llBRARY
L Michigan State
University

 

 

 

This is to certify that the

dissertation entitled

MATHEMATICS ACHIEVEMENT AND SELF-ESTEEM AT
SECONDARY SCHOOLS IN ZAIRE: THE EFFECTS OF PRINCIPALS'
EMPHASIS ON INSTRUCTIONAL LEADERSHIP

presented by

Robert S. Prouty

has been accepted towards fulfillment
of the requirements for

PhD degree in Educational Administration
K-12

 

 

WWI/w 5,

I \,

 

Major professor

Date January 16, 1987

 

MS U i: an Affirmative Action/Equal Opportunity Institution 0-12771

 

 

)V1531_J RETURNING MATERIALS:
Place in book drop to
LIBRARJES remove this checkout from
your record. FINES will

be charged if book is
returned after the date
stamped below.

_ 0.0.199;
300 A 17“; 5‘58 4

 

 

 

 

 

‘ 55' .L 9199!.

135 l ‘
'vm ism‘

 

MATHEMATICS ACHIEVEMENT AND SELF-ESTEEM AT
SECONDARY SCHOOLS IN ZAIRE: THE EFFECTS OF PRINCIPALS'
EMPHASIS ON INSTRUCTIONAL LEADERSHIP

By

Robert S. Prouty

A DISSERTATION
submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Educational Administration
1987

ABSTRACT
MATHEMATICS ACHIEVEMENT AND SELF-ESTEEM AT

SECONDARY SCHOOLS IN ZAIRE: THE EFFECTS OF PRINCIPALS'
EMPHASIS ON INSTRUCTIONAL LEADERSHIP

By

Robert S. Prouty

The purpose of this study was to examine how secondary school
principals in Zaire affect student outcomes by emphasizing
instructional leadership. Outcomes examined include mathematics
achievement and the development of student self-esteem. Principals'
emphasis on instructional leadership was measured by the frequency of
classroom supervision and faculty meetings, by the existence of
explicit instructional objectives and by other indicators of

involvement in the instructional process.

A total of 1017 students (Grades 8 and 12), along with their
mathematics teachers and principals, participated in the main study.
Students completed the Coopersmith Self Esteem Inventory (French) and
a mathematics questionnaire (reliability coefficients of .45 and .83
respectively were obtained for these instruments). The teachers and

principals were interviewed by the researcher.

Emphasis on instructional leadership was found to have a
significant positive effect on mathematics achievement and
significant but relatively small effects on self-esteem. The

frequency of classroom supervision and the number of faculty meetings

were found to be strongly associated with mathematics achievement and

pass rates on the Grade 12 national examinations.

With the exception of students in the Grade 12 sciences option,
textbooks were associated with greater mathematics achievement only
for those students whose principals emphasize instructional
leadership. The socioeconomic status of individual students did not
significantly affect mathematics achievement or self-esteem, but
classroom SES levels had a positive effect on mathematics

achievement.

Strategies used by principals to cope with a shortage of teachers
and teaching materials include 1) hiring itinerant teachers, 2) a
four-day work week for teachers, 3) increased personal teaching load,
4) increased study time for students and 5) solicitation of financial

and material support from the community.

Direct principal/teacher contacts appear to be the most important
component of instructional leadership in Zaire; there is a need for
research which examines this in greater detail. Research is needed
also on the cost-effectiveness of potential pre-service and
in-service training programs for principals and on how principals
manage innovative approaches to instructional improvement in times of

extreme resource scarcity.

The writing of this dissertation often meant long hours away from
home, but my wife, Diane, who will soon be working on one of her own,
did her best to make sure that we were always in touch. One day,
looking through my appointment calendar, I found the letters Ph. D.
followed by a series of exclamation points in her distinctive style.
When I called to thank her for the motivation, she laughed. "I guess
it worked," she said, "but you read the message wrong. I was trying
to remind you: PHONE DIANE!"...This dissertation is dedicated to
you, Diane. Thanks for the patience and the motivation and
especially, for always... keeping in touch.

11

ACKNOWLEDGMENTS

Sincere thanks...

to Dr. Samuel Moore, my committee chairperson, for help and
guidance from the very first

to Drs. George Ferree, David Wiley, John Suehr and Alemu
Beeftu, members of my guidance committee and always
supportive and helpful

to the BRIDGES group for a shared commitment

to Bob and Jan Roberts for the motorcycle and more

to Jerome and Yvonne Pontdexter for managing to put me up
and to put up with me, both at the same time

to Mwangacucu M'heta, to Mbassome, to Kanyamayaga and all
the many others who helped with the data collection process
in one way or another

above all, to the principals, teachers, students and others

in the secondary school system of the Butembo District--

gagingya kutsibgtsibgll

iii

TABLE OF CONTENTS

Page
LIST OF TABLES .................................................... vii
LIST OF FIGURES .................................................... ix
CHAPTER ONE: STATEMENT OF THE PROBLEM .............................. 1
Introduction .................................................... 1
Need for this research .......................................... 2
Statement of the problem ........................................ 5
Statement of purpose ............................................ 5
Importance of the research ...................................... 8
General uses and limitations ................................... 10
Research questions ............................................. 15
Hypotheses ..................................................... 16
Definition of terms ............................................ 18
CHAPTER TWO: REVIEW OF THE LITERATURE ............................. 21
Introduction ................................................... 21
Part I: The Role of the Principal ............................. 24
Schools as non-rational organizations .................... 24
Effective schools research ................................ 30
The instructional leadership role of the principal ........ 41
Collaboration or prescription? ............................ 43
Self-esteem ............................................... 46
Part II: Education in Zaire ................................... 50
Introduction .............................................. 50
The Zairian context ....................................... 52
Studies of schooling in Zaire ............................. 55
Summary ........................................................ 62
CHAPTER THREE: METHODOLOGY ........................................ 64
Introduction ................................................... 64
Site location and characteristics .............................. 65
Population and sampling frame .................................. 67
Research hypotheses ............................................ 72
Data collection instruments .................................... 74
Instruments for collecting data from students ............. 76
The Coopersmith Self Esteem Inventories .............. 78

The IEA Second International Study of Mathematics
Questionnaires ..................................... 79
Instruments for collecting data from principals ........... 80
Instruments for collecting data from teachers ............. 82
Data collection procedures ..................................... 83
Access to the site ........................................ 83
Transportation ............................................ 84

iv

Procedures for collecting data from students .............. 86
Procedures for collecting data from teachers and

principals .............................................. 90
Protection of data ........................................ 91
CHAPTER FOUR: PRESENTATION AND ANALYSIS OF THE DATA ............... 94
Introduction ................................................... 94
Student outcomes ............................................... 94
Principals' emphasis on instructional leadership ............... 96
Levels of parental education ................................... 97
Correlates of mathematics achievement and self-esteem .......... 99
Data analysis ................................................. 100
Serendipity ................................................... 124
Components of instructional leadership ........... ' ............. 127
General observations .......................................... 132
Principals ............................................... 132
Teachers ................................................. 136
Students ................................................. 139
Building quality ......................................... 141
Miscellaneous ............................................ 143

CHAPTER FIVE: SUMMARY OF METHODOLOGY, FINDINGS, CONCLUSIONS
AND IMPLICATIONS ................................................ 146
Introduction .................................................. 146
Summary of methodology ........................................ 146
Summary of findings ........................................... 147
Limitations of the study ...................................... 151
Conclusions ................................................... 152
Interpretation ............................................... 153
Implications .................................................. 156
APPENDIX A: INSTRUMENTS USED FOR DATA COLLECTION ................. 161
Demographic questionnaire (students) .......................... 161
Coopersmith Self Esteem Inventories (French translation) ...... 163
Coopersmith Self Esteem Inventories (Swahili translation) ..... 164
Mathematics questionnaire (Grade 8) ........................... 166
Mathematics questionnaire (Grade 12) .......................... 174
Interview schedule (principals) ............................... 181
Interview schedule (principals) translation ................... 182
Interview schedule (teachers) ................................. 183
Interview schedule (teachers) translation ..................... 184
APPENDIX B: SITE ACCESS CORRESPONDENCE ........................... 185
APPENDIX C: LETTERS 0F PERMISSION ................................ 192
APPENDIX D: EXPLANATION OF PRINCIPAL RATINGS ..................... 195
APPENDIX E: MATHEMATICS ACHIEVEMENT AND SELF-ESTEEM SCORES ....... 198
APPENDIX F: PARENTS' LEVEL OF EDUCATION .......................... 199
APPENDIX C: ACADEMIC INTAKE RANKINGS ............................. 200

V

APPENDIX H: PEARSON PRODUCT-MOMENT CORRELATION COEFFICIENT

MATRICES ........................................................ 202
APPENDIX 1: STATE EXAMINATION RESULTS, 1983-1986 ................. 205
APPENDIX J: RULES GOVERNING TEACHER AND STUDENT BEHAVIOR

AT TWO SCHOOLS IN THE SAMPLE .................................... 206
APPENDIX K: INSTRUCTIONAL SUPERVISION FORM ....................... 209
BIBLIOGRAPHY ...................................................... 210
GENERAL REFERENCES ................................................ 220

vi

Table
3-1

LIST OF TABLES

Schools by Number of Academic Programs Offered .......... III:
Academic Programs ......................................... 70
Sponsorship of Schools .................................... 72
Ratings of Principals ..................................... 97
Variables Correlating with Mathematics Achievement and

Self-Esteem ............................................ 100
Grade 12 ANCOVA for Mathematics Achievement .............. 101
Grade 8 ANOVA for Mathematics Achievement ................ 101
Grade 12 ANOVA for Self-Esteem ........................... 104
Grade 8 ANOVA for Self-Esteem ............................ 105
Academic Intake of Schools Ranked by Inspectors .......... 106
Academic Intake of Schools Ranked According to

Admissions Criteria .................................... 109
ANOVA for Grade 8 (Principals, Textbooks) ................ 113
ANCOVA for Grade 12 (Principals, Textbooks) .............. 114
Grade 8 ANOVA for Classroom SES .......................... 118
Grade 12 ANCOVA for Classroom SES ........................ 118
Grade 12 ANCOVA for Principals, Classroom SES ............ 119
Percentage of Students Wearing Watches as an Indicator

of SES, by Class and School ............................ 120
Grade 8 Mathematics and Self-Esteem Scores ............... 198
Grade 12 Mathematics and Self-Esteem Scores .............. 198
Parents' Level of Education--Grade 8 ..................... 199
Parents' Level of Education--Grade 12 .................... 199

vii

Itata Examination Pass Rate: 1983-1986...................205

Annual State humiliation Rankszos
.7 r ..= ,“!F§‘i

 

7
Me Ji‘
1.1 Map of {Interim all-'2‘. Lint “3 7
4.1 Nathan-111:2: am; Fol? ‘JSLOC'E .‘rado ’3 ., ,. . . "L.
6.2 Mathematics and Self-Hutu» Credo . ., ‘ .
4.3 11.01 M Fathers" it": stir-x. . . .. 48
4.4 Star.- :Ixauixw”l;m - 1's i‘8'w’i‘3 2
A
.3 .
A]. ~ .4"
iv. .
‘
u\
I \l J.“-

 

Figure
1.1

LIST OF FIGURES

Page
Map of Butembo School District ............................. 7
Mathematics and Self-esteem, Grade 8 ...................... 95
Mathematics and Self-esteem, Grade 12 ..................... 95
Level of Mothers' Education ............................... 98
State Examination Results 1983-1986 ...................... 123

CHAPTER ONE

STATEMENT OF THE PROBLEM

It isn't easy for a principal in Zaire because...
teachers feel as if they're on a trampoline. They
bounce up and down, getting nowhere, and as soon
as they can, they jump off.

