students bring to instruction and modify
their instructional strategies if needed. Testing can also help teachers and
administrators make promotion and retention decisions, and can be a useful
tool for measuring the effectiveness of teaching and learning (Rudman,
1989). Further, Ward stated that 'to teachers, testing should be an integral
part of the instructional programs, and there should be a high match
between tests and the curriculum' (1980, p.12).

Mehrens and Lehmann (1987) reported that teachers basically use
data from their self-produced classroom tests and other sources, such as
observations, to serve the main functions of teaching, such as setting
realistic goals for students. Nevertheless, it is really surprising to know that
most teachers separate testing from teaching“ (Rudman, 1987a, p.6).
Moreover, classroom assessment can encourage good study habits by
students' frequent review of materials, and can provide students with

feedback regarding their strengths and weaknesses (Mehrens & Lehmann,

1987).

c-r-Me-tslv h'; fifilV‘h-r'fi, a. s;- -- . . .-

         

 

 

 

18

Educational measurement and testing can play several other
leading roles in instruction, such as selecting learning experiences and
procedures of instruction, organizing learning experiences, and aiding in the
supervision and administration of instruction. They are valuable for the
continuing development of teachers and coordination of instructional efforts.
Educational measurement is conceived, then, as an integral part of
instruction, not as separate from instruction (Tyler, 1951), because it is
impossible to imagine how teachers can diagnose students, follow up
student Ieaming, and communicate with parents without daily classroom
assessment of student achievement (Stiggins & Conklin, 1988).

Educators should carefully and consciously attempt not only to help
teachers make decisions but also help teachers understand how to select
the relevant and accurate information that they need, along with their
intuition and judgment, to make correct decisions. Teachers should be
knowledgeable about the quality of information they gather and use, and the
criteria by which they judge that information. Underlying all of this discussion
is the fact that most of the information that teachers use for making
decisions about student learning and achievement comes from a variety of
sources. This can include informal or systematic observations of students,
teacher-made tests, standardized tests, students' oral presentations, and
completed projects. Hence, teachers and administrators must understand
the important potential role educational measurement and evaluation play in
integrating testing with teaching. Adopting this perspective should lead us to
understand that our knowledge of the relationship between assessment and
instruction will alleviate the persistent notion that testing takes time away

from instruction (Rudman, 1987a).

,‘I"§".-.‘-'.""'¢:|; ‘ ignilfituki-j

I

F , 1‘" altfl‘nmmmw- ..elm'flw

1

 

£31533 I‘ _ III-II; - . n: "..'l.'"_'4" . . . __ .-

 

19

Finally, we believe, as do Mehrens and Lehmann, that

...unless there is administrative support for classroom
assessment, until administrators and teachers come to
appreciate the role and value of evaluation in the
teaching-learning process, and until through our teacher-
training institutions or through high quality professional
development programs we prepare both teachers and
administrators to develop and interpret the results of
classroom assessment, our pupils will not be receiving

maximal instruction (Mehrens & Lehmann, 1987, p.42).

Teachers' Classroom Testing Practices

As noted earlier, a number of researchers who have examined
teachers' classroom testing practices have documented the fact that
teachers spend between ten and thirty-three percent of their instructional
time on testing and assessment - related activities (Stiggins & Conklin,
1988; Stiggins, 1991; Schafer, 1991; Green & Stager, 1986-1987; &
Gullickson, 1982). In addition, Green and Stager (1986-1987) found that 40
percent to 50 percent of students' course grades are based on test scores.
Further, many teachers place more weight on their own tests than on
anything else when assigning students' final grades (Boothroyd, 1990).

Fleming and Chambers (1983) reviewed 342 twelfth-grade teacher-
made tests and found that, by and large, teachers tend to omit test
directions, use illegible test copies, and omit the point values assigned each
test item. Fleming and Chambers concluded that their findings suggest that

teachers may not view their own tests 'as a means for quantifying students'

      
 

Re. I'm-nearem

  

”s: nun-4 turn aw _,.-- '- ' .

 

20

performance" (p.36). This appears to confirm previous studies according to
Fleming and Chambers.

Harris (1973) surveyed 145 NCME members to determine the
extent of their agreement on 40 controversial issues in educational
measurement. Regarding those issues pertinent to classroom teachers'
testing practices, Harris found that sixty-nine percent of the members
surveyed agreed that the majority of teachers rely heavily on their subjective
judgment in evaluating and grading their students, and that seventy-six
percent of those surveyed agreed that most classroom teachers use very
few statistical techniques in evaluating and grading their students.

Marso and Pigge (1989, 1991) analyzed a sample (n=175) of
teacher-made tests to determine item types frequently used, item cognitive
functioning levels, and the frequency and nature of test construction errors.
They found that the most frequently used item types are matching, multiple-
choice, and completion. Furthermore, they found that teacher-made test
items primarily function at the knowledge level of the cognitive domain
(according to Bloom's taxonomy), and that the matching test items were 'the
most construction error prone followed by completion, essay, and true false'
(Marso & Pigge, 1991, p.283). In a related study on classroom teachers'
testing practices, testing proficiencies, and testing resources, Pigge and
Marso (1988) surveyed 586 school supervisors and principals, and 326
school teachers regarding how supervisors can help teachers identify and
alleviate the most common test construction errors in teacher-made tests.
They reported that the 'average' teacher gives 54.1 formal tests during the
school year, and that the 'average' teacher-made test contains 37.9 items.
In addition, 50 percent of the teachers said they construct 75 percent or

more of their classroom test items. Pigge and Marso also found that 72

 

 

 

21

percent of teacher- made test items function at the knowledge level, 11
percent at the comprehension level, 15 percent at the application level, and
about 1 percent measure higher- order thinking skills (e.g., analysis,
synthesis, and evaluation). These findings are similar to those results
reported by Gullickson (1982).

It is worth noting that teachers tend to prefer certain types of items
for their self-produced classroom tests. Newman and Stallings (1982) found
that teachers' rankings of item types from most to least in terms of
preference were '(1) completion, (2) multiple-choice, (3) matching, (4) true-
false, (5) short answer, (6) calculations, and (7) essay" (p.11 ). Newman
and Stallings also noted that these rankings differed from those obtained by
Reynolds and Menard (1980) although the teachers' rank order of multiple-
choice items (ranked 2) did not change.

One will recognize that classroom testing practices and the contents
of teacher -made tests in other countries such as Japan may not be much
different from those in the United States. Yamagishi (1991) reports that,
based on a survey of schools in 10 regions in Japan, Japanese teachers
mostly favor those tests that are designed to accompany classroom
textbooks. Japanese teachers' main reason for using these 'text' tests,
according to Yamagishi, is due to the lack of time available to construct their
own classroom tests. Detailed analysis of these tests revealed that they
were poorly written, emphasized rote memory, and tested fragmented
knowledge. Yamagishi found that short answer and essay items are rarely
found in the Japanese teacher-made tests in language arts. In social
studies, she found that 80 percent of the items were of the recall or
recognition type, and 90 percent of them dealt with facts. In mathematics,

Yamagishi reports that the majority of the items measured computation and

 

 

22

understanding, whereas 5.1 percent were at the application level, and 3.3
percent functioned at the analysis level.

It seems that recent research on classroom testing practices
verifies the research findings of Gullickson (1982), who noted that short
answer and matching item types are the most popular among teachers and
both types are restricted to lower cognitive level (e.g., knowledge), despite
the fact that this contradicts what Yamagishi has reported about teacher-
made tests in Japan with respect to the use of short answer items in
language arts. Gullickson's results suggest that neither the cognitive level of
the items nor item difficulty were adequately addressed. If teacher-made
tests are targeted toward the lower cognitive level, and students' grades are
based on their achievement at this level, students' motivation will be focused
on lower cognitive level learning. Gullickson concluded that there are three
important factors that influence classroom testing practices: expertise, time,
and tools available for the teachers' use.

Further examination and analysis of the results of the teachers'
classroom testing practices suggest that there are two main factors that
contribute to the frequency of use of teacher-made tests: subject matter and
grade level. Boothroyd (1990), Stiggins and Bridgeford (1985), and Green
and Stager (1986-1987) reported that the use of teacher-made tests varies
from subject to subject. For example, Stiggins and Bridgeford (1985) noted
that teachers are more likely to use teacher-made tests for assessing
student competency in mathematics (86 percent) or science (88 percent) as
compared with speaking (71 percent) and writing (74 percent). Stiggins and
Bridgeford state that, in terms of grade level, the number of teacher-made
tests increases with grade level, and that a positive relationship has been

detected between grade level and the proportion of teachers who construct

 

 

 

23

their own classroom tests. They also report that 68 percent of the second
grade teachers surveyed said they develop their own classroom tests
compared to 91 percent of the eleventh-grade teachers. Furthermore, Marso
and Pigge's (1988) survey of teachers (n=326) revealed that secondary

school teachers test more frequently than elementary school teachers.

Teacher Training in Educational Measurement

“Historically, teachers have not been required to complete
coursework in educational measurement to attain state teacher certification"
(Boothroyd, 1990, p.19). Noll (1955) reported that only five states required
completion of one educational measurement course for obtaining state
teacher certification. Based on his survey of 80 teacher-education
institutions, Noll found that 82.5 percent of them offered courses in
educational measurement, but only 21.2 percent of these institutions that
offer a measurement course required that course for undergraduates, and
only 13.6 percent of them required it for all undergraduates preparing to be

teachers.

In a study regarding the experience of teachers with tests and
formal training in measurement, Goslin (1967) found that more than 20
percent of secondary school teachers surveyed (n=301) had never been
exposed to a course in tests and measurement. He also found that 39.2
percent of secondary school teachers in his sample have low familiarity with
tests. Goslin further reported that more than 40 percent (N=1440) of all
teachers surveyed had only one course in measurement. Goslin concluded
that his study indicated that 'teacher involvement in testing...is of potentially

great influence on the educational system and the children in it' (p.129).

 

 

 

24

In his landmark study, Mayo (1967), surveyed 2,877 seniors in 86
randomly selected teacher-training institutions concerning their training in
educational measurement. Mayo found that only 30 percent of the student
teachers surveyed reported having taken a course in educational
measurement. He also concluded that some teachers, especially elementary
teachers, have "a strong bias against statistics, apparently because they see

no relation to their work" (p.56).

Based on his nationwide survey of 896 accredited institutions
offering teacher education programs, Roeder (1972, 1973) indicated that
57.7 percent of the institutions surveyed 'did not require their prospective
elementary teachers to complete a course in evaluation" (p.239). Roeder

comments that
when one considers the increasing importance which
report cards marks and test scores play in the lives of our
children, it is readily apparent that the failure of teacher
education institutions to provide prospective teachers with
at least a minimal understanding of evaluation is

inexcusable' (p.143).

Ward (1980. 1982) indicated that while many teachers deal daily
with testing- related issues and activities, the vast majority of them have had
no formal training in educational measurement. The results of Ward's survey
on teachers' attitudes toward and knowledge of testing (n=209) revealed that
29 percent of the surveyed teachers indicated they had never taken any
measurement courses; 29 percent had taken one; and 42 percent of them
had taken two. Moreover, 70 percent of the teachers reported that they had

never taken any in-service training in educational measurement.

 

 

25

This lack of training or adequate training in educational
measurement seems to have an adverse impact on instructional practices
and student assessment in the classroom. In this regard, Green (1963)

states that
yet it is not unusual to find a teacher who has had no
specific training in measurement and evaluation and who
has evolved his testing procedures through trial and error.
Such a teacher often designs tests which measure
objectives other than those which have been emphasized
in his classroom. Because his examinations emphasize
trivial points which had been either neglected or given
cursory considerations, he has little notion of the
effectiveness of his instructional objectives. For this
teacher, tests may become a series of puzzles designed
to fool the student instead of instruments for measuring

the relative attainment of classroom objectives (pp. 1 -2).

Schafer and Lissitz's (1987) review of literature concerning
measurement training for school personnel appears to be comparable with
previously discussed research findings. They noted that 'a significant
proportion of school personnel do not receive much training in measurement
methods' (p.61). They conclude that teachers' performance on tests of
measurement principles, and analysis of teacher-made tests, revealed that

teachers lack training to utilize measurement practices.

Recent research on measurement~related issues (Stiggins,
Griswold, & Frisbie, 1986; Stiggins, Frisbie & Griswold, 1989) has shown

that there is a lack of teacher training in grading. Stiggins et al. studied 15

 

26

high school teachers, utilizing what they refer to as "intensive case study
methodology". The researchers prepared and used a comprehensive
framework of 34 grading issues to serve as a basis for observing teachers'
grading practices. The two major issues that the researchers explored were:
(1) the nature and technical quality of assessment and grading practices,
and (2) why professional training has had little impact. Discrepancies
between actual practices and 'best' practices in grading were noted in 76
percent of the 34 grading issues. A considerable number of these
discrepancies indicated that 'a lack of teacher training in grading is a
possible cause" (Stiggins, Frisbie, & Griswold, 1989). They concluded that
in order to understand the problems that seem to characterize grading
practices, the first step is that 'we begin with much needed research and
translate the results into much needed training' (p.14).

In a related study to determine teacher training priorities, and to
compare those findings with results of a decade-long study, Stiggins and
Conklin (1988) indicated that discrepancies between the actual assessment
needs of teachers and the needs addressed or not addressed by teacher
training programs suggest that there are three dimensions that need to be
reexamined: policies, certification requirements, and teacher training
curricula. According to Hills (1991), having a teacher or an administrator in
school who has at least basic training in educational measurement, can
make a difference, because 'one of the most serious problems of evaluation
is the fact that a primary means of assessment, the test itself, is often
severely flawed or misused' (Hills, 1991, p.541). What further complicates
the problem is that teachers' judgment—based assessment lacks clear and
explicit criteria due to lack of training, which leads to 'vague criteria

communicated to students... and poor performance-rating procedures

27

contribute undependable assessment and inappropriate instructional
decisions” (Stiggins, 1988, p.364).

More recently, Stiggins (1991) studied the degree of mismatch
between teachers' assessment training needs and the content of actual
training. Stiggins analyzed the assessment training curriculum in 27
undergraduate and graduate programs that provide 75 percent of all the
teacher training. He found that only 13 (48 percent) of these programs offer
any assessment training at all, and only 6 of these 13 stipulate teacher
participation in that training as a requirement for graduation. Stiggins
continues that 'I believe... that over the years—indeed, over the decades—
our training has been painfully and chronically misfocused' (p.8).

If the research on teacher training in educational measurement is
important, it is equally imperative to investigate the relevance of
measurement instruction to teachers' needs. Some of the emerging research
on this issue (Airasian, 1991; Gullickson & Hopkins, 1987) highlights some of
the limitations of measurement instruction and training for teachers. Airasian
(1991, p. 13) states that 'the content of most measurement textbooks and
courses shares little relevance with the day to day life of teachers in
classrooms. Such a disparity might have been acceptable twenty years ago
but it is not today‘. To support his thesis Airasian examined the current
status of educational measurement instruction through the topics covered in
educational measurement courses and textbooks and arrived at three main
conclusions:

(1) the topics more commonly covered are, for example, the role of
educational measurement, educational objectives, item writing, validity,

reliability, and checklists and rating scales;

28

(2) course and textbook discussion of these topics concentrates
on formal types of classroom measurement, and

(3) few realistic examples of the classroom use of measurement
techniques characterize the vast majority of textbooks. The concepts are
presented sometimes in an abstract manner, which makes it unclear how to
make these few real-life examples operate in the classroom environment.

Gullickson and Hopkins (1987) surveyed professors from 33
colleges in South Dakota regarding the nature of educational measurement
instruction provided students in preservice teacher education programs.
Their results may serve as a basis for suggestions concerning the limitations
of educational measurement instruction offered to teachers. Their findings
indicated that 50 percent of students take a measurement and evaluation
course, and the other fifty percent receive it as part of another course.
Although these measurement courses are typically taught by professors (81
percent of them hold doctoral degrees), most of them, according to
Gullickson and Hopkins, have not majored in educational measurement.
These results reinforce 'the need for specific understanding of measurement
as it applies to classroom practices' (Gullickson & Hopkins, 1987, p.15).
Boothroyd, McMorris, and Pruzek (1992, April) found tha 'permanently
certified teachers were more likely to have completed measurement courses
than were provisionally certified teachers', and that 'most of those who had
completed measurement courses estimated that only a small proportion of
the course content directly assisted them in test construction" (p.4).

At the University of Bahrain, all undergraduate students in the
college of education are required to take a course in educational
measurement and evaluation. In addition, high diploma (one year of full-time

study beyond the bachelors degree) and masters degree students in

 

29

education are required to take a higher level course in educational
measurement and evaluation (University of Bahrain, 1991 -1992).

Obviously, the problem of adequate and relevant preparation and
training for teachers in educational measurement is not confined to the
United States, but extends to other countries. As a case in point, while
extensive use of tests heavily dominates and characterizes Japanese
schools, research on testing in Japan seems to be scant (Yamagishi, 1991).
Yamagishi highlights the problem as she states that

Japanese teachers, in general, are not well trained and

prepared for tests and evaluation in classrooms, since a

specialized course in measurement and evaluation is not

required of students enrolled in teacher training

programs. Beginning teachers‘ knowledge of testing is

limited to what is provided by a brief review section of a

general educational psychology course (p. 170).

Results from a survey conducted by the Japan Educational
Psychology Association revealed, according to Yamagishi, that only 28.3%
(n was not reported) of the surveyed teachers reported that their professor
covered testing and evaluation in classroom. Further, Yamagishi continues
that 'theoretical and technical issues relevant to classroom evaluation are
generally not covered. This suggests that Japanese teachers are generally

poorly prepared in testing and evaluation' (p.174).

 

30

Teachers' Competency in Educational Measurement

Ebel (1961, p.67) stated that

It is of the utmost importance to educational progress that
the competence of teachers to measure educational
attainment be improved... I am fully persuaded that the
current problem in testing which most urgently requires
the attention of all professional educators is that of
improving the tests made and used by the classroom
teacher. To establish the importance of improving teacher
competence in measurement, it is necessary to show not
only that measurement of educational attainment is
needed but also that teachers are currently deficient in

getting this job done.

It seems that Ebel‘s words are still relevant since recent research (e.g.,
Stiggins, 1991) calls for similar ideas and raises similar concerns.

Generally speaking, research on teachers‘ competency in
educational measurement has taken two dimensions. The first depicts those
studies that consist of indirect assessment methods represented by
surveying teachers and having them provide self-report ratings of their
competency in educational measurement. The second involves more direct
assessment of teachers' competency in educational measurement.

Ebel (1961) noted seven errors teachers frequently make in
measuring educational achievement. First, teachers rely heavily on their own
subjective standards in judging educational achievement although few

teachers collaborate in test construction. Second, teachers tend to leave

 

 

31

classroom test preparation until the last minute. Third, teacher—made tests
are too short, poorly planned, and lack adequate sampling of the content
domain. A fourth error is that teachers tend to place too much emphasis on
unimportant details in their classroom tests, while neglecting fundamental
principles. Ambiguous teacher- made test items, and rarely having these
tests reviewed by a competent colleague is the fifth error Ebel found in
teacher-made tests. Sixth, many teachers usually underestimate the effects
of the sampling errors on test scores, thereby assuming that the individual
student's test score is absolutely accurate. Finally, Ebel reported that
teachers, in assessing effectiveness of their own tests, seldom compute
statistics on their classroom tests, such as the mean, standard deviation, or
the reliability of test scores. Based on the problems he identified in teachers'
construction of their classroom tests, Ebel (1962) developed a list of ten
measurement principles as a guide for teachers in their classroom testing
practices. It seems that the most important among these principles is the
one that says: 'to measure achievement effectively the classroom teacher
must be (a) a master of the knowledge or skills to be tested, and (b) a
master of the practical arts of testing' (p.22).

Mayo (1970, p.2) found that 'measured competency in
measurement was mediocre among not only graduating seniors in teacher
training but among the same persons two years after graduation“. One
would speculate, according to Mayo, that this low performance was probably
related to a 'Iack of commitment to some related factors like mathematics
and statistics. He concluded that'... professors should search for teaching
innovations in measurement courses' (p.2).

Among the first comprehensive studies designed to assess

teachers' competency in educational measurement was Mayo's (1967)

 

32

study on preservice preparation of teachers in educational measurement.
Mayo developed and administered a measurement competency test to a
sample of 2,877 senior education students at 86 institutions. The examinees'
average scores (M=28.61, SD=7.284, form A; M=24.97, SD=6.226, form B,
Total score=50) led Mayo to conclude that beginning teachers in general do
not possess or demonstrate a high degree of knowledge and skills in
educational measurement. Mayo concluded that 'two possible obstacles
impeding improvement of the measurement competency of teachers may be
(1) the lack of deep commitment to problems and practices in evaluation,
and (2) negative attitude toward statistics' (p.63).

In a related study, Noll (1955) asked 77 college seniors and 108
experienced teachers some questions on fundamental concepts and
procedures in measurement. His results showed serious lack of
understanding of the basic concepts in measurement among those

surveyed.

Newman (1981) found that the teachers surveyed (n=280)
averaged 53.7 percent correct responses on her measurement test. She
noted that teachers with high scores on the measurement test tended to
have better understanding on the affective domain portion than low-scoring
teachers. She also indicated that high-scoring teachers appeared to better
understand statistical concepts than low-scoring teachers. Newman found
that, in general, the teachers surveyed, and low-scoring teachers in
particular, view statistical concepts as unimportant. I

Ward (1980, 1982), in surveying classroom teachers' knowledge of
testing identified three areas in testing in which teachers felt knowledgeable,
namely; interpreting test score results, using tests for instructional planning,

and judging the relevance of tests.

33

Using factors such as teachers' knowledge of tests and testing, and
the nature and quality of information yielded by tests, Gullickson (1984)
surveyed 450 teachers representing different grade levels. He indicated that
although teachers perceive themselves as having sufficient knowledge of
testing, they are not necessarily knowledgeable in test construction.
Gullickson concluded that teachers have limited knowledge in educational
measurement. These findings appear to be comparable with other research
regarding lack of competence in educational measurement among
classroom teachers (Green & Stager, 1986-1987; Newman, 1981; Newman
& Stallings, 1982), which suggests that teachers' competency in educational
measurement has not changed since Mayo‘s (1967) study.

It seems reasonable to believe that many preservice and in-service
teachers depend on a 'repertoire of limited and uninformed test construction
skills when they create assessment items' (Carter, 1984, p.57). Carter
(1984) interviewed 310 teachers focusing on: (a) teachers' perceptions of
their item-writing ability, and (b) teachers' practices in developing items for
classroom tests. The study showed that teachers appeared unaware of
various item-writing principles presented to them, that their classroom tests
were loaded with item-writing problems, and that these principles were not
covered as part of their preservice training in educational measurement. The
teachers reported that they typically paraphrased textbook materials, and
rarely revised their test items. Moreover, Carter noted that teachers
incorrectly matched 20 percent of the items to their respective objectives.
He also found that the mismatch between items and objectives increased as
the level of the objective increased.

More recent research on teachers' needs and proficiencies in

testing and measurement (Marso & Pigge, 1987, 1989, 1991; Pigge &

34

Marso, 1988) has shown some results that may confirm previous research
on teachers' competency in educational measurement. For example, Marso
and Pigge (1989) surveyed teachers, principals, and supervisors concerning
classroom teachers' testing proficiencies. On average, they found no
significant difference among teachers in their testing proficiencies as
perceived by teachers themselves with regard to grade level and years of
teacher experience. Marso and Pigge concluded that more attention should
be paid to test item-writing skills that function at cognitive levels beyond
the simple knowledge levels in teacher training programs. It can be seen
from these results that teachers' proficiencies in test construction are
inadequate to meet their needs, and that although principals and
supervisors agreed on teachers' evaluation competency needs, they
disagreed on teachers' proficiency in educational measurement (Marso &
Pigge, 1987). Further, a number of errors were found in item format and
construction (Pigge & Marso, 1988), as well as guideline violations which
suggests that teachers' test construction skills are weak, inadequate and
need improvement to function at higher cognitive levels (Marso & Pigge,

1991).

Educational Measurement Needs of Teachers, Prospective Teachers,
Curriculum Specialists, and Other Supervisory Personnel

Both early and recent research have used a variety of approaches
in an attempt to identify the essential competencies that teachers,
prospective teachers, curriculum specialists and other supervisory personnel
need in educational measurement.

Mayo (1964) developed what he called 'an ideal list' of 70

competencies in educational measurement. He then surveyed a sample of

 

 

35

“experts“ (teachers, principals, superintendents, college and university
professors, and testing and research specialists) to see what they thought
beginning teachers should know about educational measurement (by rating
these competencies in terms of their importance for teachers). He found
that, in general, the vast majority of these competencies were considered as
important by these 'experts'. Among the competencies rated as highly
Important for beginning teachers were knowledge of general principles of
test construction, knowledge of the advantages and disadvantages of
teacher-made tests, knowledge of effective marking procedures, ability to
interpret achievement test scores, knowledge of test administration
techniques, ability to state measurable educational objectives, and ability to
interpret diagnostic test results to evaluate student progress. It is surprising
to know that some essential topics for teachers, such as reliability, validity,
item analysis, and statistics, were rated as moderately important by the
“experts” surveyed.

In a related but more recent study on teacher-perceived needs in
educational measurement and evaluation, Gullickson (1986) noted that
teachers usually do not use statistical analysis of tests. Despite the fact that
preservice measurement instruction has a relatively substantial emphasis on
statistics, according to Gullickson, teachers may devalue its importance
because it does not result in a level of understanding that would help them
apply statistics in their evaluations. This, however, may be due to lack of a
good grasp of statistical concepts (Rudman, Kelly, Wanous, Mehrens, Clark,
& Porter, 1980).

Ebel (1962) suggested several competencies in educational
measurement that teachers should know, such as knowledge and limitations

of tests, the criteria to judge test quality and how to get evidence related to

 

 

36

these criteria, test planning and item writing, test administration procedures,
and test score interpretation.

Goehring (1973) reviewed 66 textbooks in educational
measurement to identify what competencies are focused in undergraduate
educational measurement courses. In addition, he conducted a task analysis
of the work of classroom teachers with respect to these competencies.
Goehring found that about 67 percent of the competencies identified
constitute a major portion of the evaluation competencies that are essential
for in-service teachers, such as test construction, application and
interpretation of tests, and knowledge of statistical procedures. Goehring
concluded that, according to the texts surveyed. measurement courses
should concentrate on teacher competencies related to test construction,
application, and interpretation of classroom tests as well as on those
competencies that have direct effect on pupils and parents.

Stiggins and Conklin (1992) stated that in order to better serve
teachers in classroom assessment training, we need to develop a special
course tailored to the needs and unique demands of classroom assessment.
This course, according to Stiggins and Conklin, will serve teacher training
needs by providing teachers with skills and assessment tools they need for
their daily classroom assessment demands. Stiggins and Conklin continue
that we must plan to provide assessment training and technical assistance to
those teachers who completed undergraduate and graduate programs that
lack assessment training.

Teachers' knowledge of some core issues in educational
measurement, such as validity and reliability, needs to be established. It is
important for teachers, for example, to understand how to gather evidence

for validity, and that validity is a matter of degree. Teachers should also

 

 

 

37

know about the role and importance of the reliability of test scores (Merwin,
1989). These issues, however, will be incomplete without teachers'
knowledge and skills in elementary statistical methods, and teachers' ability
to use 'educational assessment to solve instructional problems” (Nitko,
1991, p.2). To be effective, teachers may need to be exposed to a two-year
educational measurement program beyond their undergraduate program,
along with laboratory work and experience to develop and practice the
necessary skills (Schafer & Lissitz, 1991; Nitko, 1991).

In planning a relevant and practical educational measurement
training program for teachers, some interrelated factors need to be
addressed within the context of training and its linkage to schools. As
Stiggins (1988) has suggested, such a training program must include
priorities for teachers like 'higher order reasoning skills, writing quality paper
and pencil test items, integrating assessment and instruction in oral
questioning strategies, and designing quality performance assessment
based on the observation and professional judgment' (p.15). But to facilitate
this training for teachers greater knowledge on the part of researchers and
teacher training institutions of the nature of the classroom assessment
environment and its demands is needed (Stiggins & Bridgeford, 1985).
Stiggins (1991) describes the way he approaches teachers' needs in
educational measurement in his 10-session training course. Stiggins
emphasizes the following: (1) making teachers aware of the meaning and
quality of assessment and its importance for students, (2) the importance of
designing assessment with clear achievement targets, (3) design and use of
paper and pencil assessment tools, (4) assessment of reading proficiency,
(5) using observation and professional judgment as classroom assessment

tools, (6) writing assessment and its relation to observation and judgment of

38

achievement—related behaviors, (7) how to develop sound grading practices,
(8) norm—referenced standardized achievement tests, and (9) using
computer technology in classroom assessment. In addition, Stiggins focuses
on related sound assessment factors, such as having a clear target that fits
the purposes of classroom assessment. Evaluation of the impacts of his
training course in educational measurement on teachers' classroom
assessment proficiency and students' learning showed encouraging results.
First, he found improvement in student achievement, which manifested itself
in “a higher level of student academic success in writing, speaking, higher
order thinking and problem solving" (p.11). Second, teachers appeared
better able to help students get a clear and more complete understanding of
the academic achievement target which, according to Stiggins, led to better
instruction, and increased student success rate. Third, there was an
increase in students' enthusiasm, and a reduction in assessment anxiety
“due to clear expectations, more focus preparation for tests. and improved
communication with parents and others about achievement targets" (p. 11).
Another element that must be considered in educational
measurement programs is training administrators in classroom assessment
(Stiggins, 1986, 1988). For example, principals should master the knowledge
and skills for high quality classroom assessment. Stiggins feels that
principals and other administrators need this measurement expertise to help
teachers with their classroom assessment and daily measurement of student
growth. Hence, “it would be useful to expand the dialogue about teacher
education in measurement to include the specific assessment skills needed
by those in supervisory roles" (Schafer, 1991, p.6). such as principals and
curriculum specialists who offer instructional leadership in schools. Thus,

teachers' knowledge of educational measurement should be enhanced by

 

39

those who communicate with teachers and influence their instructional
environment.

Our attitude toward relevant, integrated, and quality measurement
programs for teachers should not preclude us from emphasizing the fact that
'the adequate measurement of achievement requires not only skills and
scholarship pertaining to test theory and interpretation but also a
sophisticated grasp of the academic content to be measured and an intimate
knowledge of appropriate classroom activities" (Rudman, 1987b, p. 10).

Most educators who are responsible for teacher training and
education 'function in a world that is formal, precise, well articulated,
prescriptive, technical, content oriented, and conceptually dominated'
(Airasian, 1991, p. 14). Teachers, however, deal with real problems in their
daily classroom instruction and assessment. Thus, it is insufficient to identify
teachers' needs in educational measurement, but also, as Airasian reports,
'we need workshops that are intended to identify what measurement
specialists need to know about life in classrooms' (p.14).

In addition to the relevant topics in educational measurement
previously discussed, Airasian (1990,1991) calls for other measurement
issues and activities that he deems important for teachers, and that need
more attention in texts and instruction. For instance, Airasian states that
when, at the beginning of the school-year, teachers get to know their
students with different abilities, interests, and school experiences, they need
to gather some information based on their observation, integrate this
information with their expectations and attitudes, and make judgments about
the characteristics of individual students. Teachers, then, arrive at what
Airasian calls “sizing up" judgments. These judgments need informal

observation. Therefore measurement textbooks should address this kind of

40

observation methods, and should show teachers how the information they
collect based on this kind of observation can be incomplete, biased,
unstable, suffer from logical errors, and not valid for making appropriate
inferences about student performance. Moreover, there are three more
measurement activities that Airasian suggests that should be addressed in

measurement courses. They are:

(1) Much time should be spent on how teachers can use measurement
information to plan instruction and critique instructional materials;

(2) Informal performance assessment should be done regarding how
professors' instruction in measurement is going and how teachers feel
toward course and professor; and

(3) The packages that accompany school textbooks, such as worksheets.

home activities, and achievement tests should be discussed in measurement

COU I’SSS .

“Measurement books should...provide information to help teachers make
decisions about the suitability, appropriateness, and validity of such
materials for their own instruction" (Airasian, 1991, p.15), because these
activities and materials consume more of a teacher's time and may have a
more profound impact on learning and instruction than that of the formal
measurement procedures emphasized in measurement textbooks. Hence,
classroom assessment courses should address these assessment
procedures and show the appropriate tools to be used to carry out such
classroom assessment tasks. Airasian draws our attention to some other
relevant topics which should be addressed in measurement courses for
teachers, such as how to use information to plan instruction, how to assess

learning during instruction, and how to evaluate instructional materials.

41

The development of the Standards for Teacher Competence in
Educational Assessment of Students by the American Federation of
Teachers (AFT), the National Council on Measurement in Education
(NCME), and the National Education Association (NEA) (1991) has
essentially stemmed from the growing concern about the inadequacy of
teacher preparation in assessment.

These standards (pp.30-32) are as follows:

1. Teachers should be skilled in choosing assessment methods appropriate
for instructional decisions.

2. Teachers should be skilled in developing assessment methods
appropriate for instructional decisions.

3. Teachers should be skilled in administering, scoring and interpreting the
results of both externally-produced and teacher-produced assessment
methods.

4. Teachers should be skilled in using assessment results when making
decisions about individual students, planning teaching, developing
curriculum, and school improvement.

5. Teachers should be skilled in developing valid pupil grading procedures
which use pupil assessment.

6. Teachers should be skilled in communicating assessment results to
students, parents, other lay audiences, and other educators.

7. Teachers should be skilled in recognizing unethical, illegal, and otherwise
inappropriate assessment methods and uses of assessment information.
These standards seem important, relevant, and integrated with teachers'
needs in classroom assessment. I think that these standards, when worded

in their final form based on practice and experience, should be addressed in

42

measurement courses and communicated to teachers both theoretically and

practically.

Conclusions

The following conclusions are drawn from the review of literature
discussed above:
1. There is an inextricable relationship between testing and teaching, and
this relationship should be emphasized in educational measurement courses
in teacher-training programs.
2. Teachers spend up to one third of their professional time on testing-
related classroom activities.
3. There are many errors in teacher-made tests.
4. There is a discrepancy between teachers' perceived proficiency and their
actual competency in educational measurement.
5. Teachers' competency in educational measurement has not changed
since Mayo‘s (1967) study.
6. Educational measurement textbooks do not explicitly emphasize how
teachers can integrate testing with instruction.
7. Most research has concentrated on teachers' needs as they perceive
them or as they are perceived by principals and supervisors in educational
measurement, rather than from directly measuring teachers' competency in
educational measurement.
8. The available review of literature on teachers' competency in educational
measurement does not appear to have addressed curriculum specialists'
needs in this important area.
9. Recent research calls for preparation and assessment of principals and

curriculum specialists in educational measurement.

43

10. Recent research calls for an urgent need to make those responsible for
teacher education programs more aware and more knowledgeable about the
classroom environment in order to familiarize themselves with the daily
classroom problems, and to help them plan their programs on a reality-check
basis.

11. Recent research suggests that some educational measurement courses
should emphasize informal classroom assessment.

12. Practical educational measurement training for prospective teachers is
infrequent. This also applies to those in supervisory positions, such as
curriculum specialists.

Descriptive statistics that relate directly to classroom teachers' daily work
should be emphasized. Hence, relevance of measurement and statistical
topics should be important in order to gain teachers' attention. Teachers'
avoidance of measurement courses may be attributed not only to teachers'
'dislike' of statistics or to difficulty of measurement courses, but also
because teachers deem these courses irrelevant to their daily instruction
and assessment of their students (Farr & Griffin, 1973). Teachers' poor
comprehension of statistics (Rudman et al., 1980) may be another factor that
leads teachers to dislike statistics and make them consider statistics
irrelevant to their daily instruction and assessment. Finally, measurement
courses should be more practical. Teachers should receive training on how
to apply classroom assessment and use it as a vehicle for making sound
decisions. Further, teachers need to know how to use and integrate
cognitive theories of learning with classroom assessment. This will be a
remote target to achieve until measurement courses address it both
theoretically and practically. Practical workshops in educational

measurement for prospective teachers appear to be scant. If we believe, for

 

 

44

example, that reliability, validity, and item analysis are essential for teachers,
then we should help them practice and apply these principles. Moreover, we
should show them how to interpret the results of applying these principles in
different learning situations, for different purposes, and to make relevant,
accurate, correct, and desirable inferences that help teachers make sound

decisions about learning and instruction.

CHAPTER III
RESEARCH DESIGN AND PROCEDURE

Introduction

This study regarding the competency of Bahrain teachers,
curriculum specialists, and prospective teachers in educational
measurement investigated the level of competency in educational
measurement, and the strengths and weakness of each group in specific
areas of educational measurement (e.g., item analysis, reliability, validity).
This chapter addresses the research instruments used, the sampling
procedures, the data collection process, the follow-up of non-respondents,

data analysis, and the characteristics of the subjects.
Research Instruments

Two research instruments were used in this study: an educational
measurement test (hereafter referred to as the 'TEST') and a questionnaire.
The TEST consisted of 65 multiple-choice items. A total of 50 test items
were taken from Boothroyd (1990). Very slight modifications were made on
some of those items in order to make the context of the items familiar to the
subjects of the study. For example, if the words 'New York ' appeared in the
stern of an item, they would be replaced by the word 'Manama' (the capital
city of Bahrain). Othenrvise, no other modifications were made. The

remaining 15 test items were constructed based on measurement and test

45

46

construction criteria addressed in measurement textbooks, such as Mehrens
and Lehmann (1991) and Wiersma and Jurs (1990).

To maintain the content-related validity of the TEST items, and to
permit valid inferences from the TEST, the researcher asked professors of
educational measurement at the University of Bahrain to respond to a table
of specifications by writing the percentage of instructional time they devoted
to each of the 13 measurement areas in their measurement course.

Based on the responses of the measurement professors at the
University of Bahrain, the researcher deleted some items and added some
other items to the TEST to insure the content-related validity of the
measurement test according to the content domain of the measurement
courses taught at the University of Bahrain. The TEST was then given to a
doctoral candidate in measurement, evaluation and research design at
Michigan State University to check for any flaws that might impede or
negatively affect any inferences drawn from the test. In addition, the
researcher gave the English version and the Arabic version of the TEST to
four doctoral students at Michigan State University to check for any errors
in translation of the test items from English to Arabic that would change the
meaning and result in incorrect inferences. Finally, the researcher gave both
the Arabic and English versions of the test to the Department of Arabic at
Michigan State University to verify the Arabic translation of the English
version of the test. Based on the recommendations of the five doctoral
students and the Department of Arabic, the researcher made some very
slight changes to the wording of some of the TEST items.

The questionnaire consisted of 23 questions. The selection of the

questions was based mainly on prior research findings (Boothroyd, 1990;

 

 

47

Newman, 1981; Mayo, 1967) which showed that these variables are valid
and related, in one way or another, to competency in educational
measurement. In addition to the demographic data, the questionnaire
included items asking for information on, for example, classroom testing
practices, item types used in classroom tests, the amount of time devoted to
testing, item analysis, current competency in educational measurement, and

practical training needed in educational measurement.

Subjects and Sampling Procedures
The target population consists of in-service teachers, curriculum
specialists, and prospective teachers in Bahrain. The in-service teachers
(n=1300) were randomly sampled by strata (primary, intermediate, and
secondary school). Because of their small number, the researcher studied

the total population of curriculum specialists (N=146) and prospective

teachers (N=221).

Table 3.1 shows the sample of teachers classified by sex and
grade level. The percentage of female teachers exceeds that of male

teachers both by individual grade level and across grade levels. The

largest sex difference, as one might expect, favored the percentage of

females at the primary level.

Table 3.1: Information on the sample of teachers by sex and grade level.

 
 

———-m_
M

 
     
     

 

 

48

Table 3.2 shows the number of participating schools for the sample
of teachers by grade level and sex. Table 3.3 shows the total population of
curriculum specialists and prospective teachers who participated in the

study.

Table 3.2 : Information on the participating schools by sex and grade level.

3—_——ME_

      
 

 
     

Table 3.3 : Information on the total population of curriculum specialists and
prospective teachers used in the study.

-ME_

Soecialists

 

Teachers

Characteristics of the subjects

The two research instruments were administered to 1300 teachers,
146 curriculum specialists, and 221 prospective teachers. Table 3.4 below
shows the response rate for the total sample and for each group. The
highest (82%) response rate was for the teachers who constitute 83.4
percent of the total sample.

The distributions by sex and age are given in Tables 3.5 and 3.6.
The mean age of the respondents in the total sample was 34.3 years with a

standard deviation of 8.1 years. The highest mean age among the three

 

 

 

49

Table 3.4: Information on the response rate for the total sample and for each
group.

       

      

 

ReSoonses
1070 82%
98 67%
115 52%
1283 77%

Sam . Ie
Teachers

Curriculum S- -cialists
Pros-ective Teachers
Total

 
     
 

     

groups was that of the curriculum specialists (42.4 years). With the

exception of the curriculum specialists, there were more females than males

in the total sample and in each group independently.

Table 3.5: Distribution of the respondents' age in the study.

Group n M SD SE MIN MAX
Teachers 1070 34.9 7.2 .2 21 60
Curriculum 98 42.4 5.9 .6 31 57
Specialists

Prospective 1 15 22.2 1.5 .1 20 29
Teachers

Total Sample 1283 34.3 8.1 .2 2O 60

Table 3.6: Distribution of the respondents' sex in the study.

—--ra___
485 45.3% 583 54.5%

S . ecialists
Teachers
Sam . Ie

       
   
     

    

 

 

50

The distribution of highest degree held among respondents (Table
3.7) shows the following: First, only 0.1 percent of the teachers hold a
doctorate, 1.0 percent of them have a masters degree, and more than half
of the teachers hold at least a bachelors degree (56.6 percent). Second,
nearly all of the prospective teachers (99.1 percent) are candidates for a
bachelors degree. The 25 percent of the teachers who hold a diploma
(associate degree) might be those who graduated from the Teacher Training
College in Bahrain between the mid-sixties and early-seventies before the
establishment of the University of Bahrain in the early seventies. Among the
curriculum specialists 25.5 percent have a bachelors degree, 39.8 percent
have a masters degree, and 16.3 percent have a doctorate. The year
highest degree was obtained ranged from 1958 to 1992 for the teachers,
from 1957 to 1992 for the curriculum specialists, and from 1992 to 1993 for

the prospective teachers.

Table 3.7: Distribution of the respondents by highest degree held.

S . ecialists Teachers
M

1o7( 01%) 17( 17. 4%) 0(0%)
Di-Ioma

606 56. 6% 25 25. 5% 114 99.1%
m 267 25%

Certificate
High School
Certificate

        
 
       
   
        
 

   
 

51

Most recent year when the highest degree held was received was 1986 (6.3
percent) for the teachers, 1986 (9.2 percent) for the curriculum specialists,
and 1992 (23.5 percent) for the prospective teachers. Most of the
respondents in each group stated that their highest degree held was
obtained from a college of education (see Table 3.8).

About 34 percent of the curriculum specialists and 37.1 percent of
the teachers stated that education was their major area of study when in the

bachelors degree program (see table 3.9).

Table 3.8: Information on the highest degree held obtained from a college of
education.

___-I5—
753 70.4% 240 22.4% 77 7.2%

S . ecialists
Teachers

Table 3.9: Distribution of the respondents whose major was education in the
bachelors program.

__-
Answer ,

-
___—_I-
mm.

film-_-
S oecialists

___-___
Teachers ,

      
     

 
   
  

     
 

 

 

52

The distribution of the respondents‘ college major is shown in Table

3.10. The largest percentage among the major subjects across the three

Table 3.10: Distribution of the respondents by major in college.

    
         
  
   
   

Teachers Curriculum

   

Prospective

Specialists Teachers

   

Islamic

   
 
 

Education

  

 

 
       
     
      
       
 

   
  

Arabic 268 25% 24(24. 5%)
12311.5% 11(11.2%) i
___-_—
24 2 2% 212%
20 1 9% 414 1%

inistration

  
  

  

Elementary 121 (11. 3%) 3. 103(89. 6%

 

Teacher

      

Physical 37(3.5%)

  
 

Education

7 7 W I ~-
-MI!E_.-:-,o——u—
1°61
3 33. 1%)
___—___—
11 11.962

 
  

  
 
  

  
 
       

53

groups is Arabic followed by mathematics. The results indicate that 11.3
percent of the teachers major in “class teacher" (elementary teacher), while
most of the prospective teachers (in reality all of them) specialize in the
“class teacher" major. The next highest percentages are for the major
subject labeled “other': 7.4 percent, and 11.2 percent for the teachers, and
curriculum specialists, respectively. The “other" major subjects include

areas such as French, Library Science, Sociology, Home Economics, and

Educational Resources.

Only .6 percent of the teachers majored in Computer (not shown in
Table 3.10). Less than 4 percent of the teachers (3.2%) , and 10.4 percent of
the prospective teachers did not provide an answer about their major in
college.

The distribution for each group regarding their teaching/ supervision
subject is displayed in Table 3.11. As might be expected, the results show
that the percentages in tables 3.10 and 3.11 are quite similar, especially for
Arabic, Mathematics, English, Class Teacher, and Fine Arts.

The results show that only .5 percent of the teachers teach
computers (not shown in Table 3.11). 6 percent of teachers, 1.1 percent of
curriculum specialists, and 13.9 percent of prospective teachers did not
provide an answer.

The distribution of each group by grade level is shown in Table

3.12. The table shows that the highest percentage is in the primary grade

level.