-primary school inspector, Butembo

Introduction

In recent years, an increasing number of researchers have
examined the question of school effectiveness. Their studies
generally have concentrated on the relationship between student
achievement and school inputs or processes. Most of the activity in
this field has taken place in schools within the United States but
there has been a growing interest among researchers in developing

countries.

Although many of the effective schools studies have been
criticized both on methodological and conceptual grounds, they have
nevertheless produced remarkably similar findings. Perhaps the most
consistent of these is that the instructional leadership role of the
principal is a key to school effectiveness (Averch, et a1., 1971;

Purkey & Smith, 1982; Good & Brophy, 1985; etc.).

The instructional leadership role of principals is one whose

effect is generally achieved indirectly, through the mediation of

2
teachers in the classroom. The present study represents an attempt
to determine the relationship between principals' emphasis on
instructional leadership and school effectiveness in a country where
teachers are among the lowest paid in the world and where teacher
turnover is high. The field work was conducted in Zaire, in the
Lubero Region of the province of Kivu. It was carried out in French,
the language of all post-primary schooling in the country. A major
difference from most previous studies in the effective schools
tradition is that effectiveness was defined in terms of both a
cognitive student outcome (mathematics achievement) and a

non-cognitive student outcome (development of self-esteem).

e t s Re earch
This research responds to several gaps in the current effective
schools literature: 1) it seeks to broaden the definition of
effectiveness by including outcomes other than purely cognitive ones;
2) it explores the implications of effective schools research for
educational administration in a developing country; and 3) it
provides an empirical description of the relationship between

instructional leadership and student outcomes.

Although effective schools researchers have consistently
highlighted the importance of administrative and instructional
leadership, other researchers have emphasized the nonrationality of
schools as organizations (March & Olsen, 1976; Weick, 1976; Wise,

1979; etc.). They have pointed to the "loosely coupled" nature of

3
schools, indicating a weak link between educational administration
and student outcomes and suggesting that instructional leadership is
an inappropriate focus for secondary school principals, since
instruction at that level is a 'soft' technology, best left in the

hands of teachers.

Part of the reason for these differing perceptions of reality has
been the use of different definitions of school effectiveness and
different measures of school outcomes. The effective schools research
tradition typically has adopted a definition of school effectiveness
based on student academic achievement as measured by standardized
tests. Principal effectiveness has generally been defined and
measured in a similar fashion (Huff, et a1., 1982). Even where other
measures are used (superintendent's ratings, parent and teacher
satisfaction, etc.), there is on} rarely any indication that student
outcomes other than academic chievement have been considered.
Researchers examining schools as non-rational organizations, on the

other hand, have considered a much broader array of school outcomes.

In their review of effectiveness studies, Purkey and Smith (1983)
have suggested a need to look at other outcomes of schooling. Cuban
(1984) would expand the definition of effectiveness to include:
"sharing, learning to make decisions, developing self-esteem, and
acquiring higher-level thinking skills and aesthetic sense." Good
and Brophy (1985) add their voices to the chorus, stating flatly that
"we need...more basic research on other important outcomes of

schooling" (p.599). Finally, Rowan, Dwyer & Bossert (1982) urge the

 

4
"development of a multidimensional view of school effectiveness"
which would include the development of self-esteem and other outcomes
(although, interestingly, a 1982 study of principal succession
effects by Rowan uses only student academic achievement to measure

the effect principals have).

There is, at present, little literature on effective schools pg;
gg in developing countries, and virtually none on principal
effectiveness, yet the related studies have shared the tendency
toward a narrow definition of success (Simmons & Alexander, 1978).
Simmons and Alexander have suggested that one of the greatest
weaknesses of much of the available research in developing countries
has been the exclusive concern with simple cognitive outcomes such as
those typically associated with examination scores. They argue for
an approach which would "examine multiple schooling outputs.
Academic achievement, drop-out rate, modernity, motivation,

self-esteem should be treated as simultaneously determined outputs."

Several reviews of the literature call attention to the lack of
studies of effective secondary schools (Purkey & Smith, 1982;
Manasse, 1985). Most of the work which has been done is on the
elementary level. Yet secondary schools differ in important ways
from elementary schools and a great deal of research is needed to
show which dimensions, if any, of the effective elementary schools
research may be applicable to secondary schools. Studies by Rutter
et a1. (1979) and by Coleman et a1. (1981) do indicate that this

concept is worth pursuing on the secondary level. Rutter, in

5
particular, shows that secondary schools sharing similar demographic
characteristics differ in important ways as to student outcomes. His
study is one of the few to look beyond cognitive outcomes and his
finding that there appears to be a causal link between certain inputs
and given outcomes points up the need for further research in this

area .

Many of these comments are directed to definitions of effective
schools rather than to definitions of effective principals. This is
because effective principals generally have been seen as those who
run effective schools, even though it is widely recognized that many
factors can intervene between the actions of the principal and any
measure of school effectiveness (Manasse, 1985, p. 440). This means,
of course, that any incompleteness in the definition of effective
schooling will make the relationship between the principal's
activities and school success, or effectiveness, even harder to

trace .

Statement of the Problem

The problem which the researcher addressed is that of the
relationship between the emphasis which secondary school principals
in Zaire place on instructional leadership, and two outcomes of

schooling: mathematics achievement and student self-esteem.

Statement of Purpose
The specific purpose of this study was to examine the

instructional leadership of secondary school principals in Zaire in

6
relationship to both mathematics achievement and the development of
student self-esteem. Specific components of this leadership were
identified and described. Correlations between factors associated

with these variables were examined.

This study was exploratory in nature; it sought to identify
constructs which could be of value in future studies of educational
administration in general and of instructional leadership in
particular. It included second year (Grade 8) and sixth year (Grade
12) students from fifteen secondary schools in the Lubero Region of
the Kivu Province of Zaire (see page seven for map showing school

district studied and location of sample schools).

Each student was asked to complete the 25-item Coopersmith Self
Esteem Instrument (translated by the researcher into French) and
items derived from the cross-national pool of mathematics questions
from the Second International Study of Mathematics conducted by the
International Association for the Evaluation of Educational
Achievement (Chang & Ruzicka, 1985). Students responded to
mathematics questions translated by the researcher into French from
the IEA study. Principals were interviewed by the researcher. These
interviews were structured around a fifteen-item questionnaire which
was administered orally and which was followed by a more open-ended
format. The entire interview process varied considerably in length,
lasting from one to four hours. The mathematics teachers at the

schools were also interviewed, as were all three primary school

ITURI FOREST

 

 

Figure 1-1

Map of Butembo School District

 

inspectors and both secondary school inspectors, who provided

valuable information about schools and schooling in the district.

Impgztange of the Research

A number of studies have documented the economic returns of
education in developing countries (Psacharopoulos, 1973;
Psacharopoulos & Woodhall, 1985). These economic benefits may be
attributable to both cognitive and non-cognitive behavioral changes
(Colclough, 1980; Eisemon, 1985), yet studies in developing countries
have dealt almost exclusively with cognitive outcomes. This study,
emphasizing a cognitive and a non-cognitive outcome, will begin

filling this void in the literature.

Many researchers have concluded that the relative importance of
school effects is far greater in most developing countries than in
the more industrialized countries (Heyneman & Loxley, 1983;
Colclough, 1980; Currie, 1977). This places an even greater burden
on researchers in developing nations to examine outcomes of schooling

and those processes and inputs which affect them.

There is limited but growing evidence to indicate that the
importance of principal leadership is greater for schools with lower
socioeconomic status (Rowan, 1982) and that factors associated with
effectiveness in these schools may differ (Hallinger & Murphy,
1985). The effective schools researchers in particular have
emphasized equity concerns related to the quality of schooling for

low socioeconomic status students. It is far from clear, however, to

 

9
what extent findings pertaining to socioeconomic status in developed

countries will hold true for developing countries.

Positive self-esteem has been shown in Western studies to
facilitate the development of mature career attitudes, leading to
improved academic and work achievement (Crook, et a1., 1984). Most
studies have supported the hypothesis that a low but relatively
consistent positive relationship exists between self-esteem measures
and measures of academic achievement (Hansford & Hattie, 1982). It
is uncertain to what extent these findings can be generalized to
Third World settings. Much of the current work on research
utilization suggests that research never has direct, unambiguous
application to other settings (Buschmann, 1984). This study and
others like it should help clarify to what extent these Western

studies will be applicable to the Zairian context.

At present, the relationship between specific instructional
leadership variables and outcomes of schooling is unclear. Yet a
number of commentators (Cuban, 1983; Steers, 1975; Rowan et a1.,
1983;) have pointed out that an emphasis on test scores means that
there is a trade-off occurring somewhere (see Elmore, 1982, for a
more complete discussion of the issue of trade-offs). Something else
must be de-emphasized, whether deliberately or not. This study
should help provide both the emphasis and the information needed for
educators in Zaire to ensure that trade-offs are both meaningful and

desirable.

10
e U c a d imitations

This is an exploratory research study which has been somewhat
developmental in nature. Although the original research design has
generally been followed, several variables were encountered during
the interview process and consequently received close attention in
the latter stages of the study but none at the beginning. An example
is the pattern of itinerant teachers which was discovered part way
through the study. A significant number of schools share teachers
for key subjects. Although this was not a major part of the study,
it did shape the later teacher interviews. A second example is the
practice among principals in some of the larger schools of naming a
proviseur to be responsible for classroom visits and other
instructional matters. The government stopped paying the salary of
the proviseurs in 1984 but in the course of the study it was learned
that many schools had made internal adjustments, with teachers
picking up extra classes so that the position of proviseu; could be
maintained. Consequently, the study design was changed to include

informal interviews with provisegzs.

The major limitations faced by the researcher follow from the
restrictions in design which are necessarily imposed by a study of an
exploratory nature. The instruments used in measuring key variables,
for instance, were selected to minimize cultural distortion but had
not been previously tested in Zaire. Reliability coefficients for
the self-esteem measurements in particular were found to be

relatively low (.40 and .49 for Grade 8 and Grade 12 respectively).

11

The design of the study was also necessarily limited in scope.
For instance, while the vital importance of teachers in determining
student outcomes was recognized, the researcher did not attempt to
control for differences among teachers (other than academic
preparation and years of experience). This means that the
explanatory power of the model is not as great as it would otherwise
have been. Similarly, instructional leadership includes working with
parent groups to enhance parent participation in the learning process
and working with teachers and students to establish a climate of
security, commitment, and cooperation within the school. These

functions of principals are dealt with, but not at length.

Instructional leadership is a complex component of educational
administration and the researcher does not claim to do entire justice
to that complexity. Rather, an attempt was made to show simply that
an emphasis on instructional leadership does have a statistically
measurable effect on important student outcomes and that this effect
is independent of student background variables. An in-depth
treatment of the mechanisms through which this effect occurs is not

within the scope of this study.

Since schools from only one region were tested, with students
largely from one ethnic group (the Nande), the results may reflect an
inherent cultural bias and thus not be applicable directly to other
ethnic groups. Ethnicity may be related to instructional leadership,
self—esteem, or achievement, or to the interactions among them. The

language of early primary schooling (in this case, Swahili) also may

 

12
affect the variables studied, or at least the measurement of them.
Language also may have affected student comprehension of the test,
particularly for the second-year students, some of whom had not

attained a high level of French mastery.

The research took place in a largely rural area (although it
included one rural town of over 100,000 inhabitants). Its
applicability to urban settings, while certainly very possible, will

need to be tested with further research.