 

 

 

54

Table 3.11: Distribution of the respondents by teaching/ supervision subject.
Specialists Teachers
Education

268 25% 23123.5% _
Mathematics 146 13.6% 12(12.2% —
m 59 5.1% 616.1% —

129(12.1%) 9(9.2%) 99(86.1%)
Teacher

—m--_—
—-—-——
___-___—
___-___
-l'---—_
___-_—

42(3 9%) _-
Sub'ects

126118% 331%
272.15% 771%
M—

Education

-'---—_
M—
33 3.1% 5 5.1%) _

          
 
   

     
      
 
     
   

   

 
  
  

Elementary

 
 
 
   
   
   
  
 

 
       

Industrial

 
 
   
    
 

  

Physical

     
   
   

 

 

 

55

Table 3.12: Distribution of the respondents by grade level.

j Soecialists Teachers
———__—i
m-Mmre- as. 1

"mule-_-
, mediate

“at. 24 5
_-_I_ 10.2 7 —=

  
   
 

   
  
        

Information on the average number of years in teaching for each
group is given in Table 3.13. On average teachers have more experience in
teaching than curriculum specialists. As expected, teachers have up to 42
years of experience in teaching, while some curriculum specialists have a
maximum of 25 years of experience in teaching. In addition, the variability in

teaching experience among teachers is slightly higher than that among the

Table 3.13: Distribution of the respondents by the average number of years

in teaching.
mm
m
Curriculum 11.9 11 6.1 .7 1 25
Specialists

Prospective .01 .00 .1 .01 .00 1

Teachers

 

 

56

curriculum specialists. Apparently, the distribution of the range of
experience in curriculum supervision among curriculum specialists is close
to normal (Mean=9.3, Mode=10, Median=10, with a standard deviation of 4.5

years). The range in years of experience in curriculum supervision for this

group is 1- 20 years.

Data Collection

After having the Arabic translation of the English version of the
research instruments verified by the Department of Arabic at Michigan Sate
University, obtaining permission to conduct the study from the University
Committee on Research Involving Human Subjects (UCRIHS) at Michigan
State University, and from the officials in the Bahrain Ministry of Education
and the University of Bahrain, the researcher traveled to Bahrain to collect
the data. I

The curriculum specialists were notified about the study and it took
them a few hours to complete the materials. The data collection process for
teachers, however, was somewhat different. Due to the large sample of
teachers (n=1300), and in order to save time and effort, the researcher (with
the assistance of a team of 15 curriculum specialists) collected the data.
The researcher met with the team for 3 hours, explained to the them the
process of data collection in full detail and their role in that process, and
assigned them the participating schools. The team of curriculum specialists
then personally delivered to their schools the sealed envelopes that
contained the appropriate number of booklets of the research instruments,
along with a cover letter explaining the goals of the study and requesting the
teachers' responses. The school principals were asked to supervise their

respective teachers responding to the research instruments, collect the

 

 

 

 

57

booklets, and have them ready for the curriculum specialists to collect the
following morning. For the prospective teachers, the researcher obtained a
list of their names and their respective advisers from the office of the
registrar at the University of Bahrain. The researcher contacted the
advisers personally and sought their cooperation to have their students

respond to the research instruments and return the materials the following

day.

Follow-Up of Non-Respondents

As the response rate for the prospective teachers was low
compared to that of teachers and curriculum specialists, and to determine
whether the results of this study would be affected by non response bias (for
the prospective teachers), the researcher attempted to draw a random
sample of non-respondents. The University of Bahrain, however, denied the
researcher any access to the addresses of non-respondents. In addition,
the respective advisers were unable to help in that process. Due to these
problems, the researcher is unable to generalize to the total population of

prospective teachers.

Data Analysis
The data were analyzed using the SPSSPC. Statistical analyses
were conducted to answer the research questions and test the research
hypotheses proposed in chapter one. Research questions 1, 2, 3, and 4
(see Chapter 1) were answered by calculating descriptive statistics (e.g.,
frequencies, mean, standard deviation, standard error of the mean, and
percentages). The point biserial correlation coefficient (rpbis) and item

difficulty indices (p-values) were computed to supply information on the

 

 

58

performance of respondents on the TEST items. Pearson product moment
correlation coefficients were computed to provide information for research
hypotheses 1 and 2. Research hypotheses 3 through 12 were tested by
conducting a t-test, and one-way and two-way analysis of variance
(ANOVA). The KR2O reliability for the TEST scores was computed for the
total sample and for each group separately. Multiple regression (MR)

analysis was conducted to determine what variables best predict the total
score on the TEST. In addition, descriptive statistics (e.g., mean, standard
deviation) were also computed. A significance level of .05 was used

throughout the study.

Summary

The purpose of this study was two-fold: (1) To assess the current
competency of Bahrain teachers, curriculum specialists, and prospective
teachers in educational measurement, and (2) to examine their strengths
and weaknesses in specific areas in educational measurement. Exploring
the factors that affect the respondents' competency in educational
measurement, the differences among the three groups in educational
measurement, the current classroom testing practices, and the needs for
practical training in educational measurement are the issues this study was
designed to investigate.

The sample of teachers (n=1300), representing 21 percent of the
population of Bahrain teachers (N=6175), was distributed among 111
schools. Although all the available curriculum specialists (N=146) and
prospective teachers (N=221) were scheduled to participate in the study,
only 98 curriculum specialists and 115 prospective teachers participated.

Descriptive and inferential statistics were used to analyze the data.

 

 

CHAPTER IV
ANALYSIS AND INTERPRETATION OF THE DATA

Introduction

The study's findings are presented in these sections. First, the
results that deal with classroom testing practices, measurement preparation,
and measurement training needs (that pertain to research questions 1, 2,
and 3) will be reported. Second, subjects' performance on the TEST (that
pertains to research question 4) will be discussed for each group separately.
Third, the findings pertaining to the research hypotheses are presented.
Some findings of previously conducted research studies, such as

Boothroyd's (1990) and Newman's (1981) studies, will be reported where

appropriate.

Classroom Testing Practices and Measurement Preparation

Research Question 1: What is the actual preparation of Bahrain
teachers, curriculum specialists, and prospective teachers in

measurement?
Research Question 2: What are the current testing practices

of Bahrain teachers, curriculum specialists, and prospective
teachers?

Table 4.1 presents the findings of the amount of measurement coursework
completed by each of the three groups. From the table it is seen that most

of the prospective teachers (80 percent) had, in contrast to 35.6 percent of

59

 

 

 

 

60

the teachers and 26.5 percent of the curriculum specialists, completed one
measurement course. More than one third (36.7 percent) of the curriculum
specialists had more than one college course in measurement, whereas only
9.3 percent of the teachers and 4.3 percent of the prospective teachers had
more than one college course in measurement. Only 16.2 percent of the
teachers and 13.3 percent of the curriculum specialists had in-service

training in measurement.

The results of the analysis of the statistics coursework completed
are shown in Table 4.2. These results reveal that about one third of the
teachers have never been exposed to any kind of independent statistics
coursework in contrast to 14.3 percent and 8.8 percent of the curriculum
specialists and prospective teachers respectively. Regarding 'one college
course in statistics', the figures in Table 4.2 show that the prospective
teachers had more statistics coursework (43.5 percent) than the curriculum
specialists (34.7 percent) and the teachers (21.7 percent). The teachers
appear to have the lowest percentage among the three groups who
received one college course in statistics. The curriculum specialists had the
highest percentage among the three groups who had multiple statistics
courses (28.6 percent). This may be attributed to the fact that the curriculum
specialists were exposed to more academic coursework than the other
groups. The teachers had more in-service training in measurement than they
had in statistics (see Tables 4.1, 4.2). This may call for more emphasis on

statistics either at the college level or as part of in-service training.

 

 

61

Table 4.1: Distribution of the respondents by measurement coursework
completed.

Coursework Specialists Teachers
162 15.1% 9(9.2% 4 3.5%

Part of another 129(12.1%) 9(9.2%) 1(.9%)
course

381 35.6% 26l26.5% 92 80%

colleoe course
___—
Trainin

125(11.7% 515.1%) 13 11.3%)

      
     
       

Table 4.2: Distribution of the respondents by statistics coursework
completed.

Coursework Specialists Teachers

course

course

colle-e course
___—

trainin .

m 169 15.8% 7 7.1) 17 14.7%)

      
      
 

    
   
     

From Table 4.3 it can be seen that nearly one-half (46.3 percent) of
the teachers stated they spend more than 20 percent of their professional
time on classroom testing activities, and about one third of them said that
they spend between 11 percent and 20 percent of their professional time on

classroom testing activities. These percentages seem to be consistent with

 

 

62

prior research findings (e.g., Stiggins 8 Conklin, 1988; Stiggins, 1991;
Schafer, 1991; Green 8 Stager, 1986-1987; 8 Gullickson, 1982). The
curriculum specialists, however, devote less time to classroom testing

activities than teachers but more than the prospective teachers.

Table 4.3: Distribution of the respondents by percentage of time devoted to
classroom testing.

Time S-ecialists Teachers

    
   
 
     
    

Table 4.4 shows how the three groups use teacher-made tests.
Using classroom tests exclusively for each purpose ranged from .5 percent
to 5.1 percent across the groups.

While diagnosis, achievement, and mastery for learning,
respectively, rank the highest among the purposes for classroom testing for
teachers in this study, Boothroyd (1990) found that mastery and
understanding of the content was the most frequently purpose cited by

teachers, followed by remediation and improving instruction, and grading.

 

 

 

 

63

Table 4.4: Distribution of the respondents by purposes of classroom test

USS.

S necialists Teachers
—m—-

“m“
leamin- 3
u-roress 4
___——
leamin- 5

_lI--—_

instruction 7
—m-_—
”Fm-_—

       
   
   

     
     
     

     
     
   

Using only one item type for classroom tests seems to be rare
among the groups (see Table 4.5). Using a combination of more than one
item type in the classroom is the trend for both prospective teachers and
teachers (20 percent and 18.6 percent, respectively) who said that they use

true-false, matching, completion, short answer, and multiple choice items on

their classroom tests.

 

 

64

Table 4.5: Item types used in classroom ( percentages).

Item Type Teachers
S uecialists Teachers

___——
_ZI——‘—
_-‘l——“

___-I
Answer4
Choice5

—’l-—_—
1.2,4.5
1.3.4.5

1.2.3.4.5
1,234.56 ___

    
   

Curriculum Prospective

        
  

  

 

     

  
 

   

 
   
     

The least used item types, in general, are matching and completion.

Information on item types ranked according to their preference for
each group is given in Table 4.6 . The table shows that, by and large, essay
item type is the most preferred and multiple choice item type is the least
preferred by the three groups. Apparently, rankings of the teachers and
prospective teachers with respect to true-false, Multiple-Choice, and essay
item types are identical, whereas they are completely different from the
rankings found by Reynolds and Menard (1980) and Newman (1981). For
example, Newman (1981) found that the completion item type was ranked

the highest, followed by Multiple-Choice item type.

65

Table 4.6: Item types ranked according to preference'.

Soecialists Teachers

___—
Choice

Answer
“—_—

Note: 1=Most preferred, 6: Least preferred.

        
   
       

Most of the teachers and curriculum specialists stated that they
prepare a test plan and conduct item analysis of classroom tests. The
percentage ranged from about 30.4 percent to 76.2 percent for the test plan
variable, and from 43.5 percent to 84.1 percent for the item analysis
variable (see Tables 4.7 and 4.8). The prospective teachers, however, had
the lowest percentage among the three groups regarding the test plan

preparation (30.4 percent) and item analysis of classroom tests (43.5

percent).

Table 4.7: Distribution of the respondents by test plan preparation.

mm
815 76. 2% 227 21.2% 28 2. 6%

73( 74. 5%) 141.43% 11 (.)112%
S-ecialists
Teachers

       
     

 

66

Table 4.8: Distribution of the respondents by item analysis practice.

    

 

,gi v ___,7. _ii, . ..igj

" Group I
______
l—m--—_—

__1
S - ecialists
___-
. Teachers

Data on self-ratings of each group concerning the general

   

 

   
     

competency in educational measurement and competency in test
construction are displayed in Tables 4.9, and 4.10, respectively. Less than
one-half (42.2 percent), 30.6 percent, and 21.7 percent of the teachers,
curriculum specialists, and prospective teachers, respectively, perceive
themselves as very competent in measurement in general. Regarding the
perceived competency in test construction, Table 4.10 shows that 46.1
percent of the teachers, 35.7 percent of the curriculum specialists, and 23.5
percent of the prospective teachers believe that they are very competent in
test construction. Moreover, subjects in each group tended to view
themselves, in general, as more competent in test construction than in

educational measurement as a whole.

Table 4.9: Distribution of the respondents' self-ratings regarding their
general competency in educational measurement.

S . ecialists Teachers
M
mum-_-

      
    
 

    
     

67

Table 4.10: Distribution of the respondents' self-ratings concerning their
competency in test construction.

- S-ecialists Teachers ;
__—_—_‘
_E-E-El-l.“
MEI-EM‘
Hem—“MEI
MEI-E-
[m_____—l=-

it." 1 mm
Answer

 
  
 

   
   
       
   

Research Question 3: In what areas of educational measurement
do curriculum specialists, teachers, and prospective teachers need

practical training?
Table 4.11 gives a general indication of the practical training needs

Table 4.11: Distribution of the respondents' practical training needs in
educational measurement ( percentages).

Specialists Teachers
mm 2.0 l 1.7

___—
construction
___—
selection

W“ 7.1 l 3.5

“___“
Reliabili _—

lnte oretation
_—
methods

___

       

      
     
     
       
   
     

 

 

68

in educational measurement. The percentages in the table are small due to
low response rate for this question, and may not highlight the needs of the

majority in each group.

Performance of the Subjects on the TEST

The KR20 reliability and standard error of measurement (SEM)
estimates for the sets of scores from the TEST are given in Table 4.12. The
TEST appears to produce reliable scores for the total sample and for each
group independently.

Boothroyd ( 1990) set a standard of .54. Since this study used most
of his items, our standard in this study might also be .54. However, since 15
of the TEST items were new, we selected a standard of .50. Hence, our
standard reflects the estimated probability of an individual (e.g., a teacher)

with 'minlmal competency in measurement' correctly answering an item.

Table 4.12: KR20 reliability and SEM estimates for the TEST.

—m_
___E-

—m_

s . ecialists

---1l
teachers

Total sam Ie

      
   
     

Research Question 4: What is the current level of knowledge and
understanding of Bahrain teachers, curriculum specialists, and
prospective teachers in measurement?

(a) how competent is each group in measurement?

(b) in which measurement area(s) is each group the most and

the least competent?

 

69

Table 4.13 provides information on the overall performance of each
group on the TEST and their performance on each of the 13 areas that
comprise the total test. It can be seen from the table that the teachers' total
mean score on the TEST was 26.72 (41 percent), the curriculum specialists'
total mean score was 37.49 (58 percent), and the prospective teachers' total
mean score was 34.07 (52 percent). The scores on the TEST ranged from
9 to 48 for the teachers, from 19 to 50 for the curriculum specialists, and
from 18 to 49 for the prospective teachers. It should be noted that the
curriculum specialists outperformed the teachers and the prospective
teachers in all 13 areas of the TEST, except for the areas of "reliability' and
'types of items". The teachers performed better in the areas of 'objectives',
and 'grading and marking' than they did in all the other areas of the
TEST. In general, the curriculum specialists performed well in all areas,
except for the areas of 'types of tests' and “standard error of measurement."
The prospective teachers scored better in the areas of “objectives", “types of
items", 'item writing', 'grading and marking', 'correlation" “reliability", and
"validity" than they did in the other areas of the TEST. The table shows that
the teachers are weakest in more areas than the other groups. The
teachers need more training in different areas such as 'item analysis',
“score interpretation” and “standard error of measurement." The curriculum
specialists need more training in the areas of 'types of tests' and “standard
error of measurement." The prospective teachers need more training in the
areas of 'test planning', 'types of tests', 'test construction", 'item analysis',
'score interpretation', and 'standard error of measurement'. All the three
groups together are weak in the areas of 'types of tests' and 'standard error

of measurement'. Plake, lmpara, and Fager (1992), however, found that

teachers (n=555i demonstrated the highest performance in the areas

flit-w I '_

“I: .'i@:"'-'mifltc‘-t- . .. . _ .m.

     

.lfl' Ht 011' rat-3 a! fur: I1 In 'an? M": M

70

Table 4.13: Distribution of the respondents' performance on the TEST by
content area.

Content ft of Teachers Curriculum Prospecti-
Area Items Specialists ve Teachers

___——

Planni g

___ 4.6318751 628 1.36 70 582 180 65

___—m

Tests

___—m
Items

_M— 29615042 4.4015063 41815660

Test 178 9.145 2.39 .84 60 1.861.0046
Constru-

——m—_
Ana sis
=tatio-n
gnu—w“ ,
Martdn .

Correlatio—n _— 121.8540 200 83 67 _7710159 _
Rellablll — 1881.1638 3001.3460 3.0911762 ;

Valldi — .97 as as
__“ 374982758 340771952

       
       
         
    
         
       

0‘ administering, scoring, and interpreting test results.

Item analysis data for the TEST for each group are displayed in
Tables 4.14 through 4.16 (see Appendix F). The TEST item difficulties (p-
values), indicating the proportion in each group answering an item correctly,
ranged from .10 to .79 with an average difficulty of .40 for the teachers, from
.00 to .94 with an average difficulty of .57 for the curriculum specialists, and
from .03 to .90 with an average difficulty of .57 for the prospective teachers.

Point biserial correlation coefficients, which are an index of item

 

.-—'---—

'--’-I—

----;_n, fir. .

t.

.0

MI -

71

discrimination, ranged from -.23 to .43 with an average correlation of .16 for
the teachers, from -.29 to .58 with an average correlation of .23 for the

curriculum specialists, and from -.32 to .64 with an average correlation of .18

for the prospective teachers.

_The most discriminating items, those with a discrimination level of
.40 or above, were items 9 and 33 for the teachers, items 9, 14, 15, 20, 27,
33, 38, 39, 41, 44, 45, 49, 52, 57, and 63 for the curriculum specialists, and
items 7,19, 20, 33, 39, 44, 48, and 62 for the prospective teachers. In

general, the lowest performance on the TEST across the three groups was

on item 56.

The overall perceived needs of the respondents regarding practical
training in educational measurement (see Table 4.11), especially in the
areas of item analysis and standard error of measurement, may reflect their

weakness in these areas based on their performance on the TEST (see

Table 4.13).

Testing the Research Hypotheses

Research Hypothesis 1: For each of Bahrain teachers,
curriculum specialists, and prospective teachers there is no
significant relationship between their measurement
competency and highest degree held, recency of highest
degree held, major subject of specialization, major subject of
teaching/ supervision, grade level of teaching/ supervision,
years' of experience in teaching, years of experience in
supervision, amount of measurement coursework completed,
recency of measurement coursework completed, amount of
statistics completed, recency of statistics completed,
percentage of professional time devoted to test construction,
and purposes of using teacher-made classroom tests.

Research Hypothesis 2: For each of Bahrain teachers,
curriculum specialists, and prospective teachers there is no
significant relationship between their measurement

.q'

“'ifiupi-id-“uinu ' I .

fly is use... natal“: *"W-sm infirm

"."~.
, ‘ I. '._ -' .'J."'I Hi

 

 

72

competency and developing a test plan prior to writing test
items, conducting item analysis on self-constructed classroom
tests, self-assessment of the level of competency in
measurement in general, and self-assessment of the level of
competency in test construction principles.

Product moment correlations were computed to investigate the
relationship between competency in educational measurement as measured
by the TEST, and measurement preparation variables, classroom testing
variables, and subjects' background variables. The results of these analyses
are displayed in Table 4.17. We should be very cautious in interpreting
these findings for the following reason. Despite the fact that some significant
correlations were obtained, the proportion of variance in the TEST scores
accounted for by any one of these variables was very small; the largest r2
was .08. In other words, in some cases it is possible that trivial or negligible
relationships were significant because of the large sample size. This might
be the case with the group of teachers where there were four significant
relationships; the sample size of the teachers is the largest among the three
groups. In general, the findings in Table 4.17 indicate that the correlations
are small or negligible, and some of them are negative. Although the
highest correlation coefficient was -.37 (r2=.14 ), it is not significant, and
negative as well. These negative correlations signify that high values on the
variables in the table are associated with low scores on the TEST, and vice
versa. Some variables, such as developing a test plan, conducting item
analysis, self-assessment of competency in educational measurement, and
purposes of using teacher-made classroom tests (for the teachers group),
indicated a significant relationship with the scores on the TEST. The
results revealed that those teachers who stated they develop a test plan

and conduct item analysis scored low on the TEST.

 

 

73

Table 4.17: Correlations between the scores on the TEST and classroom
testing activities variables and measurement preparation variables.

Variable Teachers Curriculum Prospective
Secialists Teachers
1 . .

—-_——

 

'P<.01; "F<.001; . Coefficient cemot be computed; — Does not 8: these 9. 1=H' st ce held,
Hecency of highest degee held, 3=Mejor subject of spedalizafionm subje'ctiu :1 Miwwptrasion, 5:
Grade level of teaching! supervision, 6=yeers of experience In feeding, 7=yeus of experience in supervision, 8=Amomt
of meesurermnt couscwork complebd, 9=recency of meesurenmt cornework cornplebd, 10=Amount of statistics
cemplebd, 11=Recemy of statistics complebd 12: percentage of professional ime riveted t) test construction,
13=Purposes of using classroom tests, 14=Developing a test phn, 15=Condicting Item analysis, 16=SeII-essessment
of general competency in measurement, 17=SeIf-essessment of competency ‘81 test construction.

Those teachers who perceived themselves as competent in educational
measurement and use classroom tests for a myriad of purposes, scored
high on the TEST. To see the strength of association between the scores on
the TEST and each of these two variables ( measurement competency
and purposes of using classroom test ) for the group of teachers, a one-way
ANOVA was conducted on the TEST scores with each of these two variables

(see Tables 4.18, 4.19), in addition to the index of association strength
((1)2) computation.

 

74

Table 4.18: ANOVA for the TEST scores with 'purposes of classroom test'
variable for the teachers.

Source D.F. Sum of Mean

Squares Square F-Ratio F-Prob. 0:2
Between 6 471.5559 78.5927 1.5087 .1846 .03
Groups
Within Groups 88 4584.1915 52.0913
Total 94 5055.7474

From Table 4.18 we see that there is no significant difference in
measurement competency between the teachers who use classroom tests
for a myriad of purposes and the teachers who use classroom tests for one

or two purposes, where the index of association strength has a negligible

value (032:.03). This suggests that the shared variance between the TEST
scores and 'purposes of classroom test use' variable is very small. Hence,
measurement competency does not depend on using classroom tests for
several purposes. Despite the fact that there is a significant difference
between the teachers who perceived themselves as competent and those
teachers who perceived themselves as incompetent in measurement (see
Table 4.19), the proportion of the shared variance between the scores on the
TEST and the perceived competency in measurement is negligible (09:02).
In addition, the Scheffe test showed that the significant difference was only
between the teachers who perceived themselves as 'excellent" and the

teachers who perceived themselves as 'good' in measurement.