Socioeconomic status is another variable the measurement of which
is always problematic. In this study, the level of education of both
the father and the mother were measured, with the mother's education
ultimately serving as a proxy for socioeconomic status. A further
estimate of socioeconomic status was made by counting the number of
students wearing watches in each classroom. Still, it is hard to be
certain that either of these methods provided an entirely adequate

measurement .

There are other limitations which result from the study design.
Although a large number of student responses were obtained, the
sample size of principals was low. This made the attainment of
statistical significance difficult for correlations with various
facets of instructional leadership. Also, since the study was
cross-sectional rather than longitudinal, it was not possible to make
causal inferences based on changes over time. There was, however, an

attempt to determine stability of academic achievement over time by

 

 

l3
examining school records and in particular, state examination results

from 1983 to 1986.

Other limitations are clear. Perhaps the most significant of
these was the difficulty of controlling for differences in the
academic level of students admitted to the various schools. Two
procedures were followed in order to account for this variation.
First, a panel of expert witnesses was asked to rank the schools in
terms of academic intake (this panel consisted of three long-time
elementary school inspectors and one secondary school inspector, all
of whom have visited the schools in the sample area many times).
Second, principals and teachers were questioned about entrance

requirements. A ranking based on these interviews was devised.

A further limitation of this study is that findings may not be
generalizable to other than full six-year secondary schools. The
study design did not include vocational schools, primary schools, or
schools following a four-year cycle, rather than a full six-year
cycle. A number of reviews of the effective schools research have
pointed out that what is applicable to secondary schools may not
necessarily be applicable to other kinds of schools (Cuban, 1983;

Firestone & Herriott, 1982; Manasse, 1985).

Finally, a further confounding variable difficult to accurately
measure is the level to which principals or teachers engage in what
has come to be known as the system of matghishi. This refers to the

giving of financial favors or other bribes in order to gain admission

 

14
or advancement. Corruption has become a way of life in much of Zaire
(see, for example, Vansina, 1982) and, in those schools where it is
widely practiced, must certainly affect both achievement and
self-esteem (Gould & Amaro-Reyes, 1983). Some information about this

practice has been collected, but a great deal of uncertainty remains.

In spite of these limitations (and perhaps partially because of
them), the study should be of considerable value to researchers and
educators dealing with rural populations in developing countries.
The baseline information provided should give insight into the nature
of instructional leadership where resources are scarce and teacher
salaries low; it suggests areas of fruitful research for the future
and should give an impetus to further studies into school processes
and inputs and their correlations with a broad array of cognitive and

non-cognitive outcomes.

On a more personal level, the researcher plans to use the present
study as a pilot for a subsequent longitudinal study investigating
the complex nature of effective instructional leadership under
conditions of resource scarcity. An attempt will be made here to

identify variables and constructs for this further research.

The attempt to provide primary and secondary schooling has
demanded great efforts on the part of national governments in
developing countries (for some idea of the magnitude of these

efforts, see Williams, 1984), including that of Zaire, and the

 

15
information gathered here will help provide guidelines for the more

effective use of the scarce resources required.

gaggarch Questions

The ultimate question for which answers were sought in this study
is the following: When secondary school principals in Zaire
emphasize instructional leadership, does this lead to the attainment
of desired student outcomes? In order to get at this larger
question, it is necessary to answer many others: Do principals, in
fact, differ in the emphasis which they place on instructional
leadership? What are the current levels of student outcomes? What
is the effect of socioeconomic status? age? gender? type of program

followed? etc.

The research questions asked are stated here in general form.
They are listed in more precise research terms in Chapters Three and
Four. These questions are grouped into three categories: 1) those
that deal with school level variables; 2) those that deal with
classroom level variables; and 3) those that deal with student level
variables. Note that each question supposes an "Are there...? If
so..." format. That is, each question asks: Are there any
differences here? Each further implies: If so, what are they? What

do they correlate with? etc.

SCHOOL LEVEL QUESTIONS
l.‘ Are there differences among schools as to academic intake?

2. Are there differences among schools as to socioeconomic
status?

16

Are there differences among schools as to mathematics
achievement?

Are there differences among schools as to the levels of
student self-esteem?

Are there differences among principals as to the emphasis
which is placed on instructional leadership?

Among principals who emphasize instructional leadership, are
there differences in how this leadership is exercised?

Are there differences among schools as to academic stability?

Are there differences among schools as to the physical
quality of the buildings?

CLASSROOM LEVEL QUESTIONS

1. Are there differences in mathematics achievement and levels
of self-esteem between classes following the sciences
curriculum and classes following the humanities curriculum?

2. Are there differences between eighth-grade and twelfth-grade
scores for mathematics achievement and self-esteem?

3. Are there significant differences in class sizes?

STUDENT LEVEL QUESTIONS

1. What is the distribution across students of socioeconomic
status?

2. What is the gender distribution?

3. What is the age distribution?

4. Which students possess textbooks?

5. How many years do students spend at the same school?

New
The hypotheses are stated here in general form. They are stated

in statistical form in Chapters Three and Four. Since this was an

exploratory study, there is a larger number of minor hypotheses than

would be the case if more were known about principals and student

l7
outcomes in Zaire. The minor hypotheses were designed to determine
whether there are important confounding variables which may affect

the major hypotheses.

MAJOR HYPOTHESES

1. The emphasis which principals place on instructional
leadership affects student mathematics achievement.

2. The emphasis which principals place on instructional
leadership affects the development of student self-esteem.

3. Principals' effect on mathematics achievement is independent
of the academic ability of students enrolling at the school.

MINOR HYPOTHESES
4. The emphasis which principals place on instructional
leadership has an effect on students which grows stronger the

longer the student is at the school.

5. Students who have textbooks tend to have higher mathematics
achievement.

6. The student's socioeconomic status does not affect her/his
level of mathematics achievement.

7. The student's socioeconomic status does not affect his/her
level of self-esteem.

8. The socioeconomic level of the school does not affect the
student's level of mathematics achievement.

9. The socioeconomic level of the school does not affect the
student's level of self-esteem.

10. Mathematics achievement is positively related to self-esteem.

11. Schools where students have higher self-esteem scores have
done better on state examinations over the past four years
than have other schools.

12. Schools where students have higher mathematics scores have
done better on state examinations over the past four years

than have other schools.

13. Males have higher levels of mathematics achievement than do
females.

14. Males have higher levels of self-esteem than do females.

 

fii

18

Definition of Terms
1. Self-esteem

-The value judgment a person makes about her/himself
(Coopersmith, 1967). It is defined operationally as the score
obtained on the 25-item adult version of the Coopersmith Self Esteem

Inventory, with a higher score indicating greater self-esteem.

2. Principal effectiveness
-An effective principal, for purposes of this study, is defined

as one whose school is found to be effective.

3. School effectiveness

-The attainment of outcome levels beyond the norm for schools
with given demographic variables. Schools whose students score
significantly higher than the mean for both mathematics achievement
and self-esteem, with student intake in terms of academic achievement
held constant, will be considered to be effective. Effective schools
should also show evidence of stability as indicated by state

examination scores for the past four years.

4. Socioeconomic status
-This is defined operationally as the educational level of the
mother. A second measure has been obtained by observing the

percentage of students wearing watches.

 

19
5. Student intake (academic)

-The academic achievement level of students when first enrolling
at a school. It has been measured in two ways: 1) by asking an
expert panel consisting of three elementary school inspectors and one
secondary school inspector to rank the schools as to academic
intake; 2) by obtaining admissions criteria from principals and

determining which schools apply a more selective criteria.

6. Emphasis on instructional leadership

-This is a composite ranking of the principals determined largely
by principals' and teachers' responses to questions dealing with
eight components of instructional leadership: supervision of
instruction, faculty meetings, instructional objectives, teaching
involvement, awareness of the mathematics curriculum, stated emphasis
on instructional leadership, the purchasing and use of resources for
instructional purposes, and the degree to which a climate conducive

to learning has been established.

7. Stability of effectiveness

-This is the amount of variation in academic achievement at a
given school over the last four years. It has been measured by
comparing the state examination scores of the Grade 12 students from

1983 to 1986.

 

20
8. Student tenure
-The length of time a student has spent at his/ her current

school.

9. Schools as non-rational organizations

-This expression is used to characterize the models of schooling
put forth over the last decade in which the rational bureaucratic
structure described by Weber (1947) does not prevail. In particular,
it is used to describe the organizational models suggested by Hanson
(1975, 1976), March & Olsen (1979), Weick (1976), Sedlak et al.

(1986), Wise (1979), Meyer & Rowan (1978) and Powell et a1. (1985).

CHAPTER TWO
REVIEW OF THE LITERATURE

So...Why are you doing this anyway?...Has anyone
ever studied this sort of thing before?"
-student, School 7

Inggoduction

Over the course of the past decade, researchers have advanced a
number of different theories of schools as organizations. Many of
these have emphasized the lack of fit between the activities of
school administrators and teachers and the stated goals and
objectives of their schools. In this sense, they may be
characterized as theories of schools as non-rational organizations.
This, of course, stands in sharp apposition to the classical Weberian
treatment of schools as rational bureaucracies, a paradigm dominating

the research literature since its publication in English in 1947.

The recent growth of the body of literature known as effective
schools research has been understood by many to offer a counterweight
to the non-rational paradigms and a return to an approach more
consistent with Weber (and indeed, Dewey). This is particularly true
in its treatment of the principal as an instructional leader,
responsible for and capable of significantly affecting student
achievement. The effective schools literature, essentially, holds
that principals can work with teachers to create a school climate

favorable to learning regardless of socioeconomic and other

21

22
background variables. It is a positivistic approach rejecting the
inevitability of the processes described in much of the non-rational

literature.

The ongoing debate between these contrasting points of View has
been largely conducted in the Western world but is of vital
importance to the educational systems in developing countries as
well. Many of the questions raised (especially those dealing with
such key issues as access, equity and quality) are matters of even
more pressing concern for planners, policy makers and researchers
dealing with the development of educational systems in Third World
countries. These countries have demonstrated an astonishing level of
commitment to education, investing prodigious efforts over the past
ten to fifteen years toward the provision of universal primary
education and continuing to struggle with the expansion of secondary

and tertiary education.

This review of the literature is divided into two parts. Part I
examines the role of the principal within the context of recent
theory dealing with schools as organizations. An emphasis is placed
on the effective schools research and on the principal as an
instructional leader. The perspective of research and researchers
from developing countries is included and the importance of giving
simultaneous consideration to the multiple outcomes of schooling is
defended. The current literature dealing with the development of
student self-esteem is examined briefly as representative of the

literature dealing with the non-cognitive outcomes of schooling.

23
Some of the pitfalls in the application of Western research to
developing countries are also briefly considered, along with the
potential benefits of such research. Part II presents an overview of
the Zairian educational context and current educational concerns in
Zaire. Studies which have been conducted in developing countries and
which deal with concepts related to the effective schools literature
are presented. A final section synthesizes what has been learned

from the literature.

24

Part I

THE ROLE OF THE PRINCIPAL

Schools as non-rational organizations

Weber (1947) described organizations as rational bureaucracies in
which administrators exercise control based on access to superior
technical knowledge. Authority was seen as deriving from the notion
that organizational norms may legitimately be established and members
of the organization legitimately be required to obey the rules and
conventions established in order for organizational goals to be
reached. Rational bureaucracies were presented as impersonal in

nature, dominated by a spirit of formality.