 

 

75

Table 4.19: ANOVA for the TEST scores with the 'perceived level of general
competency in measurement' for the teachers.

Source D.F. Sum of Mean F-Ratio F-Prob (1)2
Squares Squares

Between 4 957.8273 239.4568 5.0034 .0006 .02
Groups

Within 659 31538.6486 47.8583
Groups

Total 663 32496.4759

Newman’s (1981) results seem to be comparable to the results of
_ this study. For example, Newman found that there was a relationship
between the scores on the measurement test and variables such as the
amount of measurement coursework (r=.10), recency of measurement
coursework (r=.15, p<.05), years of teaching experience (r=-.14, p<.05),
highest degree held (r=.13, p<.05), recency of highest degree held (r=.09),
percentage of time devoted to classroom testing (r=.06), and purposes of
using classroom tests (r=.22, p<.01). The findings of this study showed that
the variable 'purposes of using classroom tests' has a significant
relationship with scores on the TEST (r=.21, p<.001) for the teachers.
The conclusion regarding rejecting or not rejecting the research hypotheses
1 and 2 is as follows: For the teachers, research hypothesis 1 is rejected for
the 'purposes of using classroom tests' variable and not rejected for the rest
of the variables in the hypothesis. The research hypothesis 2 is rejected

for the variables 'developing a test plan', 'conducting item analysis', and

76

"self-assessment of general competency in measurement". The hypothesis
is not rejected for the rest of the variables in the hypothesis. Concerning the
groups of curriculum specialists and prospective teachers, the hypotheses 1
and 2 are not rejected for all the variables in these hypotheses. By and
large, it can be concluded that measurement preparation variables,
classroom testing variables, and subjects' background variables do not have

that large of an influence on competency in educational measurement.

Research Hypothesis 3: There is no significant difference in measurement
competency among Bahrain teachers, curriculum specialists, and
prospective teachers.

The analysis of variance (ANOVA) test results are displayed in
Table 4.20b. It is evident that there is a significant difference in
measurement competency among Bahrain teachers, curriculum specialists,
and prospective teachers. The shared variance ((02) between the type of
respondent and measurement competency was .18. The Scheffe test
indicated that the teachers' mean score on the TEST was significantly lower
than the mean score of the curriculum specialists. This may signify that the
curriculum specialists are more competent in educational measurement than

the teachers (see also Table 4.20a). Hence, the hypothesis is rejected.

77

Table 4.20a: Distribution of the respondents' mean performance on the

TEST.

__m- 56 MW
25.72 “—-

___--

S - ecialists

___--
Teachers

Table 4.20b: ANOVA for the TEST scores with the 'type of respondent'.

   
 
 

Source D.F. Sum of Mean F-Ratio F-Prob. (1)2
Squares Squares

Between 2 8726.6013 4363.3006 86.5545 .0000 .18

Groups

Within 804 40530.4570 50.4110
Groups

Total 806 49257.0583

Research Hypothesis 4: There is no significant difference in measurement
competency (for each group separately) between those who had
measurement coursework and those who did not.

Table 4.21b presents the analysis of variance data (for each group)
for the scores on the TEST and measurement coursework completed. From
the table it can be seen that there is a significant difference between those
who had measurement coursework and those who did not regarding the
competency in educational measurement (see also Table 4.213). The

Scheffe test showed that for the teachers, those who completed one

78

measurement course are significantly different from each of the following:
those who had no measurement coursework at all, those who had
measurement as part of another course, and those who had in-service
training in measurement. For the curriculum specialists, those who had
measurement coursework as part of another course appear to be
significantly different from: those who had no measurement coursework at
all, those who had one measurement course, and those who completed
more than one measurement course. For the prospective teachers, the
significant difference was found between those who had one measurement
course, and those who had no measurement coursework at all. It can be
concluded that competency in measurement is partly attributable to the
amount of measurement coursework or training in measurement. In spite of
these findings, the degree of association between measurement coursework
and competency in educational measurement as measured by the TEST is
not strong. For example, (1)2 shows that the shared proportion of variance

between measurement coursework and competency in educational

Table 4.21a: Distribution of the respondents by measurement coursework
completed.

Coursework S-ecialists Teachers
_I5_ 162 15.1% 9 9.2% 4 3.5%

129(12.1%) 9(9.2%) 1(.9%)
course

381 35.6% 26 26.5% 92 80%

100(9.3%) 36(36.7%) 5(4.3%)
colleoe course

173(16.2%) 13(13.3%)
Trainin.

m 12511.7% 5 5.1% 1311.3%

       
   

  
     
      

 

79

measurement is 7 percent for the teachers, 24 percent for the curriculum

specialists, and 9 for the prospective teachers. Therefore, this hypothesis is

rejected for each group independently.

Table 4.21b: ANOVA for the TEST scores with 'measurement coursework'.

TEACHERS
Source D.F. Sum of Mean F-Ratio F-Prob. 812

Squares Squares
Between 4 2259.8169 564.9542 12.0814 .0000 .07

Groups
Within 595 27823.5814 46.7623
Groups
Total 599 30083.3983
CURRICULUM SPECIALISTS
Source D.F. Sum of Mean F-Ratio F-Prob. <52
Squares Squares
Between 4 961.7224 240.4306 4.8253 .0026 .24
Groups '
Within 44 2192.4000 49.8273
Groups
Total 48 3154.1224
PROSPECTIVE TEACHERS
Source D.F. Sum of Mean F-Ratio F-Prob. (02

Squares Squares
Between 3 446.0193 148.6731 3.2053 .0287 .09

Groups

Within

Groups 66 3061.3522 46.3841
Total 69 3507.3715

Research Hypothesis 5: There is no significant difference in
measurement competency (for each group separately) between those who
had coursework in statistics and those who did not.

 

80

The data used to test hypothesis 5 are given in Table 4.22b. From
Table 4.22b one can conclude that there is a significant difference in
measurement competency among both teachers and curriculum specialists
depending on the type and amount of coursework in statistics (see also
Table 4.22a). The Scheffe test revealed that, for the teachers, those who
had no statistics coursework are significantly different from those who had it
as part of another course, and those who had one course. For the
curriculum specialists, there was a significant difference between those who
had no statistics coursework at all and those who had more than one
statistics course. The index of association strength (0)2) shows that the
proportion of shared variance between statistics coursework and
competency in educational measurement is 4 percent for the teachers, and

16 percent for the curriculum specialists.

Table 4.22a: Distribution of the respondents by statistics coursework
completed.

Coursework Specialists Teachers
m

course

course

colleue course
___—

trainin .

16915.8% 7 7.1) 1714.7%)

      
      
 
      

 

 

81

Table 4.22b: ANOVA for the TEST scores with “statistics coursework'.

TEACHERS
Source D.F. Sum of Mean F-Ratio F-Prob. <52
Squares Squares
Between 4 487.3506 371.8376 7.8120 .0000 .04
Groups
Within 569 27083.3550 47.5982
Groups
Total 573 28570.7056
CURRICULUM SPECIALISTS
Source D.F. Sum of Mean F-Ratio F-Prob. (1)2
Squares Squares
Between 4 660.4658 165.1164 3.2409 .0207 .16
Groups
Within 43 2190.7842 50.9485
Groups
Total 47 2851.2500
PROSPECTIVE TEACHERS
Source D.F. Sum of Mean F-Ratio F-Prob.
Squares Squares
Between 3 295.5489 98.5163 2.0064 .1219
Groups
Within 64 3142.5099 49.1017
Groups
Total 67 3438.0588

No significant difference was found for the group of the prospective
teachers. Hence, this hypothesis is rejected for the teachers and curriculum

specialists but not rejected for the prospective teachers.

 

 

82

Research Hypothesis 6: There is no significant difference in
measurement competency (for each group separately) between those who
develop a test plan and those who do not.

From Table 4.23b it is seen that there is a significant difference in
measurement competency between teachers who develop a test plan and

those who do not. 0:2 was only .06. There is no significant difference in

measurement competency between those curriculum specialists and those

Table 4.23b: ANOVA for the TEST scores with 'test plan'.

TEACHERS
Source D.F. Sum of Mean F-Ratio F-Prob. (1)2
Squares Squares
Between 1 1929.9029 1929.9029 41.7609 .0000 .06
Groups
Within 662 305930836 46.2131
Groups
Total 663 32522.9865
CURRICULUM SPECIALISTS
Source D.F. Sum of Mean F-Ratio F-Prob.
Squares Squares
Between 1 7.3500 7.3500 .1154 .7356
Groups
Within 46 2928.6500 63.6663
Groups
Total 47 2936.0000
PROSPECTIVE TEACHERS
Source D.F. Sum of Mean F-Ratio F-Prob.
Squares Squares
Between 1 23.6686 23.6686 .4204 .5193
Groups
Within 59 3322.0035 56.3051
Groups

Total 60 3345 .6721

 

 

83

prospective teachers who develop a test plan and those who do not (see
also Table 4.238). Therefore, the hypothesis is rejected for the teachers, but

not rejected for the curriculum specialists and prospective teachers.

Table 4.23a: Distribution of the respondents by test plan preparation

___-I_—
815 76.2% 227 21.2% 28 2.6%

S . ecialists
Teachers

Research Hypothesis 7: There is no significant difference in

 

measurement competency (for each group separately) between those who
conduct item analysis and those who do not.

The results on Table 4.24b show that there is a significant
difference in measurement competency between those teachers who stated
they conduct an item analysis and those who stated that they do not
conduct item analysis (see also Table 4.24a). Despite this significant

difference, the degree of association between the two variables appears to

Table 424a: Distribution of the respondents by item analysis practice.

      

___—'
___-_I-

_m
___-__-
S . ecialists
—-----m‘
. Teachers , . ‘

 

  
 

 

84

be small, where the shared proportion of variance between the two variables

((92) was 1 percent for the teachers. The hypothesis therefore is rejected

for the

prospective teachers.

teachers,

but not

Table 4.24b: ANOVA for the TEST scores with 'item analysis'.

Source

Between
Groups
Within
Groups

Total

Source

Between
Groups

Within
Groups

Total

Source

Between
Groups

Within
Groups

Total

D.F.

1

664

665

D.F.

46

47

D.F.

58

59

TEACHERS
Sum of Mean F-Ratio

Squares Squares
303.5957 303.5957 6.2371

32320.4659 48.6754

32624.0616

CURRICULUM SPECIALISTS
Sum of Mean F-Ratio

Squares Squares
8.8889 8.8889 .1397

2927.1111 63.6329

2936.0000
PROSPECTIVE TEACHERS
Sum of Mean F-Ratio

Squares Squares
163.2983 163.2983 2.9762

3182.3517 54.8681

3345.6500

F-Prob.

.0128

F-Prob.

.7103

F-Prob.

.0898

rejected for the curriculum specialists and

(92

.01

 

85

Research Hypothesis 8: There is no gender by type of respondent (e.g.,
a teacher) interaction regarding their competency in educational
measurement.

Table 4.25 displays the information used to test this research
hypothesis. It is evident that there is a significant interaction between
respondents' gender and the type of respondent concerning their
competency in educational measurement. This may indicate that the
conceptual understanding of the subjects in educational measurement from
one group to another is affected by the gender of the subject. The

hypothesis is rejected.

Table 4.25: ANOVA for the joint effects of gender and type of respondent
with the TEST scores.

Source Sum of D.F. Mean F-Ratio F-Prob.

Squares Squares
Gender 3192.194 1 3192.194 69.812 .000
Type of 8584.720 2 4292.360 93.872 .000
Respondent
Interaction 697.966 2 348.983 7.632 .001
Residual 36580.530 800 45.726
Total 49173.029 805 61.085

Research Hypothesis 9: There is no type of respondent (e.g., teacher)
by respondents' age interaction regarding their competency in educational
measurement.

From Table 4.26b it is seen that the F interaction value (between

the variables age and type of respondent) is 1.864, p>.05. This leads us to

86

conclude that the competency in educational measurement among teachers,
curriculum specialists and prospective teachers does not depend on whether
the respondent is old or young (see also Table 4.26a). This hypothesis is

not rejected.

Table 4.26a: Distribution of the respondents' age in the study.

 

Table 4.26b: ANOVA for the joint effects of age and type of respondent with

the TEST scores.

Source Sum of D.F. Mean F-Ratio F-Prob.
Squares Squares

Age 2521.112 7 360.159 7.477 .000

Type of 6490.285 2 3245.143 67.371 .000

Respondent

Interaction 538.791 6 89.799 1.864 .084

Residual 35789.126 743 48.168

Total 46802.358 758 61 .745

 

 

87

Research Hypothesis 10: There is no significant difference between
male and female teachers in their measurement competency.

Table 4.27 indicates that there is a significant difference between
male and female Bahrain teachers in their measurement competency
(p<.05). This may suggest that there is a difference in the conceptual
understanding between male and female teachers in different areas in
educational measurement. Females may be more competent in educational

measurement than males. The hypothesis is rejected.

Table 4.27: t-test for the difference between male and female teachers in
educational measurement.

———mln

Grou . of cases MEAN

“___“
24 1023 6 372

-m—-—
28 8997 6 741

   
   
     

 

Research Hypothesis 11: There is no significant difference among
teachers at different grade levels in measurement competency.

Table 4.28 shows that there is no significant difference (p>.05)
among teachers concerning their competency in measurement at the three
grade levels. This hypothesis is not rejected. Hence, the grade level has
no significant effect on the competency of the teacher in educational

measurement.

88

Table 4.28: ANOVA for the grade level of the teacher with the TEST scores.

Source D.F. Sum of
Squares

Between 2 107.0321

Groups

Within 761

Groups

Total 763 45902.4123

Mean
Squares
53.5160

F-Ratio

45795.3802 60.1779

.8893

F-Prob.

.4114

Research Hypothesis 12: There is no gender by grade level interaction
for the teachers concerning their measurement competency.

ANOVA results for research hypothesis 12 are displayed in Table

4.29b. It appears from Table 4.29b that there is no interaction between the

teacher's gender and the grade level at which the teacher practices

instruction insofar as measurement competency is concerned. This indicates

Table 4.29b: ANOVA for the joint effects of teacher's gender and grade level
with the competency in educational measurement.

Source Sum of
Squares
Gender 3936.422

Grade Level 227.500
Interaction 12.554
Residual 27871.008

Total 31924.086

D.F.

644

649

Mean

Squares

3936.422 90.957

113.750

6.277

43.278

49.190

F-Ratio

2.628

.145

F-Prob

.000
.073

.865

89

Table 4.29a: Information on the sample of teachers by sex and grade level.

Grade level
334 44% 420 56% 754 58%
Intermediate 76 49% 80 51% 156 12%

190 49% 200 51% 390 30%
600 46% 700 54%

 

that teacher's competency in educational measurement at the primary, the
intermediate, or the secondary level does not depend on whether that

teacher is male or female (see also Table 4.29a). Consequently, this

hypothesis is not rejected.

Multiple Regression Analysis

The multiple regression analysis (MR) that follows was conducted to
see which variables best predict Bahrain teachers' competency in
educational measurement. Table 4.30 displays the results of a stepwise
multiple regression analysis for the teachers. In this regression analysis, the
total score on the TEST was regressed on 10 variables; namely, number of
years in teaching, measurement coursework completed, statistics
coursework completed, gender, percentage of professional time devoted to
classroom testing, perceived level of competency in educational
measurement, perceived level of competency in test construction, major
subject of teaching/ supervision, preparing a test plan, and conducting item
analysis. The results revealed that there are five variables that appear to be
the best predictors of teachers' competency in educational measurement.
The five variables are gender, preparing a test plan, statistics coursework
completed, percentage of professional time devoted to classroom testing,

and number of years in teaching. These variables together account for 20

 

90

percent of the total variance in the TEST scores. It is interesting to discover
that 'measurement coursework completed' variable was not found among

the variables that best predict the competency in educational measurement!

Table 4.30: Multiple Regression Analysis for the TEST scores with
measurement preparation, and background variables for the teachers.

CORRELATIONS

STEP TEST 8 n ‘
(score) fl... (change)

_m-----mm--

_------mmmm-
___-W
___-“___“-
___—mmm-mmm

 

Teachers: 1 = Gender, 2 = Test Plan, 3 = Statistics coursework, 4 =
Professional time devoted to classroom testing, 5 = Number of years in
teaching.

Summary
This chapter included a review of the study's findings, questionnaire
results, TEST results. Pearson correlation results, ANOVA results, t-test
results, and multiple regression (MR) analysis results. The questionnaire
results reported in this chapter summarized the measurement preparation
and classroom testing activities. The TEST was reliable for the total sample
and for each group. Information from the TEST results indicate that the

subjects in each group independently have moderate competency in

 

 

91

educational measurement, except for the teachers who scored below
average on the TEST. By and large, the subjects appear to be most
deficient in the areas of 'types of tests' and "standard error of
measurement'. In general, correlational, and ANOVA results revealed
slight, but significant, relationships between competency in measurement
and several classroom testing practices variables, measurement
preparation variables, and some background variables of the subjects.
Multiple regression analysis results showed that the statistics coursework
variable was among the variables that best predicted competency in
educational measurement, whereas the measurement coursework variable
was not among those variables that best predicted competency in
educational measurement. Table 4.31 below shows a summary of the major

findings in the study.

 

92

Table 4.31: Summary of the major findings in the study.

HO # Stat. Test X(S) Group(s) Sig. Non-Sig. (.02
3 1-ANOVA 1 All X . 18
4 1-ANOVA 2 T X .07

CS x .24
PT X .09
5 l-ANOVA 3 T X .04
CS x . 1 6
PT x
6 1-ANOVA 4 T X .06
CS x
PT x
7 1-ANOVA 5 T X .01
CS x
PT x

8 2-ANOVA 6 All X

9 2-ANOVA 7 All X
1 0 t-test 8 T x
1 1 1-ANOVA 9 T X
12 2-ANOVA 10 T x

Note: HO= Null Hypothesis; X(S)= Independent variable(s).
Independent variables: 1: Type of respondent, 2=Measurement coursework, 3=Statistics
coursework, 4=Test plan, 5=Item analysis, 6=Gender by type of respondent, 7=Age by type

of respondent, 8=Gender, 9=Grade level, 10=Gender by grade level.
Sig=Significant, Non-Sig=Non-significant.
T= Teachers , CS: Curriculum Speciali sts , PT: Prospective

Teachers

.P‘r#_w- “...u- -v-.. .- -—--.--- —

34.1 W .918 one- is.-

 

 

CHAPTER V
SUMMARY, DISCUSSION, CONCLUSIONS, AND
SUGGESTIONS

Summary of Purposes and Procedures

The purpose of this study was twofold: first, to assess the
competency of Bahrain teachers, curriculum specialists, and prospective
teachers in the principles of educational measurement, and second, to
highlight the strengths and weaknesses among the three groups in specific
areas in educational measurement, such as item analysis, score
interpretation, reliability, and validity.

Two research instruments were used to accomplish the purposes of
the study, a measurement test and a questionnaire. The measurement test
consisted of 65 multiple-choice items. Fifty items of the measurement test
were taken from Boothroyd (1990), with slight modifications made to some
items to make their context familiar to Bahrain teachers, curriculum
specialists and prospective teachers. The remaining 15 items of the
measurement test were based on measurement criteria addressed in
measurement textbooks such as Mehrens and Lehmann (1991) and
Wiersma and Jurs (1990).

The questionnaire consisted of 23 items, some of which were based
on prior research findings (e.g., Boothroyd, 1990; Newman, 1981; 8 Mayo,

1967) that showed that these variables are valid and related in one way or

93

 

94

another to competency in educational measurement. The questionnaire
included items on classroom testing practices, measurement and statistics
preparation, practical needs in educational measurement, and background
variables.

The two research instruments were administered to 1300 in-service
teachers, 146 curriculum specialists, and 221 prospective teachers. The
in-service teachers were randomly sampled by strata (primary, intermediate,
and secondary grade levels). The 146 curriculum specialists and 221
prospective teachers are the populations of these two groups. The response
rate was 82 percent for the teachers, -67 percent for the curriculum

specialists, and 52 percent for the prospective teachers.

Discussion and Conclusions

By and large, the teachers, curriculum specialists, and prospective
teachers had taken a reasonable amount of measurement and statistics
coursework. In general, fewer statistics courses were completed compared
to measurement coursework. Most of the prospective teachers (80 percent)
reported that they completed one college course in measurement, but only
35.6 percent of the teachers, and less than one third of the curriculum
specialists (26.5 percent), said they took one college course in educational
measurement. The curriculum specialists had more multiple measurement
coursework than the other groups, with 36.7 percent of them reporting they
had more than one college course in educational measurement. Regarding
the statistics coursework completed, the picture is somewhat different. Less
than one- half (43.5 percent) of the prospective teachers, 34.7 percent of the

curriculum specialists, and 21.7 percent of the teachers reported that they

r-Mtgffil“ V's .-"~sm.l . al.-1 II

 

 

95

had one course in statistics. Furthermore, 28.6 percent of the curriculum
specialists, 18.3 percent of the prospective teachers, and 5.1 percent of the
teachers indicated that they had more than one course in statistics.
Moreover, in-service training in both measurement and statistics was sparse,
especially for statistics. Those who rated themselves as having “very good"
or “good" competency in educational measurement were between 21.7 to
42.2 percent across the three groups . In addition, 21 .5- 46.1 percent of the
respondents across the three groups indicated that they were “very good" or
“good” in test construction. Whereas 24.1 percent of the teachers and 24.5
percent of the curriculum specialists reported that they are “excellent" in test
construction, less than 5 percent of the prospective teachers felt this
confident. Except for the prospective teachers (for the statistics variable
only), those with measurement and statistics backgrounds scored
significantly higher on the TEST than those with no measurement or
statistics background at all, though the proportion of variance in the TEST
scores accounted for by measurement and statistics variables was small or
negligible (see Tables 4.21a, 4.21b, 4.222 and 4.22b).

Unfortunately, the amount of measurement and statistics
coursework completed have not translated into high performance levels for
the three groups on the TEST. The curriculum specialists and the
prospective teachers, in general, were more competent in measurement
than the teachers. The teachers had a mean score of 26.72 (41 percent),
the curriculum specialists had a mean score of 37.49 (58 percent), and the
prospective teachers had a mean score of 34.07 percent (52 percent). The
scores on the TEST ranged from 9 to 48 for the teachers, from 19 to 50 for
the curriculum specialists, and from 18 to 49 for the prospective teachers.