Corwin (1974) summarized this approach as one in which
organizations are seen to "serve as means for realizing group
goals". Under the assumptions of this model, he suggested, "any
departures from rationality can be attributed to rational mistakes,
ignorance, or miscalculation" (p. 253). This model assumes that
organizations have clear goals and that administrators will have the
information required to make the decisions needed to achieve these
goals. This rational model dominated research into schools as
organizations throughout the 505 and 60s and continues to exert a
wide influence, particularly in those models which equate management
with decision-making (see, for example, Griffiths, 1958; Simon,

1971). The rational model is also receiving renewed attention

25
because of the flurry of interest in creating computer simulations of

educational decision—making processes (c.f. Wholeben, 1985).

It has long been recognized, however, that this model of
organizations as rational bureaucracies represents a considerable
oversimplification (see, for example, Udy, 1959). Researchers in
recent years have increasingly challenged its applicability to
schools, proposing a number of alternative models, most of which may
be characterized as viewing schools as essentially non-rational
organizations (i.e., organizations for which the ideal-type Weberian
rational bureaucracy model is inadequate). Already in 1947, Parsons
saw a conflict between the roles of technical expertise and hierarchy
in Weber's model. More recently, Hanson (1975, 1976) has pursued
this question, positing the existence of two interacting spheres-~one
in which the technical experts (i.e., the teachers) exercise their
autonomy in classroom management and curriculum and one in which the
administrators wield their power through the hierarchical structuring

of the school organization.

It was Bidwell (1965), however, who moved the debate in a
substantively different direction with his seminal work, The School
as a Formal Organization. This chapter, which presents for the first
time the concept of "structural looseness", anticipates the later
work of Karl Weick (1976) and others. According to Bidwell, schools
differ from other organizations primarily in the high degree of
autonomy which is granted both to teachers and to the administrative

personnel. This phenomenon has altered the structure of school

26
systems, lessening the degree of control which administrators
maintain over the outputs and countering bureaucratic mechanisms of
control. The ultimate effect of this tendency toward a more
professional basis for teaching (and indeed, administration),
suggested Bidwell, is that coordination of school activities becomes
highly problematic and schools become characterized by structural

looseness.

While Bidwell recognized the inadequacy of the bureaucratic model
to fully explain the organization of the school, he nevertheless
continued to view schools as stable organizations with strong and
readily defined boundaries. The 'garbage can model' (Cohen, March &
Olsen, 1972; March & Olsen, 1979), on the other hand, rejected the
rational bureaucratic model altogether and proposed a model based on
'organized anarchy'. In this model, the way in which administrators
make choices was explained by the metaphor of a garbage can into
which various problems and solutions are tossed by various
participants at various times.

It is clear that the garbage can process does not do a
particularly good job of resolving problems. But it
does enable choices to be made and problems sometimes
to be resolved even when the organization is plagued
with goal ambiguity and conflict, with poorly
understood problems that wander in and out of the
system, with a variable environment, and with decision
makers who may have other things on their minds."
(March & Olsen, 1979, p. 36)
The 'garbage can model' was extended by Weick (1976) in what has

become the dominant organizational model of schooling up to the

present time. Weick presented schools as loosely coupled

27
organizations. In this perspective, schools may exhibit many of the
formal properties of the Weberian rational bureaucracy, with a
hierarchical structure, a strict division of labor and a
preponderance of formalized procedures, but there is only a weak link
between the activities of administrators and ultimate student
outcomes. This model stresses the existence of stable subassemblies
which maintain their own identity and are only marginally controlled
by administrators (for a more extended discussion of loose coupling,

see Gamoran & Dreeben [1985] and Tyler [1985]).

While some researchers have suggested that elementary schools can
be broadly characterized by the rational bureaucracy model and
secondary schools by the anarchy model (Firestone & Herriott, 1982),
most recent organizational theorists have questioned the adequacy of
the rational bureaucracy model for any educational setting, remaining
much closer to Weick than to Weber. Bidwell (1977) himself has
turned away from the concept of the school as a closed system and now
places a much greater emphasis on environmental influences on
schools. He suggests that schools are indeed loosely coupled because
of the diffuse nature of the technology of instruction. He does not,
however, understand loose coupling in the same sense as Weick.
Unlike Weick, he believes that while administrative activities and
roles are strongly responsive to surrounding events, instructional
activities and roles are also strongly responsive to surrounding
events; the difficulty he perceives is that administrative

activities and instructional activities are not surrounded by the

28
same events and hence there can exist only a loose coupling between

administration and instruction.

Other recent theories have dealt with the ways in which school
administrators and policy makers continue to be guided by the
rational bureaucratic model, to the detriment of the school
organization and the accomplishment of school objectives. Wise
(1979) describes schools as characterized by hyperrationalization.
This means that procedures and rules are imposed on the basis of
apparent logic, but do not have the effect of achieving conformity to
organizational norms. Meyer and Rowan (1978, 1983) present a similar
phenomenon, which they term institutionalized ritual, as
characteristic of schools. They see schools as responding to the
confidence which the public places in the rites of certification
rather than as organizations characterized by a meaningful
relationship between their proclaimed goals and their instructional

practices.

Yet other theorists have described the school organization as a
place where conflicting objectives place the various actors into what
are frequently contradictory roles. Powell et a1. (1985) present the
image of a shopping mall to describe what goes on in high schools,
with schools providing opportunities for students but leaving them
with the sole responsibility for choice and involvement. Principals
are generally treated as irrelevant to the learning process since
teachers and students arrive at understandings ("treaties") as to

what must be accomplished in order to graduate. This theme of tacit

29
understandings between teachers and students is further expanded by
Sedlak et al. (1986) who argue that many organizational policies have
had the unintended consequence of perpetuating the "bargain" to avoid
or minimize academic content, and that the ability of individual
schools to create a learning environment is severely affected by
social, political and financial forces beyond their control. Because
of this, efforts by individual principals to increase content and
tighten the organization are seen as relatively incapable of bringing

about meaningful change.

It is against this backdrop of theories which dominate the
current research literature that the emergence of a distinct body of
literature known as the effective schools research has been hailed by
many as a welcome affirmation of the ability of individual
practitioners and individual schools to effect significant change.
The effective schools research holds, in general, that significant
between-school differences in both equity and quality of schooling do
exist, that these differences are independent of funding levels, and
that effective schools can and do find ways of addressing both the
equity and the quality concerns. This literature holds further that
factors leading to a greater degree of success in these two areas can
be identified and duplicated by other schools. Emphasis is placed on
the instructional leadership role of the principal, who is seen as
the key player in the process of innovation and change (c.f. Lezotte
et a1., 1980). The development of the effective schools movement and
its relationship to prevailing theories of schools as organizations

is discussed below.

30
Effective sghools research
The beginnings of the effective schools research can be traced to
the early 1970's. The first studies came largely as a response to
the widely-circulated Coleman Report (Coleman et a1., 1966) and the

later Jencks report (Jencks et a1., 1972).

The Coleman Report was essentially an input-output study and
suggested that differences in school inputs do not lead to
differences in school outcomes.

Schools bring little influence to bear on a child's
achievement that is independent of his background and
general social context...lnequalities imposed on
children by their home, neighborhood and peer
environment are carried along to become the
inequalities with which they confront adult life at the
end of school. p. 325

The implication is clear: it does not really matter what
teachers and administrators do since it is the ethnic and

socioeconomic mix of students which ultimately determines

achievement.

The findings of the Coleman Report have been widely challenged on
the basis of flawed methodology and design, as well as a tendency to
extrapolate beyond the data. (Early critics include Armour, 1972;
Bowles & Levin, 1968; Cain & Watts, 1970; Hanushek & Kain, 1972: and
many others. More recent critics include Raudenbush & Bryk [1986].)
In spite of the critics, however, the basic premise of the Coleman
Report--that differences among schools account for a minimal

proportion of the variation in student achievement and that

31
socioeconomic background is all-important--continued to gain currency
in many quarters throughout the early 1970's and indeed, continues to
wield a considerable influence right up to the present time. (See

Edmonds, 1978, for one perspective on why this is so.)

G. Weber (1971) was one of the first to directly attempt to show,
in contrast to the Coleman findings, that schools could indeed result
in high achievement for students from low socioeconomic backgrounds.
Using an outlier (i.e. schools 'lying out' further from the mean than
would normally be anticipated) design, he identified, among others,
five constructs which appeared to account for this: strong principal
leadership, high expectations, an orderly atmosphere, a focus on
pphievement, and pegular testing. The Weber study, though marred by
numerous methodological and conceptual flaws, is noteworthy because
of the continued importance of the major constructs identified.
These soon became the 'big five' of the effectiveness movement and

continue to appear regularly throughout the literature.

A second study which followed the outlier design used by Weber
was published in 1974 by the Office of Education Performance Review
of New York. In this study, an effective inner-city school was
compared to one which was not effective. The purpose of this case
study was to determine which factors appeared to account for the
differences in student outcomes for the two schools. The results are

given here as typical of much of the early research in this field:

32

the differences in student performance in these
two schools seemed to be attributed to factors under
the schools' control;

administrative behavior, policies, and practices
in the schools appeared to have a significant impact on
school effectiveness;

the more effective inner-city school was led by an
administrative team which provided a good balance
between both management and instructional skills;

the administrative team in the more effective
school had developed a plan for dealing with the
reading problem and had implemented the plan throughout
the school

classroom reading instruction did not appear to
differ between the two schools since classroom teachers
in both schools had problems in teaching reading and
assessing pupils' reading skills;

many professional personnel in the less effective
school attributed children's reading problems to
nonschool factors and were pessimistic about their
ability to have an impact, creating an environment in
which children failed because they were not expected to
succeed. However, in the more effective school,
teachers were less skeptical about their ability to
have an impact on children;

children responded to unstimulating learning
experiences predictably-- they were apathetic,
disruptive or absent. p. vi, vii

The first methodologically rigorous, large-scale attempt to find
effective schools was described in 1974 by Klitgaard and Hall. They
defined effectiveness operationally as student performance on
standardized reading and mathematics achievement tests. They
analyzed 1969-70 and 1970-71 scores from Michigan 4th and 7th Grade
students (representing ninety percent of the public schools in the
state), 1960 high school data from Project Talent, and scores from
Grades 2 through 6 from 1967-71 in New York City. They then checked
to see if more schools were one standard deviation above the mean
than would be expected by chance. Fifteen of 161 Michigan schools

were above the mean six out of eight times (less than one would have

33
been expected by chance). They also investigated school districts in
New York City and found 30 of 627 were one standard deviation above
the mean five of eight times (less than four would have been expected

by chance).

This study showed that certain schools do attain a greater degree
of success than others in promoting student achievement. Since it
compares schools with similar populations, it can further be assumed
that there are schooling variables which somehow account for these
differences. What those variables are, however, is not addressed.
Further, the study offers no way to determine whether or not there
are groups within schools which are achieving success while others
are not. It gives only average achievement data for the schools.
Finally, it tends to confirm rather then disconfirm the Coleman
finding that school differences account for only a small part of the
overall differences in student achievement, since only a small
percentage of schools were found to be "effective" (an average of

about five percent).

A number of outlier studies were conducted throughout the
1970's. Madden, Lawson and Sweet (1976) matched 42 elementary
schools in California on the basis of pupil characteristics, with one
high-achieving and one low-achieving school in each of 21 pairs.
Brookover and Lezotte (1977) looked at six improving and two
declining schools. Each of these studies was cross-sectional in

nature and described findings similar to the 'big five' of Weber.

34
One of the more rigorous outlier studies was conducted by
Brookover et a1. (1979). The researchers studied Michigan
fourth-grade and fifth-grade students at 61 majority-white and 30
majority-black schools. School input variables studied included the
social composition of the student body, the school social structure
and school climate. Outcome variables studied included student

achievement, self-concept of academic ability and self-reliance.