The teachers had the lowest average score among the three groups (see

 

 

96

Tables 4.20a and 4.20b). The teachers performed better in the areas of
“grading and marking", and “objectives" than they did in all the other areas
that comprised the TEST. The teachers were weak in many areas such as
'item analysis“, “standard error of measurement" and “score interpretation".
The curriculum specialists performed better in the areas of “objectives",
“grading and marking", and “validity" than they did in some other areas such
as “test planning", “types of items", “item writing", “test construction", “item
analysis", and “score interpretation.” They were weak in the areas of
“standard error of measurement", and “types of tests." The curriculum
specialists outperformed the teachers and prospective teachers in all areas
of the TEST, except for the areas of 'types of items' and 'reliability' (see
Table 4.13). Concerning the prospective teachers, they scored better in
the areas of “objectives”, “types of items”, “item writing", “grading and
marking", “correlation", “reliability", and Validity" than they did in some other
areas such as “test construction", “item analysis", “score interpretation", and
“standard error of measurement.”

Generally speaking, even though most of the respondents,
especially the teachers and prospective teachers, reported that they conduct
item analysis on their self-constructed classroom tests, this did not translate
into high performance in the area of 'item analysis“ (see Table 4.13).

The three groups' weakness in some areas, such as 'item
analysis', may be reflected in their stated needs for practical training in
educational measurement. The results indicated that the teachers may need
more practical training in item analysis, test planning, and reliability. The
curriculum specialists and prospective teachers may need more practical

training in item analysis and score interpretation.

 

 

97

The female teachers scored significantly higher on the TEST than
their male counterparts which indicates that female teachers are more
competent in educational measurement than male teachers. Except for the
teachers, no significant correlations were found between the TEST scores
and “classroom testing' variables, and 'measurement preparation“ variables.
Only for the teachers was there a significant correlation between the TEST
scores and such variables as the purposes of using classroom tests,
developing a test plan, conducting item analysis, and self-assessment of the
general competency in educational measurement. Even so, the proportion of
variance in the TEST scores accounted for by each of these variables was
small or negligible.

It was found, for the teachers only, that those who stated that they
conducted item analysis and prepared a test plan scored significantly higher
on the TEST than those who did not conduct item analysis or prepare a test
plan. There was no significant difference among teachers in measurement
competency from one grade level to another.

The findings of this study revealed that there was an interaction
between the type of respondent and respondent's gender with respect to
measurement competency. The age of respondent, however, has no bearing
on measurement competency from one group to another. This applies to all
groups. For the teachers only, the results showed that teacher's gender has
no bearing on measurement competency in the primary, intermediate, and
secondary grade levels. The results of a stepwise multiple regression
analysis revealed that there were five variables (gender, preparing a test
plan, statistics coursework completed, amount of professional time devoted

to classroom testing, and number of years in teaching) that best predict

    

..! _.., _ r. _ _ I —' . . ”in?“ - iii.

 

98

teachers' competency in educational measurement. Interestingly, the
amount of measurement coursework taken was not among them!

In this study, the curriculum specialists had more familiarity with
educational measurement terms and principles than the teachers and
prospective teachers. Although the teachers with measurement training
scored significantly higher on the TEST than those teachers with no
measurement training at all, the entire group of teachers demonstrated
below average performance on the TEST.

It can be argued that in-sen/ice training in educational
measurement for the teachers and curriculum specialists is minimal
compared to the measurement coursework completed in a college of
educafion.

The teachers, curriculum specialists, and prospective teachers
tended to give themselves high ratings concerning self-assessed
competency in educational measurement, test planning, and item analysis.
These ratings, however, did not match their actual performance on the TEST
(see Table 4.13).

Based on the findings of this study, the following conclusions can
bedrawn:

1. The teachers in this study were the least competent group in
educational measurement . They need more measurement training in
general, and specific training in the areas where they were weak.

2. The curriculum specialists showed that they were more
competent in educational measurement than the teachers and prospective
teachers, and that they indeed had a good grasp of some measurement

terms in some areas (see Table 4.13).

 

 

99

3. The self-rating on some variables such as preparing a test plan,
conducting item analysis, and self-assessment of the general competency in
educational measurement, especially for the teachers and prospective
teachers, did not translate into good performance on the TEST.

4. Female teachers were more competent in educational
measurement than male teachers.

5. The competency in educational measurement, however, did not
depend on the grade level.

6. Competency in educational measurement among the three groups
depended on whether the respondent in that group is male or female; but the
age of respondent had no bearing on the competency of that respondent

from one group to another.

7. The 'amount of measurement coursework completed' was related
to measurement competency. This variable, however, was not found among
those variables that best predicted competency in educational

measurement.

RECOMMENDATION

The Bahrain teachers, curriculum specialists, and the prospective
teachers need more training in educational measurement in general, and in
the areas where they manifested around or below average performance in
particular. A measurement training "liaison officer" for each public school
district may assist with measurement training in schools, and may facilitate
coordination and direct contact between the school(s) and the measurement

specialists in the Bahrain Ministry of Education. The amount and type of

.,~, ”BF-11

M") W at- wir-

 

 

100

measurement training to be provided to the prospective teachers should be

based on further research.

SUGGESTIONS FOR FURTHER RESEARCH

Reviewing the previously conducted research studies on teachers'
competency in educational measurement, starting with Mayo's (1967) study
and ending with the most recent studies in the field (e.g., Piake, lmpara, &
Fager, 1992), reveals some shortcomings in how this issue was approached.
First, many studies used almost the same type of instruments to assess
competency of teachers in educational measurement. Second, some of the
questionnaires used in some research studies to assess the perceived
competency of teachers and other educators in educational measurement
did not appear to be revealing the actual level of measurement competency
of these groups, especially when the performance on the measurement
competency test was not comparable with the perceived measurement
competency. Third, those studies inadvertently ignored other factors that
may have some impact on, for instance, teachers‘ competency in
educational measurement. These factors include how educational
measurement was taught to teachers, the way they learned it either in a
college of education or through in-service training, how their learning in
educational measurement was evaluated, whether they practiced in their
schools what they learned in the colleges of education, and whether the
colleges of education evaluated the impact their measurement courses may
have had on the teachers in schools.

Based on the findings of this study, and on the shortcomings in this

study and in the research studies conducted earlier on teachers' and other

 

101

educators‘ competency in educational measurement, the following
suggestions warrant:

1. A combination of more than one research instrument or more than
one research approach in studying the competency of teachers and other
groups in educational measurement may enable researchers to detect some
factors that may affect competency in educational measurement. For
example, one might use, in addition to the 'regular' measurement
competency test, a measurement test that asks for direct performance on
some measurement topics that are very related to classroom assessment,
such as item analysis, objectives, and test construction.

2. Some research methods such as observations and structured
interviews with professors of educational measurement and prospective
teachers may provide valuable data regarding competency in educational
measurement. For example, professors of educational measurement may
be observed teaching the entire measurement course. Structured interviews
with professors of educational measurement are essential in order to explore
how these professors view students' learning in educational measurement,
the major strengths and weaknesses in understanding and applying
measurement principles, the major factors these professors found as
impediments to students' practical application of some measurement
principles, and what factors among these still persist. Furthermore,
interviewing a representative sample of students who took a measurement
course(s) may shed light on some factors that impede improving
measurement competency. These factors may include how students were
introduced to educational measurement courses, what measurement

principles they applied, what topics they found to be of direct link to their

rwn't +1!" in teem-emf .131. -.~- HIT.) w-

 

 

102

daily classroom assessment, and what made them eliminate application of
some measurement principles in their classrooms (e.g., item analysis).

3. A carefully designed continuing evaluation of the quality of
measurement courses offered in teacher-training programs and through in-

service training may be needed.

REFLECTIONS

The following are some reflections on some findings in the study and
their culturaV social implications.
1. Although there were significant gender differences in measurement
competency, there were no significant grade level differences. The
implication here is that there is a culturaV social assumption/ belief that
female teachers are more committed to their learning and professional
development, and more organized than their male counterparts. Finding
non-significant differences among the teachers at different grade levels
seems contradictory with the culturaV social assumption/ belief that those
teachers who teach at the higher grade levels (e.g., secondary) are more
knowledgeable, in general, than those teachers who teach at the lower
grade levels (e.g., primary). There are two implications here. First, it is not
necessarily true that the teachers at the higher grade levels are more
knowledgeable in their subject matter than those teachers at the lower grade
levels. Second, teachers may differ in their knowledge of the subject matter
because they are teaching at different grade levels (e.g., a primary teacher
does not need as broad an understanding in mathematics, for example, as

does a secondary teacher) rather than because of the amount of

measurement coursework taken.

-.1

.- .r- .- --.- .

 

l I
l

103

2. When one looks at the distribution of performance on the various content
areas of the TEST (Table 4.13) it is readily evident that if the last four areas
of the TEST (correlation, reliability, standard error, and validity) were
dropped, the TEST sores of the teachers would increase and be comparable
to those of the prospective teachers. This might be attributable to the fact
that because these four areas are more technical, they are less familiar to
the practicing teachers than to the 'newer' prospective teachers. In addition,
the difference might be due to better training today in these areas.
Furthermore, since knowledge of statistics appeared to be a good predictor
of measurement competency, this might account for the difference.

3. The difference in the self-ratings between the teachers and prospective
teachers (see Tables 4.9 and 4.10) can be looked at from different angles.
The large sample size of the teachers may have a bearing on that
difference. Second, the teachers may be more honest in their ratings than
the prospective teachers, because the teachers feel more responsible than
the prospective teachers and they take the issue more seriously. Third, the
prosepctive teachers, however, may have thought that their knowledge in
measurement is more recent, and hence that they are competent. Finally,
teachers' experience with test construction and classroom tests may have
and influence on their self-ratings when compared with the prospective

teachers who almost lack that experience.

.f'.-,o_ .7
, 42- _
v

3*". _l‘__ , ' "a“ _fi'miaMJ-w‘nmi
m Mill“ Etna m menu New"... -..=.-v-vh ---.-.\t 18!? nit-y ,
‘H-L'! t. -u.'..' -_.' . ".'.. .;.-

 

 

 

 

APPENDIX A

 

 

.In t.

TEST OF EDUCATIONAL MEASUREMENT

DIREQTIQNS

1.This test consists of 65 questions.

2.ln the space provided to the left of each item. please print the letter of the
option which you believe BEST answers the question.

3.Please print only DEE letter for each question.

4.You have SHQQBSto complete the test (you may take extra time if you need).
5.Please answer all questions.

i‘tfififitfifi.i.‘.i.'iifififinttiitititfiitfiifii

_1. TABLES QF SPEQIFIQATIQNS are important because they help insure
the tests'

a.content validity.
b.criterion-related validity.
c.internal consistency.
d.test-retest reliability.

___2.Essay items are more appropriate than multiple-choice items for assessing
a.application of knowledge.
b.factual knowledge.
c.recognition of ideas.
d.organization of ideas.

__3.A QISADVANTASE of multiple choice tests as compared to essay tests is
that they are
a.generally less valid.
b.scored more subjectively.
c.more time consuming to construct.
d.appropriate only for narrow content domains.

__4.A test plan (Table of Specifications) is LEAST useful in
a.judging the relative emphasis to be given to different content areas.
b.judging the relative emphasis to be given to different objectives.
c.deciding the type of items to use on the test.
d.relating the item content to performance (behavioral) objectives.

104

 

 

_5.The type of evidence with which a teacher should be most concerned when
constructing classroom tests is
a.construct validity.
b.content validity.
c.split-half reliability.
d.test-retest reliability.

_6.'chance' success in matching'items can best be reduced by
a.using statements that require fine discriminations.
b.not allowing responses to be used more than once.
c.lengthening the statements and the responses.
d.including more responses than statements.

_7.Which one of the following is N_Ol considered a sound principle in
constructing true-false questions?
a.Avoiding the use of negative statements as much as possible.
b.Confining each question to a single idea.
c.Using more true than false statements.
d.Refraining from using long and complex statements.

_8.|nstructional objectives are MQSjLuseful when they
a.are equally appropriate for all students in the class.
b.are specified in terms of the knowledge they involve.
c.relate skills and knowledge to performance (behavioral) objectives.
d.can be used as the basis for developing objective tests.

_9.For a test to be valid it must also be
a.practical.
b.reliable.
c.important.
d.objective.

_10.A criterion-referenced interpretation is MQBE appropriate than a norm-
referenced interpretation for decisions concerning
a.classroom strengths and weaknesses.
b.determination of mastery.
c.placement in a gifted program.
d.prediction of student performance.

105

 

_11.Which one of the following item types is BESI suited to measure a
student's ability to apply information?
a.Matching items.
b.Multiple-choice items.
c.True-false items.
d.Both b and c.

_12.'Manama is the most important city in Bahrain" is a poor true-false item
because it is
a.ambiguous.
b.too easy.
c.too factual.
d.trivial.

_13.Which one of the following item types is LEAST objective?
a.Matching items.
b.Completion items.
c.Multiple-choice items.
d.True-false items.

_14.Since no test is perfectly reliable, a student's score should be regarded
as
a.having little meaning unless it is very high or very low.
b.having little validity in estimating the student‘s achievement.
c.indicating a point in a range of scores in which the student's 'true
score' probably falls.
d.indicating only that the student has more or less ability than the
average student.

_15.|n the scoring of an historical essay test, all of the following are generally
considered desirable EXQEEE
a.preparing a scoring key and standards in advance.
b.deducing points for grammatical errors and poor handwriting.
c.removing the students' names from the test.
d.scoring one question on each test prior to going on to the next.

106

 

_16.The practice of allowing students a choice in the questions to be
answered on an essay exam is generally considered to be
a.undesirable, because students waste too much time deciding which

questions to answer.
b.undesirable, because it reduces the comparability of the test from

student to student.
c.desirable, because it gives each student a fairer time.
d.desirable, because it permits a wider sampling of topics covered.

_17.Arranging test items by item type tends to
a.make test taking easier for the student.
b.reduce the standard error of measurement.

c.increase item validity.
d.increase the test's reliability.

__18.A norm-referenced interpretation compares each examinee's score to
a.a performance standard.
b.other examinees' performance.
c.the true score of the individual.
d.the mean test score.

__19.Layla attained a score of 80 on her criterion-referenced geography test.
Which of the following additional piece of information is MQST useful
for interpreting her performance?
a.The minimum passing score was 75.
b.The class average score was 65.
c.The test had 90 questions.
d.Two thirds of the class failed the test.

_20. If an item is correctly answered by 38% of the students in the top third
of the class, and 66% of the students in the bottom third, the item's
discriminating power is
a.negative.
b.unstable.
c.valid.
d.almost perfect.

107

 

_21.The correlation coefficient between two sets of scores from the same test.
given to the same class is .95 . This indicates that the test is highly
a.discriminating.
b.valid.
c.reliable.
d.unstable.

_22.Which correlation coefficient indicates the highest predictability of one
variable from the other?
a. -.91
b. -.35
c. +.54
d. +.76

__23.The standard error of measurement is a useful index of the
a.amount of variability in the score distribution of all persons tested.
b.proportion of persons answering the item incorrectly.
c.degree of variability of a single person's observed score on multiple
testings.
d.test's validity.

_24.Rashid knows absolutely nothing about the material in a 40 item, four-
option, multiple-choice test. By random guessing he should answer how
many items correctly?

a.0
b.4
c.10
d.20

_25.For which of the following learning outcomes would a multiple—choice
test be LEAST relevant?
a.Defines a basic concept.
b.ldentifies the reason for an action.
c.Fielates an example to a stated principle.
d.Selects the best method to use.

108

 

_26.A sixth grade class was given two different forms of the same science
test. The test scores from the two forms placed students in nearly the
same order. This would be evidence concerning the test's
a.internal consistency reliability.
b.equivalence reliability.
c.scorer reliability.
d.stability reliability.

_27.A test is reliable when it
a.can be scored quite easily.
b.gives a consistent estimate of whatever it measures.

c.measures what it is designed to measure.
d.provides a basis for diagnosing pupil's weakness.

_28.Which one of the following is a characteristic of a highly reliable test?
a.Good students score higher than poor students.
b.The items in the test vary widely in difficulty.
c.Two scores for the same examinee agree with each other.
d.There is a uniform distribution of scores on the test.

_29.lf the scores on a test are 5, 2, 3, 27, and 8, the test's mean score is
a.5
b.7
c.8
d.9

_30.An item with a difficulty level of 85% indicates that
a.the item has a good discrimination.
b.the item is easy.
c.some students answered the item correctly.
d.a large proportion of students who did poorly on the test got the item
wrong.

___31.An item's discrimination will likely be maximized when its difficulty index
approaches
a. 0%
b. 25%
c. 50%
d. 100%

109

 

_32.Which one of the following is the MQST serious problem with the true-
false item: “Manama is the capital city of Bahrain, a major banking
center, and an important port'.
a.Tests multiple concepts.
b.Too many specific determinants.
c.Too easy.
d.Contains grammatical errors.

_33.|n terms of Bloom's taxonomy, application reflects a higher cognitive
function than does
a.comprehension.
b.analysis.
c.evaluation.
d.synthesis.

_34.'The student will demonstrate comprehension of the short story form“ is a
poor instructional objective because it
a.is too unrealistic.
b.describes the instructional process.
c.does not specify the conditions under which the performance is
expected.
d.does not specify an observable performance.

_35.Which one of the following item types would be MQSI appropriate for
determining students‘ ability to compose a short story?
a.Essay items.
b.Completion items.
c.Multiple-choice items.
d.Short-answer items.

_36.lf Ahmad had a test score of 78 on a test with an estimated standard error
of 6, we would be 95% confident that his 'true score' would fall between
a. 66 and 90.
b. 69 and 87.
c. 72 and 84.
d. 75 and 81.

110

 

_37.A percentile rank refers to the percentage of
a.items correctly answered.
b.items needing to be correctly answered to obtain a passing score.
c.students scoring above the mean on the test.
d.students scoring below a given raw score.

_38.An advantage short-answer items have over multiple-choice items is that

they

a.are more reliable.

b.can be used on either norm-referenced or criterion-referenced tests.
c.have more desirable discrimination levels.

d.minimize random guessing.

_39.Which one of the following statements MQSI suggests a nonn-

referenced interpretation?
a.Noora scored higher than the class average score on the science test.

b.Mariam correctly answered 80% of the items on the mathematics test.
c.Hassan passed the minimum competency reading test.
d.Ali exceeded the mastery level on the 5th grade social studies.

__40.A history teacher uses the following item: 'List three reasons why the Gulf
war started". To classify this question on Bloom's taxonomy we would
need to know
a.what the students received for Gulf war lectures and readings.
b.the general intellectual level of the students in the class.
c.what the teacher keyed as acceptable answers to the questions.
d.whether the test is norm-referenced or criterion-referenced.

_41.A correlation coefficient indicates the
a.extent of variation within a set of test scores.
b.extent to which two variables measure the same thing.
c.degree to which one variable causes the other.
d.degree of relationship between two variables.

111

 

‘1
D‘

n.-

 

42.

 

The relationship indicated by the above scatter plot is
a.highly negative.

b.moderately negative.

c.moderately positive.

d.highly positive.

__43.Which one of the following is appropriate for assessing a test's internal
consistency?
a.Equivalent forms.
b.Kuder-Richardson
c.Coefficient of stability.
d.Test-retest.

_44.A DISADVAN [AGE of a norm-referenced grading scheme is that
a.percentage grades cannot be calculated.
b.an individual's class ranking cannot be determined.
c.grading is presented in relative terms.
d.performance is assessed in terms of a fixed criterion.

112

 

 

_45.Which one of the following statements B_E_S_T_ illustrates a criterion-
referenced interpretation?
a.Nada earned the highest score in the class.
b.Amal set up her experiment in five minutes.
c.Khalid's test score was average for his class.
d.Ebrahim solved the fewest arithmetic problems.

_46.An important consideration when constructing classroom tests is to
a.construct items which require students to make fine distinctions.
b.develop two parallel forms of the test.
c.match the items to instructional objectives.
d.test materials not taught in class so that students have an opportunity

to demonstrate their out-of-class learning.

___47.Which one of the following item statistics would be of MQST serious
concern to the teacher?
a.D'rfficulty index: .80
b.Difficulty index: .05
c.Discrimination index: .00
d.Discrimination index: -.05

113

g -*.."'"'. I‘M r‘ f: «fin-u"?! mrl nu ‘
,-.'-- _... .. __.. _. _._-:—__ . gag-r- «.1! 1"

;i": :-
l'."-- ' " '

      

h.

 

ul
U

 

48.

    

0 IO 20 30 40 506070 30 90100
,usyr v‘ 95ml cake);

The graph above indicates that the test is
a.too easy.

b.too difficult.

c.moderately difficult.

d.moderately easy.

__49.0bjective scoring of selected-response items enhances the test's
a.reliability.
b.validity.
c.validity but not stability.
d.reliability and validity.

_50.lf the reliability of a test is 1.00, the standard error of measurement is
a. .00

b. .10
c. .50
d. 1.00

_51.The point below which 50% of the scores lie is known as the
a.median.
b.mean.
c.mode.
d.standard deviation.

114

 

 

_52.Which one of the following is a measure of variability of scores?
a.Mean.
b.Median.
c.Mode.
d.Standard deviation.

_53.According to Bloom's taxonomy, at what level can the following statement
be classified. 'The teacher said to one of his students: Your composition

passage is of poor quality'?
a.Application.

b.Analysis.

c.Synthesis.

d.Evaluation.

_54.From a measurement viewdpoint, using classroom tests consisting
entirely of essay items is undesirable because
a.scoring requires too much time.
b.it is difficult to construct the items.
c.content sampling tends to be limited.
d.constructing model responses is too time consuming.

_55.Which one of the following is NQT generally associated with criterion-

referenced testing?

a.A performance standard.

b.A well specified content domain.

c.Test items selected according to difficulty level.

d.The use of test items which closely reflects the instructional objectives.

_56.A poorly defined domain results in items that are
a.too difficult.
b.dissimilar.
c.ambiguous to the student.
d.unreliable.

_57.The term objective , when used to label educational tests, MQST
appropriately describes
a.a characteristic of the scoring procedure.
b.the degree of the standardization of the test.
c.the content limitation of the question.
d.whether the test interpretation is norm-referenced or criterion-

referenced.

115

 

 

__58.Which one of the following would indicate the highest level of learning
for a student?
The student
a.applies a principle.
b.gives a precise definition of the principle.
c.identifies an example using the principle.
d.rephrases a principle in his/her own words.

_59.Which one of the following educational goals is liOl stated in
performance (behavioral) terms?
a.Design.
b.Prepare.
c.Understand.
d.Write.

_60.A major strength of criterion-referenced testing is that one can
a.generalize the results to a larger set of items.
b.compare students in terms of relative standing.
c.establish objective performance criteria.
d.measure a specific trait.

_61.Problems arise when attempting to develop measures of ultimate goals

mainly because

a.group norms with which to compare the results are generally not
available.

b.measurement methods have not given proper weight to all goals.

c.such goals concern behaviors usually not observable under classroom
conditions.

d.teachers are reluctant to depart from traditional testing methods.

_62.Which one of the following educational outcomes is classified in the
affective domain?
a.Developing handwriting skills.
b.Fostering an interest in music.
c.lmproving reading comprehension.
d.Understanding the classification of interests.

116

  
 

       

 

 

_63.What is the major problem with the following completion item in a seventh
grade history test: 'The Portuguese invaded Bahrain in "?
a.Too easy.
b.Unclear.
c.Has a specific determiner.
d.measures factual knowledge

_64.When constructing multiple-choice tests, it is best to
a.make all options the same length.
b.put the main idea of the item in the options.

c.use options such as a and b , but not 5:.
d.repeat key words from the stem in the options.

_65.Completion and short answer items should have the blank(s) placed

a.at or near the beginning of the item.
b.between the beginning and the middle of the item

c.as close to the middle of the item as possible.
d.at or near the end of the item.

'.......".'I..".m'."."i.'t '.... '....

THANK YOU VERY MUCH FOR YOUR COOPERATION

117

 

 

THE QUESTIONNAIRE

W

1.This questionnaire consists of 23 questions.

2.Please use the space(s) provided to print your response(s).

3.Please note that some questions require that you select only ONE number
from among the numbers that correspond to the options provided, and print
that number in the space provided.

4.Please note that in some questions you can select more than one option.

5.Please answer all questions.

..iiitfifiiiilfifiim'iittifiIt...

1._ Age?

2._ Gender?
[1] Male
[2] Female

3._ Education level (the highest degree you hold). Year obtained
[1] Ph.D.
[2] Masters
[3] High Diploma (beyond bachelors degree)
[4] Bachelors
[5] Associate Degree (Diploma) (two years beyond high school)
[6] High School Certificate
[7] Less than High School Certificate

4. is the highest degree you indicated in Question 3 above obtained from a

college of education? _ Yes. _ No.
If your answer is NQ, please specify the highest degree you obtained from a

college of education.

5. Was EDLJQA [ [QN your MAJQB when you were in your senior year in the

undergraduate program? _ Yes. No. _.
6. Major subject of specialization?
7. Subject you teach/ supervise.

118

 

 

8. _ Stage at which you teach/ supervise?
[1] Primary
[2] lntennediate
[3] Secondary

9. Number of years of experience in teaching? (for in-service teachers and
curriculum specialists only).

10. Number of years of experience in curriculum supervision? (for
curriculum specialists only).

11. How much coursework in W have you completed?
(check as many as apply).
_ None
_ Part of another course.
_ One college course.
_ More than one college course.
__ As in-service course/ training (as offered by the Ministry of Education).

12. How much coursework in statistiss have you completed? (check as many as
aPPIYI-
_ None.
_ Part of another course.
_ One college course.

_ More than one college course.
__ as in-service course/ training (as offered by the Ministry of Education).

13. _ In what year did you complete your mgsttecsm coursework in
educational measurement?

14. _ in what year did you complete your mostteceutcoursework in statistics?

15. Approximately what percentage of your professional time (both in and out of
school) do you devote to constructing, administering, scoring, grading,
interpreting, and reviewing your self-constructed classroom tests? (check
me only).

_ Less than 5%.
_ 5% to 10%.

_ 11% to 20%.
_ More than 20%.

119

 

 

16. For what purposes do you use your self-constructed classroom tests?
(check as many as apply).

_ Diagnosing strengths and weaknesses.

_ Assessing academic achievement.

_ Assessing mastery of instructional units.

__ Assessing academic progress.

_ Motivating learning.

_ Assigning grades.

_ Planning instruction.

_ Other.

17. What type of items do you write for your self-constructed classroom tests?
(check as many as apply).
_ True-False.
__ Matching.
_ Multiple-Choice.
_ Completion (fill-in-the-blanks).
_ Short Answer (one or two paragraphs).
_ Essay (one or two pages).
_ Other (specify).

18. Rank the following item types in terms of their preference to you (1=Most
preferred, 6=Least preferred).
__ True-False.
_ Matching.
_ Multiple-Choice.
_ Completion (fill-in-the-blanks).
_ Short Answer (one or two paragraphs).
_ Essay (one or two pages).

19. _ Do you generally develop a tsstptan (e.g., table of specifications) prior
to writing test items?
[1] Yes.
[2] No.

20. _ Do you generally conduct an jtsm_ana[ysjs on your self-constructed

classroom tests?
[1] Yes.
[2] No.

120

         
        
 

c..'-.". “finwhwmuqflwm :-
m M much?“ I
'...-J": a:{ol.'.f_'. a ..'- _ 3"" :I-;;."| .'n , I _ _ -,- In

. u ' '4 " (“HEN-fl wall}:

1 T I. . .' I ' ...‘I-‘I'. ‘-

 
   

21. _ How would you assess your competency in educational measurement in
general?
[1] Excellent.
[2] Very good.
[3] Good.
[4] Adequate.
[5] Poor.

22. _ How would you assess your competency in test construction principles?
[1] Excellent.
[2] Very good.
[3] Good.
[4] Adequate.
[5] Poor.

23. In which of the following topics in educational measurement you feel that you

need Qtastjsatttajmng ? (choose as many as apply).

_ Planning classroom tests (e.g., table of specifications).
_ Constructing different test item types (e.g., true-false).
_ Knowing what test item type to choose and why for constructing your

classroom tests.

_ Conducting item analysis on your classroom tests (e.g., item difficulty,
item discrimination).

_ Test validity (e.g., content validity).

_ Test reliability (e.g., internal consistency).

__ interpreting classroom test scores.

__ Observational methods.

_ Other (please specify).

ifi'ifiitiiiiiififififiiflfiiiii‘

THANK YOU VERY MUCH FOR YOUR COOPERATION

121

 

 

 

 

 

APPENDIX B

 

 

APPENDIX B

THE TABLE OF SPECIFICATIONS OF THE TEST

Content Item Level Item Total

Area Knowledge Knowledge Explanation Application
of Terms of facts & of

Interpretati Numerical

Problems

Test

Planning

Objectives 8, 33, 53, 59 34,40,58,61,
62

Types of 57 18, 56, 60, 10

Tests 55

Types of 2, 11, 13, 25. 3, 35, 54

Items 38

Test 16, 17, 46 47 4
Construction

Item 51, 52 31 30 20 5
Analysis

Score 37, 44, 49 14, 19, 39,

Interpret- 45, 48

anon

Grading & 15 24, 29 3
Marking

Standard 36 50
E_rror

 

122

    

  
 

. .c'ru’l meat q

_ -- ... __

 

 

 

APPENDIX C

 

 

THE TEST ANSWER KEY

ECECDAAAADDCCBAACACEE

3 .EurC-InCC.O4l-4.JA_C.:_l:~3.0.l2nd
4 44444EVEvE~2v£~C~Eu£uE~C~€666

A

D
‘0. B
40. A

22. A
23. C
24. C
25. A
26. B
27. B
28. C
29.

31. C
32. A
33. A
34. D
35. A
6. A
37. D
38. D
39.

41 D
42. D

a
a
v

ADCCBDCCBBDABCBBABAAC

O.1..234567n0.0~nv.4l.
1234567aov1111111111a42

123

 

 

 

 

 

APPENDIX D

 

 

 

‘...,” 0.,11 1111‘...,

1119-95“ 9.11.111qu .143!
OIJL—_.:J‘l

.'~11_-,..-1o,,.,1,.-i~1111.;,,;.-._-1
Jiiaa631,1,i1,.;,_u1.asu.cesi11,i,jrogyuu.di {1,119.

.J...‘\11,1§...11._,J1,...11,...,.:

.1... .i,1, 1.1.1,,1—1"

.ALYIbiosGUJaiwwI — 1.

1911.21 .23, 1.1.4.151 ..1i_>_11 31.1.. 331: 1.Ja_..LI — o

01..., d1. .U—LJ 14.-'1 11'... cute-.311 3114151321., . 1
65.-Jig»

. 1

(4:9) ,1,.._1.,:,11,;i..11 «.2

.uu1i111511,.11)51...-j11 u,

.JLeéyI :JLGL, beI 53L; .3

0.141 .m,,1,:i11111i.10. 1.1.. ,si i',1u1121;..‘\11 . r
.Lifilgflias . I

. .JUJ-‘I 13,-; . g

.,1§.s~11a1...1,..-11. ..

.JK‘YI r35: . 3

1.31,. 411111.31 1,511. 11,11. .i..-.. o. ,1,.-,~11.;.1,1_.;-,1 1.4).:- .L>I . ij
.uJLLEis “J51 . 1

4.,5.I3_,:5IJS...1 M ‘41 . 1,:
.11..1i.111.:1.i.c.s,uts>1...1,si .i
.,;.‘,...11d,;,.11.;.15 1.1,.11).:.1.,.;.,.11 1,..1... ..

g. zisu Lui 131a..1,11,1,i,1,1,.',j11i1.s . 1._

.111...“ 6,...11 r“? L,1... 6311,4111...ku r5" . 1
- in.» dim-1,1... $119.11 r1....\11,_,1i,.$i1 . ._.
131.11.511.11).1.‘11..11i.1,11,..11,,,.~ 1m, . 1

124

 

 

_7”

11:.11.1.1.1).pufi1w1,,i11.r..,1.~,nly.11ty .

_,A 94.4“ JL,»\JI

.(erI 9.1.4) “It,“ 9.1.4.” .I
-15.~=’~“134~’ -'-.—'
4.,1.43I4JJ"1.3I QL,.‘ . .7
.JL,»‘J| iaLaI, ,L,»‘1I 0L1: . a

@IyMYIJSJQdJBUSg (QJIJLI)ZL1U:1I4.L:..IQJ«UA.4II» CLf1 .
‘ _‘1I,_.,£..LI:.-Jol,L:.rI.1$1...l

3:,» 0“ ,5I 411.119.11.41. CL...JIr .1; .

.oLng‘lI, 2.1:...“ OLA» EJLLI .
.1135” OIJLrn ..1.-. L,a_,.‘SI1:JLaI.?‘JI 9.1:. J.»

L-II ‘C '...-

1‘11111, ._.1,..1111..i 1.1.1.11 1.9.141. 1i..,,..'.. 19,511,. i.1, g1 .
dbl: '.vg'g—‘JU-‘J‘x" -

11.11 .-.1,1,. U.,.1’i1_.1,..11.~.1,1,. J... .

.041», 1L1} ObLJ: . rI.L§1.:...I 0:- Lb?! .

v "V 'C '...<

1.11.1111; ,511‘,..1..11.11.1.'~11;,,:.- .
.LAmJIsrigikJqu-JMMQDQS
1,...w...11....11..»..,,:..
..1.\11.11i.'1.1.',.11, .-.1,1,111..,: .
19y.» able-fl ...th 1.1-L515 1..1.1;...1 1:1“.

afloII-zeiwbaLvJLe—"Waéfl -
.14: .

I. .11 'C :_.>

u N '( .—..)

..1,1,111....11.¢.,119.1.1.1,..1.11,...1..,51¢.,11 ,1_..11,,....11 . .
al.-.11" I

..iAIIvbMI, iflIQIgu . I

.QLUYIdHJJJ’LD .9

wry“ 6.1., .1 1411111131111 . .
...111.11.1.1._,...11. . 1

125

 

 

 

 

11.1.11}. 11111111111 u11,11 1.31. M¥|}¢LVLNIEI;IO. 11,151.11

1.211111

117,51111111111 .1

-'-‘vmouL=S—"?I - 1—2

31111.1J 1.1,.” ..

{weary- >

J15.111..,1.1,..t,-0.1J15.11 111.1(134191H'x111111118fl11111. .111
13311.11,

.9411; .i

11.1.. v

IIAag-‘IfJIJ ..g

.4.iL7 . .1
sggyflflqmwmtwy1bgi .11:
.1mb11111u111 .1

«IL-5?" «.4

.JMJO‘ngI ..g

.1111, 1,11,..11 .1

L131 0"" 1191.111 .115 311.111.11.511112 cw 1.1-11.11,. 11 L1 1111
1.1111151111141101? bywols

. ...JILIIMJJBQSJQQuoIS .
1.11111111.131.311.111-01141.012.111.516111J1fiz .
1.411111110111111 ,m :11; 11.11 1.11111 1,1 L111.111,..1: .

.\L

o L '( '....

rumtyxfig‘b. 1:215:19 .eriqsymwem-MQC
“___.

.51.» 915145655431ch #Lau 3.5!

1.1.1 1.11,, 14.111.11.181... c111). ‘...; .
.J1,.--.;1151,,1C,..,.>11.11.1..121151 .

.11 111.. c... .11 1111131115 11.511.51.111 1%. 1.1, .111. c—.-' .

126

vob'c;

 

‘1-
1 .

 

 

cu 511.1, 3...; 911111 01.11311 1.1-...1 q.- ,1..1j11._ 9311.11 C1._..11:-,1.11
.1; ‘...-geo'v-thD-«g‘ afiadfl‘wnfi’bm vwlflvfixfl
v“ J! 141111.113?” 'a,121_1;;1.u01,1lz.fi'11 113.95.111.11}- -
. nfil‘bjgwyegu‘lgfif .
.u;.1,.¢29111ou,1,110.£5111,12.11;.c_..5~19,1,.

1, .11 '...

.11 11.1. 111.11? 1» 1111.111 was .w
1111.11 21.-:11. 111.111.11.11 1.1 1.1.. .11?

.v.1_.z11 ¢,1__.1111.11J.1z.- .

.J1_;.J1 5.1.4:;19' .

.,1_..1}11 1:11.: 1.1.,

1,111 ‘...

C“ ‘.Ju’ Jib); {uni C?!“ 944" .n-J‘“ - ‘5.
.4530 (,Lgu) 6,2...- . I

.vj‘llyWI-Iai . 5.11

«Aw 14.411 22.111 ..1

-JL.-="?' Luv-113:9“ 5

91.,1111o.;,1.3,11._,,1,11111_.s1,111,1,;11g111. 25101191311111.
L‘Jlal ,...1: 913.1%".5—1144‘” 1,1511 4.51.13le

1111;115:1111 u1911.11.13”; .1

.‘10 DUMwalbé 113.92.. . y

11111313111211.1111.

.,1,.:3le93 1,,.., 11.111, 1,511.11 1:1: . 1

x11 91,1 1.11.11, 11.1.1111. “111.115.1111“; I.11 01.11193112110. [YA 91.11111.
3,111 3,11 11.11....» 2.111.111.111110. 1.1.111. U1.1..‘111¢.1:11(‘,5 r“ c.111 @1111 a.
11.55 11':va egg-=3]

1

.1.11..
Jpni . y
.déLA ..y

.LJJJJU . .1

127

 

t -_"°yw‘flflJbehJumQHJJJ'OOWNQRJ,L¢€J?IJALLA.*\ _.
s.111,1,11j11;,1‘;111111,..2.
.H' .l

“3.1.4” . 5.11

.,1,z...~_11r1191 .1
1,1TJ1JQAuO152UQUJJ 911,11. 1.1911117111011111101311 _

.,11— .1
-,1"0— .5)
~J°£ .1’.
”V1 . J

ufl¢111u,11,151!:1,w¢,1..1111111,,11;.1r _

1.11.11 ‘1.- 0.111.111.1111? 11151118,; 91' 511,111,111.
.1111; 211,111,410; 1,1111w111011111111 2....- .
.11lmobgélsgblwwsgfigawmw .
131,3?151.’ .

u-V'c;

,wytyy,1,111¢,1.11§1133111,1110€b111V1.51 11,..111‘11,“ _
W%ol&qybg.yb£- OOWJaObte’ll-UV‘JS 34:90:
wan-11116111111191.1111;

 

Jun .1
L .1...)
\. ..‘g
Y. ..1

_J_31 111.110.,1.111_11t,.-¢,.,1.111;,,51_ 111119.111 (€111) 51,101.51.” _

11.31.-

.rbirfiingéa‘fl .1
.1.J...r1.11191‘1.....1_101115,..-11..1
. 'MJLDLU ..7
.11111111115‘112111111911 .1

128

     
   
 

‘ 1 1'.- . -' I ' I ..r- _ .r“ ". '. ' - - ' _ -.I

_, _ 111.3631. ”It; {ftp-i '7 1;." P” .51" i '.L .; ._; saga“? ‘1' :i _.

.

 

911,1 #1111111 t111.11.11.11 f.,11111,1,;11‘,.1~.- (,1. .111.» 0,11,: 11111.1 “..- .11 _
11411111211 11.-.11. .11.: 111-1:11 11x 1.3 915“ 1.1.1, 01131.11 >15 “,.- 11,1711
1:11.11 111.9,; 31,11 1111.1

mky,1,;;311,,§..11{_
.1511.- ufi... 111,115.51 .1
.11...z..",.:@1(1251_,:1)11_.-L1m9h~ .9
.hwrfidflfb‘wi: .e
.WIwWLLICa. .1

101,:11d15,1,:1j11;a11;14é‘11¢1251.w1 _
..u1111u1111yg111o11,1a111,10;,,1,112,1111 .1
151$» @wawb1b?'flhi .1.
~H—haybw'figw'flbbefi a?
.,1,.-1§111-111,11.1:1.11¢.,,:111111 .1

3,1,11,11,11 11,1165... gy1Jw1.r. _
b11413 JLz-JI

,.111- 1111—" .

5111111111111: 111.111.3711.; L.» r£1111 111511131 1211:1111- 91.521.»- .
.1111.

v N '( .._.

129

 

J.1..._.,.2. 1.11211911141J54136PMO1M10111 _
arid!
U—v -'
[1'0 .9
X“ --?
Z\-- .a

19.-S111111,..1,.11.11,..9.-,,5.1121;...11‘,...:‘_,.'<11,.¢1 r1 _
1(rlh- IL,” ,..‘wfiaffquowfluwtiumlp
“DWHLL-fibu .1

111,1.11110111...
. .Tqua .
.bfl‘m'kqfl .

1.9191111111131211111111115111.1.2.142131-1'1:—

1
“1.1.3
-’

a

 

v-V '(

1112111111241 . ,
.1..:,111 .

12.111.21.1111111114...;,..2112..21122,..11...1;.111.11‘,.1-1.f1.1,....n _
.251
.wabni: .1

1;..1..11z;1..111.2..._.

..1181 11.169394111115111. v.
.(u..ul.u) 1:31.91:- L11 and

LN '1.

.1...1U12..1111111,1111Jm3ufi1fi21311111111t1iwt,¢11'1 _

9
.1‘1.;.‘,.,1.:1~11...
J

.M‘JI 1113-le 1:113 .

130

 

 

 

0_\.L5I_,U)§.1 13.1.3: 'WyéJLralIAJIJNIJbI‘JvVA oJJJIuLLwIP ISI.

I"

011C]: q1.i._.iJ-I1_>JJ»SI1 /‘101,....1

1. ,-1'1
‘ AV ,‘1'1
M ,Yo
M. ,W

QAIWIJIHQIM‘JI
.WflLWwIflIm

{1._-111,1;wa.11. 1.2.1.111...“ng2111511.
.,1.:.\112._,1111.,:.O.U1.1.-11.,1L,12 1,L...,.J1.1>11:11 .
.rU-ILIJIJJJJIdJa-deaIObJJ‘JI-fl Wain?“ .

91 111.1,,1...11121..1u1.;,._4211..1.11121..1 1...,,1:.-9.11.11;._11.,.1.1.

 

 

1.1.: ,51 (p1

_,11L_,L_.:111,I ,lluLJ-I 051...»?! dI L4.I.1$.;..I AJI .
C” C’ 0" 9"

r-{IJL‘LU-‘fm‘UILfi-fi IHOI -

.JflIAJ-IJIE, 'IEJIMIVJU: L‘EI .

_EC‘JIIg‘ix-A-i‘ -buIJJKJI¢A,1KJIoIJLaJIL-fgI
rfilJLolwwAIIofiLkuhIofiJ;my...»

.oLbLJIJLoIquhgl IW/A- ywublrfslqu .
111,211.91. 115.11..1\11111,1...1,...,1..1.
.g.11:.111,..111.2..11 .'..1.:.j11 11,119,.- .1211 .,,:... “,1. L1;.- .

1.1-c

.Y’V

U JV '( :_.

o -b “C *L“;

.1~1

o N '( :_.

13.111.111.11 .-.‘.1 ..1...1 3311,1111, U111111,..11{11,111,311.111... L
JJuQICLzaJa:JLMU..pJIJ.JIIJAMr.1éJ

.CJiIuJ’on-Ifj opbquIbUb L

.1.th UM rL-JIqfiaJIdfllI .
..1:.\11,.21,2112.1.>_11C1:2.5,,11111111. 1.. .

8,11 1.99“ ,1 c.,11 910,1...111011 111 1. .

L111111911 1.11,...

‘rJIwJJdIWUHdJJIdJ-U .
flrUUJbHL-JM‘éJJIdrmll .
.OJW WUWI (L5H) QJJ .

131

L1)

.1,‘

It

.H

u k 'C .._.

 

 

1a

 

.tT __

 

.3111,1..11t..,1121..1,. 1111,1111 2131.11
+111: 1:141 1.11- . I

45‘?ng .1.JL. -
#0331me -

v-V “C

“EH-‘- in-H he.» -

514710.121 1_...1211 ,1. 11.-.111 :11..- .,11...L 4.1.4.11 1.411 1.11.1511, .51 11' __
14,1.:.>1J_‘.,.1.1.1111.21.311i
4,1491”... 0;) 151$...“ C31...” .
u-‘rl-‘JL-L-u- J); 9
.J1J2:..j11J.1.. ..
91.11.112.111, 4.1.5311 . 1

giyflg1gq1ufi10H 1.1 .u _
.1..1...Q$.5 *1 .11.,111 14:11 2,...11

.muflvwvgwg.
1...... (131.1111...) olmuwwofiolnmxm .

I; W '( '....

ffiId,_—:L4._,_.1.J1afbj J.4J‘JI¢1.JYIQIJL.JIO.¢I.£O _.
.ualIwbfiUkIulsgxd"

.9155wé95 199111.131.

.1L4J1...JL11.I:1._,.: aUJbY|¢JBbJQ .

..;._1..J.1 351.11 .1» 1.1. 115' J. r-‘Ix' £11.41 .

v 0‘ ‘C ‘....

132

 

 

,. 91.151118111111111. g; 11.11.111.1111 .1'1 _
.ugmufzJJUwIWLJL-JuJumn 1;. .1
JbYlwwwbuu-ayffi . 12
1.1.11 111.11% 11.511 “.11... ..

”WIF‘JyJLfiYwflIuWICJMLMfirardeg’JJLoI . .5

1111.- ,51 “3,1111% 1:1-9131;...” 41.31..» qu .tV _

 

‘JA' =JI5.11IL_NL;J.:.44 . I

'J‘°=JWIL~M6’:‘M.EJ

-,oo—=JL‘...1I;._£_- '5,“ . .1

.. .£A _
25 ..
20 ..
JIJSSI 15 ..
10 -
5 -
0 10203040506070 80 90100
11.521193 .1111 c1111:
Jl,;.‘>’¥l 2,1 UJI 11:1 9.4L...” ‘...,II_

.121»pr .I

new-v

.‘Jf-AJILF ..>.