The five factors defining school social structure were parent

involvement, differentiation of student programs, openness of
classroom organization, principal's time allocation and staff
satisfaction with school structure. School climate was defined as

the norms, expectations and beliefs of those involved in the school.

Academic achievement was measured according to mastery of reading
and math objectives (from the Michigan School Assessment Test).
Self-concept of academic ability and self-reliance were measured
according to questionnaires specially developed for this purpose by

the researchers.

Brookover and his colleagues provided a very complex view of how
effective schools operate. 0f key significance to their work was the
concept of a school social system composed of three elements: social
inputs, social structure, and social climate. Social inputs they
defined as student body composition and personnel characteristics;
social structure referred to class size, open classrooms, etc

.; while

social climate was defined as the norms and expectations about the

35
school held by staff and students. They found school social climate
(and, to a lesser degree, school social structure) significantly
related to student achievement, even when socioeconomic status was

held constant.

Perhaps the best known of the effective schools researchers
during this period was Ronald Edmonds. Working first in New York
City and then with Brookover and Lezotte at Michigan State
University, Edmonds soon became a leading spokesman for the effective
schools movement. He argued vigorously that "all schools be held
responsible for effectively teaching basic school skills to all
children". (Edmonds, 1978; p. 31). He held further that "we can,
whenever and wherever we choose, successfully teach all children
whose schooling is of interest to us...Whether or not we do it must
finally depend on how we feel about the fact that we haven't so far."

(Edmonds, 1978; p. 35)

One of the most important studies which can be placed in the
general framework of the effective schools tradition is the study of
inner city secondary schools in London conducted by Rutter et a1.
(1979). The major findings of this study, which was longitudinal in
nature, include the following: 1) secondary schools differed
markedly as to the level of student outcomes; 2) student intake did
not account entirely for outcome differences between schools; 3)
variations in outcomes between schools were relatively stable over a
period of five years; 4) schools whose students exhibited better

levels of behavior also exhibited higher achievement; 5) differences

36
were not due to physical factors such as school size and building
quality; 6) differences between schools were related to their
characteristics as social institutions; 7) factors outside the
teachers' immediate control also influenced outcomes; 8) academic
balance in students admitted was related to a decrease in
delinquency; 9) there was a stronger relationship between outcomes
and the combined measure of overall school process than with any
individual process variables; 10) the association between school
processes and outcomes appears to be at least partially causal in

nature .

A number of objections have been raised to the effective schools
studies. One of the most important concerns the outlier design which
many of these studies have employed (Purkey & Smith, 1983; Rowan et
a1., 1983). Follow-up studies have frequently failed to show
consistent gains in achievement over a number of years at schools
identified as effective for the purposes of one study or another. In
other words, the statistically predictable pattern of regression
toward the mean has occurred in virtually every sample of outlier
schools studied. Weber (1971), for instance, found upon return to
his original schools several years later that only one of them

continued to be effective.

The failure to show stability of school effects is a more serious
problem than has generally been acknowledged, for it means that
long-term educational planning based on the effective schools

research is not empirically defensible at present. Although this has

37
not stopped educational planners in this country from moving forward
with school improvement projects based frequently on intuition (for a
defense of this position, see Lezotte, 1986), educational planners in
developing countries are unlikely to embrace the effective schools
concepts with the same enthusiasm. In the wake of the failure of
vocationalization and other expensive approaches to live up to their
seeming promise (Foster, 1965; Urevbu, 1984; Chapman & Windham, 1985;
Psacharopoulos, 1986), there is a growing skepticism of Western
educational models and a concern about being used as little more than
a research laboratory for unproven and frequently unworkable

educational theories (Salmi, 1984; see also Kagia, 1986).

A further difficulty of the effective schools studies has been a
failure to control adequately for differences in student intake.
This means that some of the variation in school quality observed may
have been due to the composition of the student body, rather than to
any differences in effects which the school itself may have had.
Ralph and Fennessey (1983) discuss two important studies in which

student intake appears to have been inadequately controlled.

An additional methodological problem is that there has been no
real attempt to demonstrate causality. This has commonly been
precluded by the cross-sectional nature of the studies, which have
shown only the level of correlation between variables. This leaves
unanswered questions such as whether high achievement may lead to a

stronger leadership role for principals rather than the other way

around.

38

Perhaps of even more concern is that the effective schools
movement has generally adopted a very narrow definition of success
(c.f. Rowan et a1., 1983). Edmonds defined it in terms of minimum
competency in basic skills as measured by scores on standardized
tests of mathematics and reading. Weber and the 1974 study by the
Office of Education Performance Review defined it in terms of scores
above the norm on standardized reading tests. These narrow
definitions have helped educators bring into focus what had
frequently become very diffuse and unfocused programs, yet virtually
everyone acknowledges that the role of schools is to accomplish far

more than these definitions would indicate.

The strength of current school effectiveness research
is perhaps also its greatest weakness. The focus on
one outcome dimension (student achievement) and the
measure of that outcome through standardized test
scores in the area of basic skills promotes a
unidimensional view of school effectiveness. (Dede &
Freiberg [1986], p. 78)

Unfortunately, while there is general agreement that an attempt
must be made to take into account the multiple goals of schooling
(Purkey & Smith, 1983; Cuban, 1984; Good & Brophy, 1985; Rowan et
a1., 1982), there is no consensus on how this is to be done, or on

what the relative importance of these multiple goals really is.

In spite of methodological difficulties, however, the effective
schools studies display a great deal of face validity and have proven
to be remarkably consistent in nature (Purkey & Smith, 1983). In

recent years, attempts have been made to overcome many of the

39
methodological shortcomings and the Rutter study in particular, is
able to make meaningful causal inferences because of this
methodological rigor. This literature has done much to increase our
understanding of schools as social organizations with the potential
for a high degree of collaboration on shared goals and objectives and
the establishment of shared norms and a shared culture. While
recognizing the complexity of school processes and the role of school
culture in shaping the learning opportunities of students, it has
placed an emphasis on individual schools as the focus of change, thus
presenting a bottom-up approach to change consistent with the
literature on innovation and implementation. Berman and McLaughlin
(1978), for instance, conclude in their study of federal funding for
educational change that the success of innovations depends ultimately
on the quality and commitment of local staff. Elmore (1978, 1984)
argues persuasively that solutions require forming consistent
expectations among people who have a stake in the outcome. His
”backward mapping" approach underlines the importance of
participation by those specific groups ultimately responsible for
implementing policies. In the same vein, Purkey and Smith (1985)
argue that effective policies are those which alter behavior at the
delivery level (i.e., the school) and that this is best done from a
bottom-up perspective. Goodlad (1984) adds his voice to the chorus,
calling for site-based school improvement. The conclusion of these
researchers is that local responsibility and sense of ownership must

be maximized.

40
It is in this sense that the effective schools research offers
the possibility of change in keeping with the findings (although not
necessarily the conclusions) of current theories of schools as
non-rational organizations. Loose coupling theories support the
bottom-up approach, holding that top-down directives cannot control

such critical factors as school climate and school processes.

As Elmore indicates, this perspective allows for the
identification of the multiple objectives which exist at the
implementation level (i.e., individual schools and classrooms) and
for the development of strategies sensitive to the complexity of
trading among these objectives. This approach also holds the
potential for addressing the concerns of those who see schools as
places where teachers and students, from the vantage point of
differing objectives, strike various kinds of bargains in order to
satisfy these objectives (Powell et a1., 1985; Sedlak et a1., 1986).
The establishment of group norms based on consensus about valued
outcomes can go a long way toward meeting these concerns, as can a
recognition of and a sensitivity to the multiple objectives of

schooling.

The effective schools movement has moved towards a recognition of
the multiple objectives of schooling and away from a narrow academic
definition of effectiveness in recent years (Lezotte, 1985). This
evolution is in keeping with the findings of the non-rational
theories of schooling. One area in which conclusions of the

effective schools researchers differ consistently from the

41
conclusions of the non-rational theorists, however, has been that of
the role of the principal. The single theme which resounds most
loudly throughout the effective schools literature is that individual
principals do make a difference--that they control enough of the
parameters of the learning process through their role as

instructional leader to significantly influence student outcomes.

Principals have an important role in the school
generally, and they have an especially important role
in clarifying and changing a school's learning climate
and the resulting instructional effectiveness.
Principals represent the organizational authority of
the school and in that regard, they symbolize what the
school stands for, how it will operate, and what is
important. The research on effective schools,
effective educational innovations, and effective
strategies for planning change all point to the
principal as a singularly important person in the
successful school system.
(Lezotte et a1., 1980, p. 93)

It is to this role that the present review turns next.

5 ruct na eadershi role of the rinci a
The principal's role in promoting instructional improvement is
identified time and again in the current literature as the key to
school improvement. Wellisch et a1. (1978) found that administrators

in schools where achievement was on the upswing l) were more

concerned with instruction; 2) communicated their views about
instruction; 3) took responsibility for decision-making about
instruction; and 4) coordinated instructional programs through

regular discussing and reviewing of teacher performance and

emphasizing academic standards. Other studies supporting strong

42
principal roles in instructional leadership include Edmonds (1978),
Irvine (1979), McLaughlin & Marsh (1978), Squires (1980), Goodlad

(1984), Young (1986), and a host of others.

Research findings reported in the literature indicate that the
influence of the principal can be both direct and indirect.
Principals may exert an indirect influence by buffering the
instructional process from discipline-related disruption (Thompson,
1967; Bossert, et a1. 1982; Cohen, 1983), by shielding teachers from
criticism through support of innovative projects (California State
Department of Education, 1977); and by protecting the learning
process from community opposition and the dysfunctional consequences
of community apathy by fostering strong community support and

involvement on an ongoing basis. (cf. Dove, 1980)

Principals do more than act as buffers, however. They may also
exert a direct influence on learning through a number of different
approaches. First, principals may be the crucial players in the
establishment of a school climate in which teachers are encouraged
and indeed, enabled, to give priority to instructional goals
(Brookover, et a1., 1979; Rowan et a1., 1983; Berman & McLaughlin,
1978). This is accomplished by a variety of means, including symbol
management (Manasse, 1985), defining a fair and effective evaluation
system (McAndrew, 1981), and working through the bureaucratic and
cultural linkages that govern teacher behavior (Firestone & Wilson,
1985). Principals may also influence student achievement by direct

attention to the school curriculum and its relationship to student

43
outcomes--influencing what is and what is not emphasized (c.f. Spady,
1982, for a discussion of outcome-based instructional management
using this approach; see also Bloom, 1980). Finally, principals
affect student achievement by supervising and working to improve
teacher instruction and methodology. Martin and Willower (1981)
found that secondary school principals spent 17.4 percent of their
time working directly or indirectly with teachers for instructional
improvement (classroom visits, etc.). Bloom has suggested that
effective and ineffective principals be compared on this sort of
measure to determine to what extent instructional time-on-task may

affect student outcomes.

Collaboration or prescription?

The ways in which an effective principal can and should exert
direct influence on learning have remained a matter of considerable
controversy. Hallinger and Murphy (1985) suggest that effective
principals in lower SES schools are much more directive and
prescriptive than effective principals in higher SES schools. Cohen
(1981) wonders if unusually effective schools "are more tightly
managed and more collectively committed to basic skills
instruction". High and Achilles (1986) state that "principals should
involve teachers only on important issues and should rely more on
their legitimate authority" (p. 19). They see this authority as
deriving in large part from the principal's instructional leadership
expertise. Rutter et al. (1979) argue for a tight control by

administrators of school processes.