.1I,¢..1IJa4_,;J . a

3;;1L4ng101uigymmg‘,» 911,112.11.“ _

_'— 1.1.111 .1

.64-‘11.?

@1415 al.:JI . .1

133

 

 

(”‘11 ‘.u—LLU dJerII MISU (‘1 b1) éJL-u JLJVICJ—U at; 131.0-
___ _1,..
. )\.

.Jo.
\

L111 1.

5341,1115. xo. 14:99:11 1,111.“
.1,..,11

.1...,;11.

..11,;11 .

.éJLgalILJIJaJ‘jI .

I; ‘9 .C '—-I

141,115.; 91.1L1L,,;.;§;<115.51.ov_
.1...,11 .1
Jae-o" .1.
..11,;11 ..
.‘chLrnlIJIJaile . .1

“VIDJWIMflJHGI‘I’, cjlguiyflm. of
I¢M3JUyulanhYI¢1Lnde :31.)1:»Yw,11|JU.

.5,,1..11 .
1.1-.11 .
1,1911 ..

11,1111 152.11 ..

5.1;.1___1. 5,1: 9.115111051111r1m1311.5,,;1151,z11,11 1..., 5. of.
aVuExMHQ—WJth:

5915,4113...» .i

12.51111...» 12.11.11.115. .1

.0);#a0_,$~76fa.1112?5.q

.a,ui.,.~..3,..;..1.~_11£3111.1111.1.-. ..

£115,141,1,.-1~11,,1.5;4U.,:,.,,_.. 895115.51...
.oIJIdw
.115111’3 1.14... ”1115;... JLsu .

,.JJKJ. WI JIJEJI 9.51... ‘rJlJin‘lI 2.1:..1 {1151.41 .

u '5 '[ :-.

I34

 

nguczshthubfllnmdeWAWi _

.16.;ng . .1

z.»;<i..1§.11..1.51a..3,,;11.-.1,1.-,‘v1z,...;15.,-.,.1.1:r1s.-..- 1.....ov _

 

.w
.5...» ..,1.1 5.11... 5.1.1

51.5511 “,1 u15.-.1155... .

. .JU-JI 15.1% 13:34 .
6,115,1,.,1¢.,115.1..,1.;j1151<1311. .

1.11.11 1...-11.1.;115...51.1511,.:._9;§115.‘5I.u1 _

v -V '( '...‘t.

 

1.1.51. .1

1);“ LL; '_'31.i1f«kq.y

.1.,11.1131.1;..-....'~11:.;.1.._...

.,...51..‘1.1.1.1,.251...1,.. ..
s.1.S115.;.;11.S112355;11.11»'~115.5i.M _

L111_

c—-I'-~' 111,3,11L51,.11¢1,1,.j1151 51111.11... 1.131. _
..1-..~1'15. .512.” 51. e11.11....-

..1.>11;...15,.-....111.112.-,u. .

.-|aSU¢,&,.b,-,.._Lu (8;) ILL; .

.1... 14.1.1 ..2. 5.1,;

L111 '...

5S1 1.....111 5,431.1.5u5..u.,51.;1,1~..,.s51s.:.o>1£;11‘-1.s.11 _
.;,a1,;.,,a5,§.-c:1;.-112.-,1z. 14.-1.15.55.55.11u1..11,.1..

5111;115:1115”...- 15.1.1111 1.41.11 511,1.

...11a,,1sJ>1;5.a1.L:.5$1_~1.d,1..51.-;.111.‘11u11:5:..

051.53}! 4.1.1.11 ..11511 5. 1,... 515,... 1 (5.5.111 .

o 'b it .—-O

135

 

 

“1...,”Jw1¢wogq¥1gfim1e1g1wqinx _
___- ...umobwwa: .1
-u5-e-‘)‘U1J>:“}J‘J“:’
.ulfilquwlw a,
.JAIMHJ . .1

bi» gala-5‘1”,“ d.a.U'CULJI bl; 5;?! JWIJljuqi 5),“! 111.2.“ Ln .11“ _
C .................................. Vi 011“?” £,~,.Jl.nn..'J.J|
.TJqJ... .I

'c‘bbfi‘.’ .

41.21.101.11... .

u-V'C

13.-K110. £11.53... 0.15911 11:..1 1.1.1.1) .1. .u. .11 __
.JAJ' u-w our?! J5 Jae
$11122“; 115-111 11.-4.11%“th -
. ;c_u..:JJiJ__l_:Jb oLngl ”Jim! .
$11.11 9.1 1;..1 14.-41:5 3.1.1, £115.41... g;.,.,11.;..,:,11c.1.1111.1'.1 .

v '9 'C 2._.

ou1,u1¢1.,5- 01.... .,,_..zJ1 91.31, 41.571291; .'\o _
-J|;»-J|L.|~Le 9.11 .I 4:5

-J'5-J' J...” is“: W -

0&9.“ J41: JU—J' 1w); 305-! -

«lb-fl 11W 9J1! J' 43‘ -

u k 'C '_.

..JLdI a1...“ 1:5 u1_.:..~;1 11:..1 L,11 w... o. 091 .1251

136

 

 

apL—uyl
559/) Q)!

81,. 1r”...1...\11..1.,,$.—\
..1;._1.1..1.g1.11..11c.1.1,.11 “5.1.11me—
.115111.,~11Wu.1.u..1,r.,,1...1.11....1..‘110...,1.11...wa 1'
‘11.11 015.111,, r5,111.1. 91:51.3...11141011211
..1,-1,.U.,.:1,1..1..11...1..S11u...,1.11...u.1..\1- 1
.LhayIHULJLAwa1-O

. .............................. 1.1.1.412. ................................ suyyuui. r
..1,,:5.11111

.HUI [Y]

1...-1.1111 / “My...“ .. 11111.1”,11111111

v.1...111 /:,.,.,;.1§.11 111

uh." 119.1.“ 3314.2 [1]
.LL-JI 4.911..." 35H... 1')" JJI [V]

0.1.,11a113u1...1..1..11~1 P.,.11,...111,...11..,...1‘,1..,,.‘,1.1.1,1.. 1.1.1
..................... y 1...; 1Lyj 1'21;

 

.wdb1y9uw1yauiwo. ...171¢c1:.1.,1c.;1<151
. ......................................................................................................... .Lijl

[,...1wa/wnggw1m1¢wbmgyMko. o

137

 

. ............................................................... 10....“ 3.1.. 1

. .......................................... 11.51.)..11) 1P3.- 5,112.111. v
. .......................................... 111-.1 c.1.11 9.1.511 1.1.... ' 9.11 ..111

14.1. arm/10.91.11) 14.3.- g.11 4.1.11 45,11. 11
. .......................................... 11.595211)

. .......................................... 1WJ.:11§,1.J,51.1.9L....... '1

. ..................... 1111: 01.11.53.111 &.1. 931.51: 9,111 .91... 1... 1.

album J..11w.1,111¢1‘.1.:iq;11o1,,z11z.:1..11
1.1.1.,1..1.......1,u.,51J1...1.15..)

.bbwobJJL-yrfi ..................
.MJJ-II U1 “Jun/J” ..................

1...... ..1,0.,:51)1..51.11;§.51.1...§11.d.1,.91 1.451 9:11.91Jfi1g 1.11

.UJJb'l wiWJI‘U /J,io ..................
. .................. igyfiIrUizlbwgflngpabiwiugiquf'

._ ................ s.L..jJI.L;.1,.g5JJ1..>.>i.-..¢ch.giqj.\1

138

  
  

  

1. $13!?"

   

 

01,1.5511 14......- J51 1..,.11 .015, J.1.) 4.3, J. .J;11.,..;11J. 1. ‘1...- \o
1,1,;5511.1:.1...1,., .-.1.,.11,,..1.-, .-.1._,.11 ,_.11:, C._.....11, .1..11.._. J. 41.11
.1...1_, .1.1,-.i)

/00aJJi ..................
/\ 1411/0 ..................

14.1., 1....- J.11 11,1..11 .:.1,1..~11r.5..-..: .,.-‘11J.1,5'~11J.J.,5J51.1'1

.s—ij'dJJMboflle fun-a.a ..................

r5111J...J1J.1,5 ..................

J.1,.11r11.11 J.1_..- ..................

:LE” «AMI SJUI ..................

.CJLyJJJI c3011 ,1»: ..................
.J.J.;111.,15.: ..................

. .................................... 1.111;.1J. 1....)J,51J.1,11 ..................

1.1..114:1,1,.51:._151 41.....- .,:‘11_t1,-1.1..\11 J. t)" J1. w
1..-.1,1.J.1.. 1.4....1,,-,.,51,1,.-.145.)

311.11, ..1,..11 ..................

.1..,1,11) 112.1111 ..................

.JL.S\_‘I_ .................

..,,..211..15j11.-.11 ..................

.11... J.,1,:5§11 ..................
.1,51,i ...1... 11.1.1411. 1,511.;11211 ..................

........................................................................................ (191L233 Cr‘ Ln») ‘5,“

1.1.51.1..11....11.1...1...1.....'1J111J...S11.1..‘11 £1,11J.t,;Jr.1.:,.1/1
151,...J5511=-1.51..a.-,5‘11=

.(L'DJ‘JH) Liam] ..................
.J-LI-n.‘ O‘J‘QY' ..................
.JL5§11 ..................

..,,.x11.,1.§11.;.1s ..................
. 15,1111! ..................

139

 

22.1.ng1.1...141.J,.1.),1,.-.1:L.,‘,1.:,ic_.:J..r1.J4.-. .................. .11
1,1..‘11

1...;[1]
.5111]

1..:1 1.‘.1.:J.1.-.1,1,.-.711:11:.51J,u:1;1..r,z:J. ................... 1'.

‘4’

Sjké‘ll (5'45” L. “153L993 047-1151..» 1.4.5 .................. .YT
.15L:I[\]

31.9142? [Y]

.11.”)

4.31.5 [L]

.wlo]

“[4: 9.9.13 “114,199.. d.ULljadJ JszJIJ.1.z11¢ 451-19.411.15! 9.1.]?

114,1.J... 1......1,J.,51,1,.-.14-.4.)

.1..1..1,11J,.1. -.11:.1 (,...1 ..1,1.-.~11 1.1.... ..................

.11.“, .w1..11:.).1.511J._ .11... [1,1 .1. ..................

..,...11.1.-1,1..1 1.1.1111, 1.1.51... ...511 1J.1t,: J1 .1,» ..................

.J1,.11.,...J,:.. .11..) .,...11.1:1,1..51.1. _I___J,1£tl',l... 1,.1 ..................
(Jb-J'he-sz-‘fl

1.;,...11J.1....11..),1..‘11J.. ..................

.1J1.1.111J11,:11 J15), 1.711 .11.: ..................

J1..1,1..~111.1.,§ ,_..:_ ..................

..1'..>111 1.11.1 ..................

. ...................................................... (cl-Lb! 0" bib) .5}! ..................

 

..u,1..J,..1.: J,,4.:J.11..11

140

 

 

 

 

 

APPENDIX E

 

 

MICHIGAN STATE UNIVERSITY

 

(III'ICE OI‘ VICE PRESIDENT HIII RI-ZSI'LAIICII ‘ I::\N'I' IANSINC: I MICHIGAN O «timid-[046
ANI) III-IAN 0! TIME GRADUATE SCHOOL

April 13, 1992

Mr. Rashid H. Aldosary
919-1 Cherry Lane
East Lansing, MI 48823

RE: ASSESSING THE COMPETENCY 0F TEACHERS, CURRICULUM SPECIALISTS, AND
PROSPECTIVE TEACHERS IN EDUCATIONAL MEASUREMENT IN BAHRAIN, IRB #92-167

Dear Mr. Aldosary:

The above project is exempt from full UCRIHS review. The proposed research
protocol has been reviewed by a member of the UCRIHS committee. The rights and
welfare of human subjects appear to be protected and you have approval to conduct
the research.

You are reminded that UCRIHS approval is valid for one calendar year. If you
plan to continue this project beyond one year, please make provisions for
obtaining appropriate UCRIHS approval one month prior to April 10, 1993.

Any changes in procedures involving human subjects must be reviewed by UCRIHS
prior to initiation of the change. UCRIHS must also be notifed promptly of any
problems (unexpected side effects, complaints, etc.) involving human subjects
during the course of the work.

Thank you for bringing this project to my attention. If I can be of any future
help, please do not hesitate to let me know.

Sincerely,
12.32. 1.3 ,1‘
David E. Wright, Ph ., Chair
~rUniversity Committe Research Involving
Human Subjects (UCRIHS)
DEW/pjm

cc: Dr. Irvin J. Lehmann

“HI .1 .m ,lHnmum-v A. 1.....” wall “Nam...” Inunlulmn

l 41

 

 

1

MICHIGANSTATEUNIVERflTY

_——_—_____——————————-———_

DEPARTMENT OF LINGUIS'ITCJ AND GIERMANIC. Telephone: HUSH-0740
SLAVIC. ASIAN AND AFRICAN LANGUAGES Fammxk: “WHO-27%

A-6H WEILS HALL
EAST LANSING. MICHIGAN 488144027

.April 24 , 1992

To Whom it May Concern

This is to certify that I have reviewed the Arabic version of the
"Test of Educational Measurement" (MET) written in English by
Rashid Aldosary, Ph. D. candidate at the Department of Educational
Psychology & Special Education, College of Education, Michigan
state University.

I found that the Arabic version is true and correct translation of
the original document to the best of my knowledge.

Sincerely,

lei—ELM M .

Dr. Malik Balla
Instructor of Arabic Language.

impartsram of Linguisiizo and
Cermznic. Slavic, Asian and
’ 'r-“n Languages

~ "#3:: University

' ”Ti-1027

142

Trln (n30-I77-Il-IK MCI Cal-Ir MKLIINII'RD “SIT

AHU u an ,Ij/muuln'r zhlum/Luuul ()plmmmur lumnuum

#_—4

 

 

MICHIGAN STATE UNIVERSITY

 

COLLEGE OF EDUCATION - DEPARTMENT OF COUNSELING. EAST LANSING - MICHIGAN - ("N-l0“
EDUCATIONAL PSYCHOLOGY AND SPECIAL EDUCATION

May 18, 1992

Dr. Abbas Adeeby

Dean of the College of Education
University of Bahrain

Isa Town, BAHRAIN

Dear Dr. Adeeby:

I am writing this letter on behalf of Mr. Rashid Aldosary, a
doctoral candidate at Michigan state University, who is working
on his doctoral dissertation research.

Rashid is conducting a study on "ASSESSING THE COMPETENCY OF
TEACHERS, CURRICULUM SPECIALISTS, AND PROSPECTIVE TEACHERS IN
EDUCATIONAL MEASUREMENT IN BAHRAIN". This study will provide
data that hopefully, in the coming years, will improve the
competence of teachers, prospective teachers, and curriculum
specialists in the area of educational measurement. In addition,
it is hoped that the results of this study will serve as an
important element in bringing about more coordination and
integration between the Ministry of Education and the University
of Bahrain in planning preservice and inservice measurement
training programs for these three groups.

I would be very appreciative if you would permit Rashid to
conduct his study on the senior students at your College of
Education.

Thank you in advance for your cooperation.

Respectfully,

saw no...

Irvin J. Lehmann
Professor of Measurement, and
Doctoral Committee Chairperson

143

Mm" an /II/l"”lu“l'l' ’ .. (If/Hulululv [mum/mu

 

 

MICHIGAN STATE UNIVERSLFY

 

COLLEGE OF EDUCATION ' DEPARTMENT OF COUNSELING. I-IAST LANSING - MICHIGAN - {KIN-I0.“
EDUCATIONAL PSYCHOLOGY AND SPECIAL EDUCATION

May 18, 1992

Dr. Ebrahim Y. Alabdallah
Director of Curricula
Ministry of Education
Manama, Bahrain

Dear Dr. Alabdallah:

I am writing this letter on behalf of Mr. Rashid Aldosary, a
doctoral candidate at Michigan State University, who is working
on his doctoral dissertation research.

Rashid is conducting a study on "ASSESSING THE COMPETENCY OF
TEACHERS, CURRICULUM SPECIALISTS, AND PROSPECTIVE TEACHERS IN
EDUCATIONAL MEASUREMENT IN BAHRAIN". This study will provide
data that hopefully, in the coming years, will improve the
competence of teachers, prospective teachers, and curriculum
specialists in the area of educational measurement. In addition,
it is hoped that the results of this study will serve as an
important element in bringing about more coordination and
integration between the Ministry of Education and the University
of Bahrain in planning preservice and inservice measurement
training programs for these three groups.

I would be very appreciative if you would permit Rashid to

conduct his study on the u. .
— - curriculum specialists and
teachers in Bahrain.

Thank you in advance for your cooperation.

Respectfully,

Mg 75%...-

Irvin J. Lehmann
Professor of Measurement, and
Doctoral Committee Chairperson

144

MN I u an III/Irmuln'c A(hnn/l‘.quu/ ()[ifmrlvllilv Iumwm»

 

STATE OF BAHRAIN
MINISTRY OF EDUCATION

 

HJJ‘QAA’JI m\,_....

 

,_i,.n L__.,1.1i,_.1. sis—'an
. 1.1., LIPI, ‘.....k Es;

'H)‘ e-‘W‘ 5‘4} 0:93“ ray-‘3“: u-W‘c-‘ke‘ drfl‘ 4“; -‘-' ‘2 Ju-I‘“ 03!
,_,.,ml, u'W‘ a" “2:354“ L‘UJ Jr” 4‘43: ' drfl‘ oil—‘9'.) cram-h“ 3"“ 9:45 J,» '—=~
- a.sv'xl 14.1.2.“ guru. gulls, mi, 3.1..\_-. Q. 0:24“

'._..'\1. WM 11.1.4.» ,.Lg'ulc: .le in...» ' I0, 2,...» ,ufii o' u...» v.3:
_-.—...JI;I..I;1'._..IO‘ Lbzl‘g “Jump guru ”Jug“ H 9.111, 93.3.... 99-;44‘
-( you.“ ,..- gam Wrw ...u. .3;le ml ()1 3... 51,.)

"H.111 ..u...: d.a eua, _-._L.JI ml Qatié...) cu Wadi ,3, bus: ,.ng Us,

e._‘_.u,_,s (”gum 0.91.1: 31.4.1.“ ,ufin

‘r‘v‘w L224“: 0154‘ CPU «Jr-L ‘.r‘A-‘W 'J

: 1_..Ll- LE4);

. 61L.“ 1J|1;._.,.\« a; WJJ‘.“ Q.- L,.L\\.I.;..~I .u. 5.5..” guy—:- rel-45 ”your”,
- (mar/film got. em,

145

 

 

run ‘......
.oyylwul 3.1M / 0,9.“ / your-Ls;

"L—-——‘“J 3;" 1:3

..b. .139 you guy, ,podl ”an, 01,211me ,5...” .L, 1.1, sum (9;.
94,211 9.; (yflI 2...L._.._) 4.3.1.11 ..w, you you-,1, 03.,311 LU! r33. ,1,
c» tbs-5‘ sic-r” r943: gyfi' rob-”1 94:1“ 9’ '55:le in) J23 #1332 ngfl'
is»: will own ah—u Lil. M.

‘...m, ..Qyfil 01,111 vs fies}, ,. ml r..zn 1..... 9. 0,5..- .szI -511! 31.1
..utajll. _,. pen

.1an gur'arizfiub‘u .ufilobfilfiwohhumfllwrfl rfiflljg,
...Ipn

ell». a. ..‘1— 23.-a. L»- J-L-s“ a. nun-1|. is?! ..a‘ w. ...u- 9.; cat-.14! J!
«low ..4 Leis

fits-E“. .st .5 ‘MI C9? ..a 49.11. .32“ W.

.rsi,t.a,_.1.n, page 1.1. I'M,

' ll.
guitar” ’

146

 

 

/

UNIVERSITY OF BAHRAIN
COLLEGE OF EDUCATION

Department oi Education

St‘fj-U‘ L315
13)—“IF“-

 

 

-pini/,—a‘—-/iv : fut-U
' “___,Jul '.____J.'. .a__.u ; Q__.
. gsJuingJJJié, 5.ng v;,.‘..._..n.:.~u 0,1;J...n Lil—ii ; u._n

. .5J__,.ui 1.:inth ..,.. j...- ”fir—a...“

AJSJLélsa;

u... ,.n zU-‘w , sag...” ei__xJ.JL.,.,,.J.\, .5J.J..n .-_-iJ ..Pn 0L5. Lu

:_~.., EIJI 5,; a... Jew . ”#1414 .-_1L..J cred- a.a...“ say...”
._-..uL_. Lexus. Q. edgy! 0.1.; ,__-_.J....n_-JI ”1:; J...” .,.:1-_u v1.
..Jzux eL.J..L..Ji v1. M. j... Us. ,.qu: o_._iJ-_- 9.111 mg; JL¢:.;'.i
L__'Js:. 5:... 0. ,.LI J__.1.;..i_-_i ,sutfl ,gJLg 0.. ”an, .‘i L'_‘.i_,

. L;..,'_-._,

. 14:1 'l._‘_L_ 2.1L; 5.1,.

___—___—

P.o. Box : 32038 - Sm: cl Bahrain _._.....'I up- rvu. -1.
Telephoneudmu li-'ll:..n..a
Telex 19552 JAMEA 8N

3,..-luL. “ltd—S:

Fax :u2225 ’ unnzvsu

147

 

 

STATE OF BAHRAIN
MINISTRY OF EDUCATION

 

UHJ-WHJL'J'J seal 2....

(“yr-V1“ ‘15-)“ r\\\1/\./o “J;

J_;,11 /:.u..au
.b—u-E‘J'LJ‘LP — LHJ-U'US

.JHJriJhyinflinLLgfLQfiiLéfi

c1111: Jami, gram nus ,...1. J).- z..iJ. .iJ.,L_. 11.11. “:1 U111. .12.in 1,1
wow .2 alas.“ lewd-.41 cm. . My.“ u-LAJIUJMHJI inks: ) 0.4.11 vile-“J
is...“ .1...“ our». .An-s.» w. a... a.» “mums-m.

+91 zuwweé.uleylw¥aw1~m.

,..31...iwuisumIJWIwwi:I.ib.m,il‘c_1iic__.JJ;JrJg-JIJ_,J1 111
c111J a...“ 211': b. «3......1J ..Juui pit-M ,J. g...” ,st e...“ mi my 1.911 .111,

.LHASU
.C_L,.U_1,_s..u_ml;,ix_.iy.u1.,._ss,usf_sii,su
A.“
6...,11IJI....~.IJ
a.a-:23“ {-9335 wi—efl'g—il—Ai!
(Jpll)‘iT\.\. :CLJLA

(J..Jl)nrrwi
(MUHIJHJHLH

148

 

 

STATE OF BAHRAIN

MINISTRY OF EDUCATION
5.1.1:“ mm. ...Ls'il 1_....