44
There are, of course, a number of arguments which can and have
been advanced in support of a more directive role for principals,
particularly in developing countries. Teachers in these countries
frequently do not have independent access to either the information
or the materials needed for optimal professional growth. A solid

educational background upon which to build is frequently missing.

Windham (1985) has suggested that with the overwhelming set of
difficulties facing the teaching profession in sub-Saharan Africa,
the most promising approach is some form of programmed teaching which
places the important curriculum decisions in the hands of the
experts. He sees this as one way to overcome teacher deficiencies in
short order, contributing to the teacher's ultimate growth while at
the same time providing the policymaking branch of government with a

means of ensuring that its goals and objectives are being met.

This may be an approach, however, which locks teachers in
developing countries into a position of increasing dependency on
external materials and imported technology (not to mention
high-priced Western consultants). Most researchers and practitioners
place a high value on teacher autonomy in the classroom and have
consistently questioned the appropriateness of any activities on the
part of principals which would tend to diminish that autonomy.
McLaughlin and Marsh (1978), reporting on findings of the Rand change
agent study, conclude that "principals need to give clear messages
that teachers may take responsibility for their own professional

growth" (p. 92). Hurst (1981), using Third World examples, argues

45

that augmented, rather than diminished, teacher participation is
essential. He suggests that teachers can and will respond to
innovation if they perceive it to be appropriate. He calls for a
much more collaborative approach, with administrators working with
teachers in order to strengthen the institutionalization of
innovations. Erickson (1984) supports this position, arguing
forcefully from an anthropological perspective that teacher
involvement in all aspects of the learning process is the sine gua

ppp of success. He concludes that the prevailing tendency to control
the ways teachers teach restricts students' access to knowledge and
creates a separation between what goes on at the school and what goes

on in the society in which the child lives.

Firestone and Wilson (1985) describe the principal's role in this
context as one of establishing bureaucratic and cultural linkages
with teachers in order to improve instruction. Bureaucratic
linkages, such as supervision, procedures and so on, "establish
constraints on and opportunities for how teachers teach". Cultural
linkages refer to the organizational ethos--shared beliefs and
meanings. The principal may use these cultural linkages to influence

how teachers think about what activities they should undertake.

The heterogeneity of teachers as a group points up the need for
consensus as to instructional goals and it may well be that effective
principals learn to adjust their administrative style and the use
they make of these linkages according to the particular needs and

abilities of the individual teachers with whom they work. Ideally,

46
this will be achieved within a framework which respects the autonomy
teachers need for growth, avoiding a mechanistic view of teachers and

the teaching profession.

Se -esteem

This section does not purport to be a comprehensive overview of
the research relating to self-esteem. Such a task, if possible at
all, would be far beyond the scope of the present study. The purpose
of this section is rather to explain the inclusion of self-esteem as
a dependent variable in the study and to show the major emphases of

recent research in this area.

Elmore (1984) and many others have discussed the nature of
choosing among objectives as a process of trade-offs. Focusing on
one objective implies that less attention will be given to another.
This raises the possibility that valued outcomes will not be achieved
if not given priority, or at least explicit attention. It also means
that the price to be paid for gains in one area may be that of
accepting loss or at least no gain in another area. This is
frequently the choice which policymakers in developing countries must

face in terms of equity and quality (c.f. Behrman & Birdsall, 1983).

Recognizing this dilemma, a number of educators have expressed
concern that the exclusive attention often accorded to academic
achievement in the effective schools research may lead to losses in
terms of other valued student outcomes. Self-esteem is one of these

"other" valued student outcomes seen as meriting particular attention

47
(Rowan et a1., 1983; Purkey & Smith, 1983). The value which is
generally accorded to self-esteem by educators can be seen by the
high level of continuing research interest in the topic, yet there
has been little attempt to see in what ways a deliberate focus on
academic achievement affects the development of student self-esteem.
Educators have generally assumed automatically that achievement is
strongly related to higher self-regard, yet the literature on
self-esteem does not provide a clear verdict about this. Most
studies do appear to find a positive relationship, but there is a
remarkable variation in the strength of the relationships found,
depending often on the particular instrument used. The few African
studies which have been conducted also report positive findings.
Mwaniki (1973), for instance, found a significant positive
relationship between self-concept and academic achievement in a
sample of Kenyan students, but only for a restricted subset of the
components measured (mental ability and school subjects). His
results and the results of many similar studies must be considered as
ambiguous at best. Perhaps one of the most remarkable aspects of the
self-esteem research is that in spite of the tremendous body of
literature, we really do not know what effect schooling in general

has on student self-esteem.

Morse (1964) was one of the first to ask whether or not schools
did in fact lead to the development of self-esteem. Using the
Coopersmith Self Esteem Inventories, he found that for many of the
items, the longer the students remained in school, the lower their

level of self-esteem. He reported that "whatever else we have done,

48
we have communicated a sense of personal failure to many of our
pupils" (page 27; c.f. Glasser, 1969). More recently, Wylie (1979)
has questioned the relationship between self-concept and achievement,
suggesting a number of reasons for which this relationship may not be

as obvious as had been previously believed.

In an attempt to find answers in the apparently contradictory

research literature, Hansford and Hattie (1982) conducted a
meta-analysis on the relationship between achievement and
self-concept. They included studies representing a total sample size

of 202,823 persons. Over three-fourths of the correlations reported

in the studies came from the U.S. literature.

Meta-analysis is an inexact science at best but that of Hansford
and Hattie provides an intelligent discussion of its own weaknesses
and does not claim to provide the final word. While cautiously
reporting the finding of a low, positive correlation (averaging
approximately .21), they point out that higher quality studies which
report reliability coefficients tend to report relatively low
correlations. The relationship between achievement test reliability

and correlation was -.32.

One reason that the relationship between achievement and
self-esteem appears somewhat ambiguous may be that the process of
achieving is as important to the development of self-esteem as the
end product of that process. It may be also that certain modes of

learning do not lead to increased self-esteem while others do.

49
Nimnicht et a1. (1969) explain the concept of the "responsive
environment" as a way to encourage the development of self-esteem.
This is an environment which accords a great deal of autonomy to
children, within a general framework of reasonable limitations
established by the school. It may be that other learning
environments lead to cognitive improvement without contributing to

the development of self-esteem.

The question which surfaces from the above discussion is whether
instructional leadership promotes the sort of climate in which
self-esteem will flourish. As has already been noted, some
researchers (Hallinger & Murphy, 1985) have suggested that effective
principals in lower socioeconomic environments tend to be more
directive and prescriptive. Many researchers report that children in
lower SES environments appear to exhibit lower self-esteem (see Rio,
1979, for a review of this literature). If this is true, it may be
that techniques used to promote effectiveness for these students,
when effectiveness is defined in purely academic terms, may actually

lower effectiveness when measures of self-worth are included.

50

Part II

EDUCATION IN ZAIRE

lBELQQEQELQD

The development of education in Zaire has been greatly influenced
by the turbulent history of the country over the past century. From
1885 until 1908, it existed as the personal fiefdom of King Leopold
II of Belgium. The levels of cruelty and exploitation attained by
his agents during this period attracted increasingly wide attention
until finally the other colonial powers, in spite of their remarkably
high tolerance for this sort of thing, were compelled to take
action. Le0pold was forced to relinquish his personal claim to the
Congo Free State, which then became a Belgian colony known as the

Belgian Congo.

The Belgians, who themselves possessed almost no natural
resources of any importance, soon established the Shaba (former
Katanga) region as a major producer of copper and other minerals.
Important supplies of industrial and gem-quality diamonds were also

found in the Kasai region of the country.

With the emphasis on mining a few select areas, little attention
was given by the Belgians to developing the infrastructure of the
Country. Few roads were built and only the most rudimentary system

of communication was installed.

51

The work of education in the Congo was largely left to the
religious missions, with the Catholic church receiving the lion's
share of such financial assistance as was available. Emphasis was
placed on primary education to the virtual exclusion of all further
schooling. Incredibly, by the time of independence, from a population
of some 14 million, only sixteen individuals had graduated from
college in the country (this total does not include Catholic

seminarians).

The colonial patterns of education still have important
ramifications for Zaire today. Although secondary and post-secondary
education have expanded rapidly since independence in 1960, the lack
of qualified personnel is a persistent problem in Zairian schools.
Many primary schools in the interior cannot count a single high
school graduate among the teaching staff and most secondary schools
have few, if any, college graduates teaching. Typically, most of the
upper secondary school classes are taught by those with 'graduat'
status (two or three years of college). The lower secondary school
grades are frequently taught by secondary school graduates with no
post-secondary training. There is no training program for

principals.

The conditions of schooling in the Kivu region, like those in
most of rural Zaire, are generally very rudimentary. Primary
students use cardboard 'slates' upon which they write either with a
small stylus or a chalk-like rock. Mud-and-stick buildings with

thatched roofs are common. There is, however, a wide variation in

52
school building quality, often on the basis of school sponsorship

(Youdi, 1972).

Schools in Zaire can be placed in five general categories on the
basis of sponsorship: Catholic, Protestant, Kimbanguist, Moslem and
government. (The Kimbanguist church is an indigenous church founded
by Simon Kimbangu and enjoying special official recognition within
Zaire.) The Catholic church sponsors the largest educational system
while various Protestant school systems tend to be regionally
important. The government of Zaire does not recognize the Protestant
organizations individually but has created an umbrella organization,
the Church of Christ in Zaire (generally referred to by the acronym

ECZ--Eglise du Christ au Zaire.)

The Zairian contexp

Zaire is a country of some 2.3 million square kilometers and as
many as 250 different ethnic groups, most with their own language.
It has Africa's most important system of inland waterways, although
navigation is seriously hampered by a number of rapids and falls.
These rapids and falls give Zaire the greatest hydroelectric
potential in the world, more than one billion kilowatts. The huge
Inga-Shaba project, recently completed, is one of the largest in the
world, yet only a relatively small portion of the country will share
in the electricity generated. Before Inga-Shaba was completed, the
Shaba province with its mining industry consumed about 75 percent of
the electricity produced, the region around the capital city

(Kinshasa) consumed about 23 percent and the rest of the country

53
competed for the remaining 2 percent (Gran, 1979). With the
exception of the Kasai region, this situation will change very little

as the Inga-Shaba project is implemented.

Zaire has faced a steady economic deterioration for a full decade
and more. The downturn can be dated from mid-1974, although it
resulted in part from the Zairianization measures of 1973 (Gran,
1979). These were designed to remove all foreigners from the
commercial and agricultural sectors of the Zairian economy and to
replace them with nationals. It was done with virtually no advance
planning and resulted in widespread economic disruption. This
worsened when, between April and December of 1974, the price of
copper plummeted from $3370 per metric ton to $1350. At the same
time, the price and production of cobalt and other minerals declined
rapidly. The production of diamonds of gem quality, for instance,
fell from 117 million carats annually at the beginning of the decade

to only 305,000 annually by 1976 (Kaplan, 1979).

Two external factors aggravated the situation. The oil crisis of
1973 resulted in greatly increasing expenditures of foreign exchange
and the Angolan Civil War resulted in the closing of the Benguela
railroad, Zaire's most important route for the exporting of goods and
minerals from Shaba province. The zaire, the official unit of
currency, dropped from an official value of $2.00 in 1976 to $.018 in
June, 1986. Zaire is belatedly making a concerted effort to bring
about economic reform and has been cited recently as a prime example

of a country willing to make the hard decisions necessary for reform,

54
but so far its economic position has not improved. The net outflow
of capital in 1985 and 1986 has exceeded $200 million annually
(Farnsworth, 1986). Widespread mismanagement and corruption, along
with the devastating debt service charges, make any immediate

economic turnaround unlikely.

Much of the country is dependent on subsistence farming for a
livelihood. It has been estimated that upwards of 75% of the
population, following agricultural pursuits, subsists on between $25
and $50 a year. Teacher salaries reflect the economic situation. As
of June, 1986, monthly salaries for teachers who had completed
college were 1173 zaires, or about $20. Monthly salaries for
teachers who only possessed the secondary school diploma were 796
zaires, or about $15. Principals received up to 3396 zaires, or
about $50 monthly. Because of the cost of purchasing food and
clothing, it is not possible for teachers with families to live off
these salaries alone. They must, and do, supplement their salaries

in one way or another.

In 1970, 20% of the annual national budget of Zaire was going for
education. In the boom year of 1973, when copper prices were
skyrocketing, the budget prepared for 1974 included only about 16
percent for education, but as the economy declined the percentage
taken up by education grew rapidly. By 1976, it was back over 21
percent. A year later, it was up to almost 24 percent and by 1978, a
crippling 31 percent (Varon, 1979). This was not because more money

\vas being allotted to education but because the overall budget was

55
shrinking so fast. Funding for education was not close to keeping
pace with inflation. Between 1979 and 1986, the salary of a teacher
with a high school diploma went from 180 zaires to 800 zaires. Had

it kept pace with inflation, it would have gone to over 3000 zaires.

Studies of schooling in Zaire

Because of the tremendous variation in the Zairian context over
the past fifteen years, studies of schooling in Zaire during this
period must be categorized as pre-l974 or post-1974. Studies
conducted before 1974 can certainly cast light on current problems
and possibilities but they must nevertheless be handled with care
because of the fundamental differences in the era which they

represent.

In addition to the economic factors discussed above, the effect
of nationalization on education merits specific attention. In 1974,
the program by which individual church organizations held primary
responsibility for running the schools was abolished. The national
government, on paper at least, assumed control of all schools. In
actual practice, however, many of the rural schools, such as those of
the Kivu province, were not immediately affected by the change.
Within two years, following a period of tremendous confusion, the
churches were again assigned responsibility for administering those
schools which they had earlier sponsored, although the government
continued to pay teacher and administrator salaries. This
arrangement has remained intact until the present time, although a

Ibolicy announced in the spring of 1986 would provide for three

56
categories of schools: 1) entirely directed and funded by the
state; 2) 75 percent state-directed, 25 percent church-directed,
with the churches required to pay 25 percent of salary expenses; and
3) 100 percent privately funded and privately run. Although this
plan was officially scheduled for implementation in September, 1986,
a conversation held in November, 1986 with Zaire's Secretary of
Technical Education indicates that this plan had not yet been

implemented at that date.

A relatively early study of influences on academic achievement in
Zaire presents the effects of school sponsorship as it was practiced
at that time. This study of 25 secondary schools was conducted by
Youdi in 1970 and found school sponsorship to be the most important

determinant of student success.

It is this one characteristic which primarily
determines the internal culture of the institution and upon
which most other factors seem to depend--the quality and
nationality of teachers, the composition of the student body
in terms of quality, religious affiliation and sex of
pupils, the administrative style and teacher-student
interaction, the origin of imported educational equipment,
and even the architecture of school buildings. pp. 53,54

Youdi found that the Protestant students scored higher on tests
of modernity of attitudes, individualism and activism, but that the
Catholic students scored higher on the French and mathematics tests
which he administered. He attributed the variation in outcomes to

the apparently differing "internal culture" of the school systems.

57
Youdi's study provides strong evidence of the unequal access to

_schooling which existed prior to 1974.

A study by Sheline, Papagiannis and Grant (1984), in spite of its
publication date, must be classified as belonging with the earlier
studies since it examines data collected by Youdi in 1974. These
data were based both on national examinations and university entrance
examinations. Unlike Youdi, Sheline et a1. entered school
sponsorship last into their regression analyses. Entering personal
and socio-economic background variables first into the analysis, they
found that these variables accounted for 25 to 28 percent of the
explained variance. Like Youdi, they found that students attending
schools sponsored by the Catholic church tended to achieve greater
academic success, followed closely by the Protestant-sponsored
schools and distantly by the government and Kimbanguist schools.
These differences in achievement level are striking. Catholic
schools were found to average approximately 55 percent on the
national examinations; Protestant and government-sponsored schools
averaged approximately 50 percent, while the Kimbanguist schools
averaged about 44 percent. From 3 to 6.5 percent of explained

variance is found attributable to sponsorship.

Sheline and her colleagues suggest that the sponsorship, or
charter, effect found in this study can be explained in terms of the
differential ability of the schools to control access to future

status. They hold that students in the higher status schools (i.e.,

58
Catholic, then Protestant) believe that they will be accorded higher

status upon graduation and are thus motivated to study harder.

It is contended that the perception of social power has
an effect on the motivation of students, and influences how
well students will do on examinations. The higher the
social power perceived, the more likely students will study
harder since the status rewards are perceived to be higher
and are highly desirable. p. 25

It is unfortunate that this study utilizes data from 1974, a
period of time when the nature of school sponsorship was quite unlike
what it is now (see above discussion of the changes which occurred at
the time of nationalization). In addition, the researchers
acknowledge their difficulty in controlling for all of those factors
which may have been confounded with school sponsorship. While they
cite student IQ, years of teaching experience and expenditure per
pupil as examples of factors which were not included and which could
have an intervening effect, it seems more reasonable to be concerned
about the absence in this study of any discussion of the possible
importance of variables such as teaching methodology, principals'
leadership emphases, discipline patterns, time on task and other
process variables which could be confounded with school sponsorship.
Of more interest to the current study than the findings relating to
sponsorship is the finding that a relatively high percentage of
variation in student achievement levels is attributable to

socioeconomic status.

59

The question of socioeconomic status and its effect on schooling
in developing countries is currently the center of a great deal of
spirited debate. Heyneman (1976, 1980), Currie (1979) and Heyneman
and Loxley (1982) have argued that socioeconomic factors have only a\
minimal effect on learning in the sub-Saharan African countries.
They hold that schooling, on the other hand, accounts for a
dramatically higher percentage of learning in sub-Saharan Africa than
it does in most developing countries. Simmons and Alexander
disagree, as do Cooksey (1981) and Lanzas and Kingston (1981).
Cooksey suggests that much of the confusion is due to selection
effects. In other words, the lower SES students are generally
eliminated early from the process of schooling which means that those
who manage to remain in school are of above-average ability. He
concludes that "levels of parental education and the use of French in
the home were closely related to class background and are almost
invariably associated with performance variations" (p. 417). His
results, however, indicate that the effect of parental education is

negligeable except for the upper secondary levels.

Lanzas and Kingston (1981) offer a different perspective as to
the relationship between socioeconomic status and achievement. They
suggest that the problem is a lack of research attention to the
different nature of family status in sub-Saharan Africa. Their study
of English achievement in Zaire argues that children of the same
father may have quite different statuses. The residential disruption
‘which occurs when students are forced by circumstances to move in

‘with various relatives is seen as having a significant impact on

60
learning. They argue that this finding means that a more
sophisticated measure of socioeconomic status is needed. Test scores
on English performance are found related to family status only among

the residentially stable students.

The Lanzas and Kingston study appears to overemphasize the degree
to which polygamy exists in much of Zaire. Their explanation of the
nature of family relationships appears simplistic, as when they state
that "survey questions designed to elicit information about the
paternal father...would have little meaning for students" (p. 432).
It is also unclear to what extent the correlation found for the
residentially stable students (.293) is independent of other effects
and to what extent English scores can be taken as a proxy for
overally academic achievement. The sample size of residentially
stable students in this study is very small (n-104), but the study
does, nevertheless, underline the difficulty of transferring
constructs from one cultural setting to another. It also
demonstrates the need for increasingly sophisticated models of

socioeconomic background.

Several studies in recent years have examined the relative
importance of the educational system in Zaire. A study by Mukweso et
a1. (1984) interviewed 702 male employees of 15 large-scale
enterprises located in Kinshasa and develOped a path model
illustrating the occupational attainment process. The authors
conclude that "the effect of years of schooling on occupational

attainment has remained as powerful and as strikingly constant after

61
independence as it was before. The regression coefficients of .523
(preindependence) and .566 (postindependence) ...have proven to be
remarkably consistent with findings in other nations as well.
Education is an important factor in occupational attainment in Zaire

now, and it was so before independence too." (page 65)

The sample used to arrive at this conclusion was far from
representative and the results reported must be interpreted with
considerable caution. Nevertheless the strength of the correlation
between education and occupational attainment is striking. It is
little wonder that education in Zaire remains highly prized and
continues to be the object of considerable sacrifice, both on the

part of governments and individuals.

In spite of the important role which education in Zaire continues
to play, a number of writers speak despairingly of current trends.
They foresee a failure to maintain current levels and suggest that
there has already been a dramatic dropping in the overall educational
level of the schools. Apumbi-Lomema (1984) presents a typical

analysis:

La situation de l'enseignment primaire n'est pas

 

 

 

brillante. Le niveau est manifestement en baisse

L'explosiop demographigue pree un grave probleme de
scolarisation, (The outlook for primary education is
not good. The level is clearly dropping. The

demographic explosion is creating grave problems for
schooling.) p. 34

In a situation such as that described, where resources are drying

up and the overall level of schooling is dropping, it could be

62
supposed that the literature would be replete with references about
how administrators are seeking to cope with the difficult
circumstances. However, the literature remains silent on this
topic. Indeed, there is no indication that questions of
instructional leadership of any nature have received explicit
attention in studies of Zairian education. Meunier and Ketelslegers
(1985), writing for the Zairian education journal, Educateur, in
1985, discuss the importance of detailed educational objectives;
this article includes a fairly comprehensive presentation of Bloom's
taxonomy of educational objectives. Nothing is said about the role
of the principal in helping teachers focus these objectives on the
instructional goals of the school, however, and there is nothing else
in the literature to suggest that this is generally considered to be

an integral part of the principal's job.

Summa

The image of schools as loosely coupled organizations is
generally accepted throughout the research literature, but there is
considerable debate as to the leadership role of the principal and
the degree to which it is possible, or even desirable, to tighten up
these links, particularly at the secondary level. Within the
effective schools literature and the literature on implementation,
there is wide acceptance of the key role of the principal in defining
and helping achieve a consensus on the objectives of schooling.
There is acknowledgment of the multiple objectives of schooling as

well, but little evidence is presented to show that other than

63
academic objectives are being measured or monitored on a regular

basis.

Equity and quality issues are very much at the forefront of
discussion in developing countries. After a period of rapid
expansion at the primary level, there is concern about whether or not
current levels of funding can be maintained and considerable
discussion about how to expand secondary education without

sacrificing quality. The social demand for education remains high.

In Zaire, tremendous efforts have been made to provide education
throughout the country, but access to education appears unequally
distributed. The economic and political crises which the country has
faced have reduced funding for education but education has
nevertheless maintained its role as an important determinant of
employment. Socioeconomic factors appear to be important
determinants of the quality of schooling which a student will receive
in Zaire and of subsequent outcome levels. Questions of management
and efficiency have received little attention in the Zairian research

literature.

CHAPTER THREE

METHODOLOGY

I spend half of my time at each school. Every
Tuesday and Thursday, I leave home at 4:00 in the
morning and I walk until 8:00 or 8:30 to get to
(School 5). I head back home again at 4:00 and
get in around 7:30 ...It's nothing, really...Where
I was principal before, I used to walk 101
kilometers once a month to pick up the salaries.
-principal, School 5

Introduction

Educational researchers in Zaire, like the educators themselves,
face a number of frustrating constraints. Many school sites are
remote and transportation is always a problem. Respondents are often
difficult to locate since, as the above illustration indicates, there
are frequently extraordinary demands made upon their time and

energies. This is particularly true for the principals.