 

H I...“ 5L1;
(IN/f - I/‘V'I :9)“ (\SAY/L/I :V‘
Pi,“ / (Ii—'...“

'b—ue-“Lnie - 9.21191:

1»... (1.2.11. ,...1”... L»!- 1.1:: ‘—.-~

e.L-.1I ”4.1.11, 9...,111 2.11: P.... 4,. 1.1,. .131. uh. ~11 0111. from 1,1

,,.L_.;11,,1.I_,_,:.s.111._,..1._.u.1111_,.w.;.11,umJ1(,-,._.J..,111..L,._.);¢.1.11..>LI.11J
- 1.2.9.91 :...-.11 «1.1.19 116.322.- 1.., we I» ass-111.9311.
.flgIZngiciJigrSJungJfliwaLL-wkm;

,..JL...ImulsuwIJuuluuqu.1.I,,.1...,11c...11c_.,,.‘,p5.11_,.,1111

.111, e...“ 3111,). “'...-.1 .1qu (11:..1 c»- ce-u: JsL'. ..._.11:1.11=.1...Y 1.1.1.41 .1111,
9....11 mi»... «)4,)\ 5.4% L5" JVC’JI car 1.111
5.5.) .190); ..f- fig/JJM //~/“e

.wwWWWH‘MuC/Angirsu

r—w
“’...-yell JLn Ami-IJ

Urn-1“ fist-“.1 u-HUIVJMI

/ : a.5/vi.) (gym-Int :._1:.I..
,5...) Mix-f W
”(4%; ‘1‘ )3} //4) i" ' (M1)\uj._._,.-.\'11.\I

149

 

 

   
   

    

STATE OF BAHRAIN ,‘ , . ‘ 4,411 L) J3

MINISTRY OF EDUCATION 3.9,! Mbkfxgdibb
C {—21.11 IJIJI
*——J‘“H““J~*“‘“““
‘.I‘ifir/i-Hl z”.
J_i,.II / 11.1.3“

o—.,AJ' i—M-t: — "—99%“ LAS

' VOL... - ,.Ifiqr/h/o U. 14.1ngqu 4.11.1.1 Q’s: oi .r-PJ‘

 

c
,._n...Js (,...1! LXEUJ.‘ e_...II 211i c—~£~—-'1r‘ (J. _—_-.s.;, . .1”_-.s..1.1
. - a “-5.174- I ‘1‘ 5 1h“) i _._.;I5I
,—,_111_:.1..1 1... _-I_.1u: ,J. _~._-..JI 511i ,J. &_..'.J| a)... _-1_.1u..1 A,
1_..__... __15. '_J-_.J.II d.a. 1_...116. I_,,..I.\_.IJ PJ_<.;JI ,..Ji . I_,_~.11..'>1

'"WijIK—JS

‘ P—S‘Jw d.r——!-*3‘1 dz“- LPJL‘ dr“ 5““:

,._sJ_.i
,3J_..,.II 1g. 1_..IJ

sir/'4‘ 903m: u'I-U‘ ELL"

1rI-I-2JJ..II..1LE

I’IFI'VA : \J‘J'“ _uu.

[OI L19... runs

150

STATE OF BAHRAIN
MINISTRY OF EDUCATION

 

..mr/u/N :__,,

,_._i_,..II / Jl_:_.'¥l

Fm“ b-gflILAs

 

”_m 0L... - ,.Iwr/I-la c. uh.» ”LIL.” “LA ...I 0,5.- OI fiJi

,__,_;.L4.;.I .u._. .a_..:u= Q. ._uJI “J a HIM 41an A...
A—“. —-—-s~. ' —h-' .-J' d—nf w. ‘e l—v-‘w‘; n—‘u‘r‘fi' \——«-L-—~" '1

- L_g.‘.:.:| L45

' fi—Afl-‘J 03—9413; 0,3- ‘plL: drih [dc—DJ

,_5,4I
QJ_...:AJ| ALA; ;_.'.'.|J

6:1,?“ mm, u-w ans—«I

'II'I-I~ I‘bédlmb

I'II'I'VA Z d.a.“ ASLD

“fit 40...: ru-Iq

151

 

 

STATE OF BAHRAIN \ ~/¢,~£_J\ 1_J J3

MINISTRY OF EDUCATION MbQfiIDUJ
a—‘L-J‘ 5m.

 

d:——-‘.r”‘ run“) cu“ 14*

FIIIAr/I-lra :6».

,4,“ / “3._/an
o—o-u“ l—hl-z- — 1_..Jul__4s' -

. u.) ,5;qu ,..er “I Lk'rh ’..PI ...u, ‘

« 9+"me u'L‘ _._-_.I\ “4‘ e—u cw'u Glue U WSW oe-ka‘ mu .JLAJ‘“ us
- WISI A”. JSJ MI ..an Jug-mu; by“, _-ll...z OI”: 6*“
.—n es»: “Pr“ A.“ w‘ e- em ”sum—v a» may, wk w:
I_,_;.u_.'u 1.14.4...1‘u.‘ H... L,.;u..|, ( QIJALEJI -L.:I) UL; Hy,» ww .4342
L—~.J-U‘ wayl‘S' a.» u: ...LHI 0,2. J J=r~ mum unfit E44 u‘ w.» a»
- 0102-“ mix“, china .Ji'wuu um, Isu- W4,
.3_,u ...zsa 12......“ J4: MI 6‘ ..xEJI ()1 34"...“ 6““ in”: ”and!

' i—Fflll—fils

' r—‘HJLLU dr‘”: 6.7:" at“: dr'“ 5““:

e—‘P‘
o,_..,.\JI.IL_._1 ...LIJ
oy—Jwe—ow', avao—w
:_A.:\.n
WrI-I- :Jr‘J‘

TU'FVA : JuJI

\01 d.,_,.; ru-Iq

152

STATE OF BAHRAIN

 

MINISTRY OF EDUCATION Mb@fi\bw
Ht.“ gm; '

6,—23“ rah": 9.9313,...
rH-II/II/ I :U,‘

,_;,.JI /J LN!

. 4...,E,_..IE;:,;.I;_.,.:

mg 5.), Wm c.ULLJI, 3,11,» 0. a?“ ml t-' FL... 54...: ..5 9,5; OI ”J!
. 14‘“ cu:

«‘2—9“ 2"“ (IL-(51)“ ‘5‘ ‘ LL99“ vb r54»)u“ r56“: r9 r‘” 5" H:
d;—-": 24““ at)? u‘ tS-v-E‘i' ‘15-'- dL. JJUV—u 1+}: ‘ u] 39“!" 321‘“ ‘4‘ J r-v‘
"my“ .14» +—-“ c‘ an.“ J,» v-I—id- 6‘

”(L—l Lr._.\..-,,,S-..LE._,.I....-J\¢_.._. (NE-7 4‘:- r'TJ-t-rd'u-awL-l" rm!
-,IMI/II/Iv -I-.)l-.J| (2: ,.n uh,“ J_,L

' (Sal-‘3’ 62:43") k‘JS“: \rJL‘» J,—‘3< 514‘,

I——-S>"‘

d,—————,.IJI at... .L—Zl)

69———v'-“ HFJ‘) u-UJ‘ u—‘L‘é‘
‘ITI-I~ :Ji-JI ”44.
IU'TVA : J...“ ..cu

m J”); Iu~I1

153

 

APPENDIX F

 

 

Table 4.14: Item analysis data for the TEST: Teachers.

lte

(”\ICDUIAGN

10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25

p=ltem difficulty index, robis=Point biserial correlation coefficient.

2

.40
.39
.49
.30
.57
.43
.28
.58
.39
.44
.55
.38
.48
.36
.58
.20
.54
.56
.28
.32
.49
.25
.28
.57
.29

Ipbis

-.08
.32
.26

-.02
.04
.22
.20
.06
.42
.10
.05
.20
.36
.1 1
.32
.12
.19
.34
.07
.33
.13

-.03
.04
.17
.09

item

26
27
28
29
30
31
32
33
34
35
36
37
38
39
4o
41
42
43
44
45
46
47
48
49
50

154

2

.46
.33
.39
.66
.11
.34
.44
.44
.56
.73
.10
.12
.66
.37
.48
.41
.49
.16
.29
.25
.79
.23
.31
.08
.45

pris

-.13
.13
.36
.28
.10

-.06
.12
.43
.33
.27

-.O4

-.01
.24
.39
.21
.23
.13
.15
.22

-.01
.24
.09
.18
.05
.31

Item

51
52
53
54
55
56
57
58
59
60
61
62
63
64
65

.22
.51
.62
.28
.35
.05
.30
.20
.68
.22
.39
.57
.56
.28
.56

Ipbis
.18

.35
.16
.05
.07
-.18
.19
.09
.37
-.23
.29
.33
.24
.27
.24

._—. ———‘-—'—
.**—- ——.-. -‘..I—u-- -

1 E3553! : '- 4‘.

 

 

Table 4.15: Item analysis data for the TEST: Curriculum specialists.

Item

mflmm#wl\)—*

(D

10
11
12
13
14
15
16
17
18
19
20
21

22
23
24

2

.64
.61
.74
.47
.65
.69
.45
.49
.82
.60
.40
.53
.72
.45
.76
.40
.74
.70
.41
.71
.74
.47
.16
.62

Ipbis
.35
.28
.33
.28
.15
.33
.23

-.24
.47
.23

-.08
.07
.35
.50
.52
.36
.37
.17
.19
.54
.28
.1 1
.33
.07

m
25
2e
27
28
29
so
31
32
33
34
35
36
37
38
39
4o
41

42
43
44
45
46
47
48

P.

.32
.58
.70
.74
.88
.06
.43
.50
.82
.71
.90
.15
.29
.84
.68
.67
.79

.63
.29
.64
.52
.94
.28
.69

r

—pbis

.13
.15
.55
.23
.03
-.16
.16
.16
.44
-.17
.32
-.29
.11
.42
.58
.19
.41

-.09
.34
.40
.46

-.O4
.19
.28

Item

49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65

P.

.37
.62
.48
.78
.76
.30
.50
.00
.53
.36
.94
.09
.62
.85
.76
.67
.74

rpbis
.44
.38
.23
.48
.31
.36
-.13
.00
.41
-.04
.26
-.14
.37
.25
.58
.16

.18

“—_—

.p=ltem difficulty index, rpbis=Point bisen‘al correlation coefficient.

155

 

 

Table 4.16: Item analysis data for the TEST: Prospective teachers.

Item

mummawmal

(O

10
11
12
13
14
15
16‘
17

18

19
20
21

22
23
24
25

E

.40
.68
.77
.30
.49
.64
.49
.48
.70
.42
.70
.58
.70
.28
.62
.26
.59

.75
.42
.63

.70
.47
.22
.39
27

Ipbis
-.05
.18
.23
.03
-.12
.33
.44
-.11
.21
.04
.19
.04
.23
.12
.26
.14
.26

.37
.40
.47
.33
.37
.05
.09
.04

m
26
27
28
29
30
31
32
33
34
35
36
37
38
39
4o
41
42

43
44
45
46
47
48
49
50

E

.56
.53
.62
.64
.15
.26
.53
.71
.74
.77
.10
.18
.81
.66
.49
.68
.60

.68
.48
.28
.71
.26
.44
.13
.61

Ipbis BEE
.01 51
.19 52
.37 53
.19 54

-.01 55

-.19 56
.03 57
.64 58
.24 59
.34 60

-.23 61
.03 62
.30 63
.40 64
.19 65
.38
.09
.35
.46
.11
.36
.29
.41

-.02
36

2

.33
.62
.80
.36
.31
.03
.51
.26
.90
.15
.44
.81
.80
.57
.55

—pbis
.34

.09
.36
.26
-.01
-.32
.07
-.13
.44
.02
-.05
.41
.19
.27
.25

___—___—_—__—-—————

p=Item difficulty index, rpbis=Point biserial correlation coefficient.

156

 

£-

 

 

 

BIBLIOGRAPHY

 

BIBLIOGRAPHY

Airasian, P. W. (1990 April). | r h f
tigw teacher rs size up their pupilsm, Paper presented at the annual

meeting of the National Council on Measurement in Education
Chicago.

Airasian, P. W. (1991). Perspectives on measurement instruction.

WW. 10(1) 13-16.

Alabdallah, E.Y. (1989). i I i n m v I m n
field fQIIQw-up (in Arabic). Unpublished report. Bahrain: Ministry of
Education.

Alabdallah E--Y (1990) WNW
LQLteacbeLsfiLAabic). Unpublished report. Bahrain. Ministry of

Education.

American Federation of Teachers, National Council on Measurement in
Education, and National Education Association (1991). Standards for

teacher competence in educational assessment of students.

Edusetienal Measurement; Issues and Prestige, 10(1), 30-32.

Boothroyd RA (1990) W

W
W Doctoral Dissertation. Albany, NY.
State University of New York.

Boothroyd, R.A., McMorris, R. F., 81 Pruzek, RM. (1992, April). Magic
| | w | I | I. I I' I ,7

Paper presented at the annual meeting of the National Council on
Measurement in Education, San Francisco.

Carter, K. (1984). Do teachers understand principles of writing tests? mm
W, 35, 57-60.

Ebel, R. L. (1961). Improving the competence of teachers in educational

measurement. The clearing HQIIse, 36(2),67-72.

157

 

158

Ebel, R. L. (1962). Measurement and the teacher. mm.
20. 20-24, 43.

Farr, R., Griffin, M. (1973). Measurement gaps in teacher education. Jewel
hanImn‘nT hr in7,19-28.

Fleming, M. Chambers, B. (1983). Teacher-made tests Windows on the

classroom. In W. E. Hathaway (Ed..) W
MW. No. 19. (pp. 29— 3-8).

San Francisco: Jossey-Bass.

Goehring, H.J. (1973). Course competencies for undergraduate courses in

educational tests and measurement. W191, 9, 11-20.

Goslin, DA. (1967). Ieaebersafldjesflng. New York: Russell Sage.
Green. J.A. (1963). Ieaghegmadetests. New York: Harper and Row.

Green K. E. Stager, S. F. (1986-1 987). Testing: Coursework, attitudes, and
practices EducatmaLBeseamhfiuactemi. 11 (2 ). 48-55.

Gullickson AR (1982 November) W

W (ERIC Document Reproduction Service No.
ED 229391).

Gullickson, A. R. (1984, April) hin h r r ini wih h rn
mesting. Paper presented at the annual meeting of the American
Educational Research Association New Orleans. (ERIC Document
Reproduction Service no. ED 249254).

Gullickson, A. R. (1984). Teachers perspectives of their Instructional use of

tests. .LQumaLoLEdNaticNLBeseaLcn. 77 244- 248

Gullickson A. R. (1986). Teacher education and teacher perceived needs in
educational measurement and evaluation. Jeur rnal sf

EducaticnalMeasuLemem 23(4 ) 347- 354

Gullickson, A.R., Hopkins, K..D (1987). The context of educational
measurement instruction for preservice teachers. Professor

perspective EducaticnalMeasummenLlssuesaNflactice. 6(3 )
12- 16

 

159

Harris, W.S. (1973). Agreement among NCME members on selected issues

in educational measurement. Jegirnal Qf Egggeatienal Measurement
10(1), 63-70.

Hills, JR. (1991 ). Apathy concerning grading and testing. Wm,
72(7), 540-545.

MarSO. RN. (1985, October). WW9!

9W. Paper presented at the meeting of the
Midwest Educational Research Association. Chicago. (ERIC

Document Reproduction Service No. ED 268145).

Marso R N Pigge F L (1987. March) Astawdeaeeeeemenmnm

testing ev avluatien needs an nd nrefieienei es at beginning teachers:
ImelieatimsieLstaflmflQnmem Paper presented at the meeting

of the Association for Supervision and Curriculum Development,
New Orleans. (ERIC Document Reproduction Service No. ED
283833).

Marso, R. N. Pigge, F. L (1987, October). Theag er- madetests anatesting:
WWW Paper presented at

the meeting of the Midwest Educational Research Association,
Chicago. (ERIC Document Reproduction Service No. ED 291781).

Marso, R.N.. Pigge. F. L. (1989. March). IhestatumLelaeememJeachere‘
. t" . 'I l . . l.

li ' f l r-m
Paper presented at the annual meeting of the National Council on
Measurement in Education, San Francisco. (ERIC Document
Reproduction Service No. ED 306283).

Marso, R.N., Pigge, F.L. (1991). An analysis of teacher-made tests: Item
types. cognitive demands, and item construction errors. QQmempeLam

WW. 16, 279-286.

Mayo. ST. (1964). What measurement experts think teachers ought to know
about educational measurement. WW
Measurement. 1 (1). 79-86.

 

 

160

Mayo, ST. (1967). WWW
measurement. Final Report, project No. OE4-10-01 1. Washington,

DC: United States Office of Education and Loyola University. (ERIC
Document Reproduction Service No. ED 021784).

Mayo, ST. (1970). Tr n in h hi I fir r in i |
measmement, Chicago: Loyola University. (ERIC Document
Reproduction Service No. ED 047007).

McMorris, R..,F Boothroyd, RA. (1992, April). MW
analxsiseLcIaeeLQQmJeststcjeneeanemathematics. Paper

presented at the annual meeting of the National Council on
Measurement in Education, San Francisco.

Mehrens, WA. (1987). Validity issues in teacher Iicensure tests. JemnaLo!
BeLeonneLExaluatioLirLEdueatiQn 1. 195-229.

Mehrens, W.A., Lehmann, I.J. (1984). Measutementandexatuatimm

W (3rd Edition). Orlando, FL: Holt, Rinehart,
& Winston.

Mehrens, W.A., Lehmann, I.J. (1987). Using teacher-made measurement

devices. W, 71 (496), 36-44.

Mehrens, W.A., Lehmann, I.J. (1991). MeasuLememjndflaLuatiemin

Wflh edition). Orlando, FL. Holt, Rinehart,
&Winston.

Merwin, J. C. (1989). Evaluation. In M..C Reynolds (Ed). M5353
WWW 185- 192) New York Pergamon

Newman. D.C. (1981)-Wm

Doctoral Dissertation. Atlanta, GA: Georgia State University.

Newman, D.C., Stallings, W. M. (1982). W
. . i i . . _

Paper presented at the annual meeting of the National Council on
Measurement in Education, New York. (ERIC Document
Reproduction Service No. ED 220491).

 

 

161

Nitko A..J (1989). Designing tests that are integrated with instruction.
in B L Linn (Ed) EducatienaIMeasuLemem(pp 447-474) New

York: American Council on Education and Macmillan.

Nitko, A.J. (1991). What are we teaching teachers about assessment and

why? EdueatienaLMeasuremelessueeaneflraetim10(1).2-

Noll, V.H. (1955). Requirements in educational measurement for prospective

teachers. 999mm. 82, 88-90.

Pigge, F. L. Marso, RN. (1988, November). We;
MW Paper

presented at the meeting of the Ohio Association for Supervision and
Curriculum Development, Columbus. (ERIC Document Reproduction
Service No. ED 304450).

Plake, B. S., lmpara, J..,C & Fager, J. J. (1992). W
W. Lincoln, Nebraska. Kellogg Foundation

and Buros Institute of Mental Measurements.

Roeder H.H. (1972). Are today's teachers prepared to use tests? fieabedy
l fT h r i ,59, 239-240.

Roeder, H. H. (1973). Teacher education curricula. Yourfinal grade is F.

JoumaLQLEdueatLQneLMeasuLement. 10(1) 141- 143
Rowntree, D. (1981). Ajiqtjenemgueetjen. Totowa, NJ: Barnes & Noble.

Rudman, H.C. (1987a). Classroom instruction and tests: What do we really
know about the link? NASSE Bulletin, 71 (496), 3-22.

Rudman, H.C. (1987b). The future of testing is now. Edueatjenet
MeaseLemenLIssuesancLEractice. 6(3). 5-1 1.

Rudman, H.C. (1989). integtettngjestingmnjeaehjng. Washington, DC:

Office of Educational Research and Improvement. (ERIC Document
Reproduction Service No. ED 315432).

Rudman, H..,C Kelly,J. L. Wanous, DH. Mehrens, WA. Clark CM. &
Poner AC (1980) Willem.

Institute for Research on teachIng Research Series, No.75. East
Lansing. Michigan State University.

162

Schafer, W. D. (1991). Essential assessment skills in professional education

of teachers. Eduaatienal Measutement: Issues and Practice, 10(1),
3-6.

Schafer, W. D., Lissitz, R.W. (1987). Measurement training for school
personnel: Recommendations and reality. MW

Edueatien. 38(3), 57-63.

Stiggins, R..J (1985). Improving assessment where It means the most: In the

classroom EdueatieuaLLeadeLsnip 44(2 ) 69- 74

Stiggins, R.J. (1988). Revitalizing classroom assessment: The highest

instructional priority. Wanna, 69(5), 363-368.

Stiggins R.J. (1991). Relevant classroom assessment training for teachers.

EducatienaLMeasuLenlenanseeeandfiLaetice 10(1) 7- 12

Stiggins R.J., Bridgeford, N.J. ( 1985). The ecology of classroom
assessment. Jeurnal Qf Eduaatienal Measurement, 22(4), 271-286.

Stiggins Fi--J Conklin N F (1988) Wm

Washington, DC: Office of Educational Research and improvement,
US Department of Education. (ERIC Document Reproduction Service

No. ED 303439).

Stiggins RJ Conklin N F (1992) We
W. New York: State University of

New York Press.

Stiggins, R.J., Conklin, N.F., & Bridgeford, N.J. (1986). Classroom
assessment: A key to effective education. Edugetionatfileasugememz
IssuesaneELaetice. 5(2). 5-17.

Stiggins, R.J, Frisbie, D..,A & Griswold, P.A. (1989). Inside high school
grading practices. Building a research agenda. Eeueatmat
Meesurement;l ssues and Prestige, 8(2). 5-14.

Stiggins, R..,J Griswold, PA. &Frisbie, DA. (1986). i hi h I

W. Program report. Portland, OR. Northwest
Regional Educational Laboratory. (ERIC Document Reproduction

Service No. ED 2 79713).

163

Tyler, R. W (1951). The function of measurement in improving instruction. In

EF LindQUIst (Ed) EducatienaLMeasurememIpp 47- 6-7)

Washington, DC. American Council on Education.

University of Bahrain. (1991 -1992). QQHegeefflueatjenjetelggue. Isa

Town, Bahrain: Author.

Ward, JG. (1980). T h r ' ' w
attitudes. Research Report. Washington, DC: American Federation of
Teachers.

Ward, J. G. (1982). An overview of the AFT's 'Teaching and Testing".

S. B. Anderson, L..V Coburn (Eds). Agademiclesflngenmm
WZMWEM. No.16

(PP 47- -52). San Francisco: Jossey-Bass.

Yamagishi, M. (1991). Testingin Japan. In K..E Green (Ed.). Edueatjenet
lesting;_ssues_and_Apnlieatiens(pp 169-195) New York Garland

 

 

 

 

MICHIGAN STRTE UNIV. LIBRRRIES

    

I ll lllll lllll ll

' 31293009087739