While these constraints undoubtedly give the researcher a more
complete understanding of the context in which schooling occurs, they
also pose unique methodological difficulties. These difficulties and
others of a more general nature are dealt with in this chapter, which
is divided into five sections: 1) site location and characteristics,
2) population and sampling frame, 3) research hypotheses, 4) data

collection instruments, and 5) data collection procedures.

64

65

Site location and characteristics

This study was conducted in the Kivu province of Zaire, the
largest and most populous French-speaking country in Africa. The
schools described are all located in the Butembo district
(spps-division) of north Kivu. This is a district straddling the
equator and characterized by remarkable geographical diversity. It
is bounded to the east by the Virunga National Park and the former
Lake Edward (currently known locally both as Lake Idi Amin Dada and
as Lake Virunga). This lake, together with the snow-clad Ruwenzori
Mountains to the northeast, forms part of the border between Zaire
and Uganda. To the north and northeast, the Butembo district is
bounded by the Ituri Forest and to the south and southeast by the
great savannahs which form the plain of the Rwindi River (see Figure

1-1 on page 7).

The climate depends largely on the altitude, which ranges from a
low of three thousand feet along the lake to a high of sixteen
thousand feet in the Ruwenzori Mountains. Most of the area is
between six and eight thousand feet of altitude. The temperature at
eight thousand feet rarely exceeds seventy degrees during the day and
remains considerably lower when there is rain. The temperature in
some of the more lowlying parts of the district frequently exceeds

ninety degrees during the early part of the afternoon.

66

The entire district is very hilly and heavily cultivated.
Principal crops are bananas, Irish potatoes, sweet potatoes, wheat,
corn, beans, cassava, coffee, pineapples, papaya, and peanuts, but
there is a high degree of local variation according to climate. Well
over ninety per cent of the area's inhabitants are engaged in some
form of traditional agriculture, although there are five rural towns
within the Butembo district. The largest of these, known as Butembo,
has a population of approximately 100,000. It is a trading center
with a busy commercial district running for approximately one mile
along the dirt road which links the provincial capital of Bukavu 250
miles to the south with the city of Kisangani 650 miles to the
north. Thirty-five miles north of Butembo is the town of Beni, an
industrial center with approximately 50,000 inhabitants. The
administrative center of the district is the town of Lubero, with
10,000 inhabitants, while the towns of Kayna and Kanyabayonga, each
with another 7,000 or 8,000 inhabitants, lie to the extreme south and
the smaller town of Oicha lies to the extreme north. The overall
district population is estimated at around 500,000, although there

are no accurate figures available.

The Butembo district is ethnically and linguistically
homogeneous, populated almost entirely by the Nande. (Other than
small communities in some of the larger cities and towns, there is no
significant Nande population living in Zaire outside of the Butembo
district.) With the exception of a few thousand Tutsi and Hutu, ten
to twenty thousand Pere to the northwest and several hundred Mbuti in

the Ituri Forest, there is no significant representation of any other

67
ethnic group. There are no six-year secondary schools serving the
Pere people and only a two-grade mission school serving the Mbuti

(also known as the Pygmies, a term which they consider insulting).

In addition to Nande, Swahili is widely spoken throughout the
district, particularly in the towns. It is the language of most
early primary schooling, although this varies somewhat. French is
the language of the educated. This means that it is spoken much more
widely by men than by women and much more by the younger than by the
older generation. It is the official language of secondary education
throughout the country, although within the Butembo district, a
significant percentage of students in the first several years of
secondary schooling have not yet attained a high degree of fluency.
Surprisingly, although the Konjo people of nearby Uganda and the
Nande form in reality a single ethnic group and speak the same
language, there is virtually no English spoken in the Butembo

district, even along the border.

Po u at on and sam rame

The population for this study consisted of principals, teachers
and students at those schools in the Butembo district where the
principals had assumed their current position no later than
September, 1983. Since the study was performed during May and June
of 1986, this means that for the schools studied, principals were
completing at least their third full school year. Of exactly thirty
secondary schools in the Butembo district offering full six-year

programs, fourteen met this condition. All fourteen of these were

68
studied, as well as a fifteenth for reasons which are explained
later. In each school, at least one Grade 8 and one Grade 12 class

was studied. No other grade levels were included.

The reason for restricting the sample to schools where the
principal had spent at least three years was to masimize the
possibility of measuring the effect of the principals' emphasis on
instructional leadership. It was reasoned that this effect could
hardly be measured in schools where the principal had just arrived,
or had not had time to establish him/herself. Rowan and Denk (1984)
have indicated that the mere fact of changing principals tends to
create an improvement in the academic success of schools
characterized by students with low socioeconomic status. This is
seen for the first two years, after which there is no more effect. A
study of schools which includes new principals might, then, falsely
report this "honeymoon" effect as an effect of the principal's
emphasis on instructional leadership. Further, a carefully
documented study of a school improvement project (Young, 1986)
concludes that positive results from innovations implemented by
school principals take about three years before becoming noticeable;
other references in the literature (Rowan & Denk, 1984; Berman &

McLaughlin, 1978) offer corroboration of this general time framework.

The sampling frame selected for the schools which met this
criterion consisted of the principals, the eighth- and twelfth-grade
students and the eighth- and twelfth-grade mathematics teachers. As

noted previously, informal interviews with assistant principals

69
responsible for supervision= and instruction were also conducted at
several of the larger schools. There were several reasons for
restricting the study to the eighth and twelfth grades. The first
was strictly a question of time and money--it would have been too
costly and too time-consuming to have tested all of the grades in
each of the schools. The second reason was that this frame
corresponds with Populations A and B of the Second International
Study of Mathematics conducted by the International Association for
the Evaluation of Educational Achievement (IEA) and thus permitted
use of an IEA instrument already tested cross-nationally in both
developed and developing countries, including French-speaking Belgium
(the country whose mathematics curriculum has generally formed the

pattern for the Zairian curriculum) and several African countries.

Secondary schools in Zaire are organized on the basis of early
specialization. This means that after the first two years of
secondary schooling (known as the pyplp d'orientation, or orientation
cycle) where the same national curriculum is followed at all schools
throughout the country, students must choose an option with a
specialized program. The schools in the Butembo district, both in
the population and in the sample, offer a total of six different
areas of emphasis (known within the country as sections). These are
math/physics, biology/chemistry, teacher training (pedagogie
ggpgpalg), business, literary, and social work. For purposes of this
study, these have been collapsed into two categories based on the two
different mathematics curricula which are followed. These two

categories are: 1) humanities, with two hours of mathematics each

70
week at the Grade 12 level, and 2) sciences, with five hours of
mathematics each week at the Grade 12 level. The pgppipps which
follow the humanities curriculum are: 1) teacher training, 2)
literary, 3) business, and 4) social work. The sections which follow
the sciences curriculum are: l) math/physics and 2)
biology/chemistry. Four of the secondary schools in the Butembo
district offer two sections; two offer three. Of the schools in the
sample selected, there are none offering three but there are four
offering two. See Table 3-1 for a listing of schools according to
the number of secpiong offered and Table 3-2 for a listing of schools
according to section. Note again that this classification does not
apply to the eighth-grade students, all of whom follow the same

six-hour mathematics curriculum.

Table 3-1
Schools by Number of Academic Programs Offered

# of sectiops Schools in population Schools in sample

 

 

 

 

 

 

 

 

 

 

 

 

1 22 12

2 6 3

3 2, 0

Table 3-2
Academic Programs
Section Schools ip population Schools in sample

Matthhysics 10 2
BioZChem 6 3
Sciences (Total) 16 5
Ipacher training 14 8
Bpsiness 6 3
Literary 3 l
§ppipi Work 1 1
Humanipieg (Iptal) 24 13

 

71

A further distinction which must be made here is that of
sponsorship. Under an arrangement established in early 1977, all
teacher and administrator salaries are paid by the government, but
the responsibility for administration, upkeep and construction is
generally held by any of a number of religious organizations. This
relationship is referred to as 'sponsorship' by a number of writers
who have described it, including Youdi (1972) and Sheline (1979).
There are also a number or schools for which the government assumes
sole responsibility. (See Table 3-3 for a summary of schools in the

population and sample listed by sponsorship.)

The government schools throughout the country have tended to be
characterized by lower quality buildings and high administrator and
staff turnover. This is certainly the case for the
government-sponsored schools (referred to as official schools or as
pppips non-conventionnees within the country) in the Butembo
district. Because of the high turnover rate, there are very few
government-sponsored schools where principals have remained for three
years and, in the sample, none at all in the towns. In order to have
a more representative sampling, one government-sponsored school
located in a town was studied, even though the principal had not been
present for three years. Results from this school are included only
for those comparisons (such as correlations between mathematics
achievement or self-esteem and academic intake, socioeconomic status,
textbook availability and in-town/ out-of—town status) which do not

deal with principals.

72

Table 3-3
Sponsorship of Schools

Sponsorship # of schoola in population # of schools in sample
_gatholic 5
_QEK
CEBK
Government
igiamic
SDA
CECA

r‘h‘h303uéo\¢>
PJC>P‘£‘h*bJ

CBK— Kivu Baptist Community CBK- Kivu Baptist Church Community
SDA- Seventh-day Adventist CECA— Heart of Africa Evangelical
Community

Reaearch hYDothesea
The general hypotheses listed in Chapter One are relisted here
in bold face, immediately followed by a restating in null hypothesis

form.

MAJOR HYPOTHESES

1. The emphasis which principals place on instructional
leadership affects student mathematics achievement.

Hypothesis 1 : There is no significant correlation
between principals' instructional leadership ranking
and student mathematics achievement.

2. The emphasis which principals place on instructional
leadership affects the development of student
self-esteem.

Hypothesis 20: There is no significant correlation
between principals' instructional leadership rankings
and student self-esteem scores.

3. Principals' effect on mathematics achievement is
independent of the academic ability of students
enrolling at the school.

Hypothesis 30: There is not a significant positive
correlation between academic intake and principals'
instructional leadership ranking.

73
MINOR HYPOTHESES

4. The emphasis which principals place on instructional
leadership has an effect on students which grows
stronger the longer the student is at the school.

Hypothesis 40: There is no significant interaction
effect between principals' instructional leadership rating
and student tenure.

5. Students who have textbooks tend to have higher
mathematics achievement.

Hypothesis 50: There is no significant correlation
between possession of a textbook and mathematics
achievement.

6. The student's socioeconomic status does not affect

her/his level of mathematics achievement.

Hypothesis 60: There is no significant correlation
between a student's socioeconomic status and that
student's mathematics achievement.

7. The student's socioeconomic status does not affect
his/her level of self-esteem.

Hypothesis 70: There is no significant correlation
between a student's socioeconomic status and that
student's level of self-esteem.

8. The socioeconomic level of the school does not affect
the student's level of mathematics achievement.

Hypothesis 80: There is no significant correlation
between the socioeconomic level of the school and
students' mathematics achievement.

9. The socioeconomic level of the school does not affect
the student's level of self-esteem.

Hypothesis 90: There is no significant correlation
between the socioeconomic level of the school and
students' level of self-esteem.

10. Mathematics achievement is positively related to
self-esteem.

Hypothesis 100: There is not a significant
positive correlation between mathematics achievement
and self-esteem.

74

11. Schools where students have higher self-esteem scores
have done better on state examinations over the past
four years than have other schools.

Hypothesis 110: There is not a significant
positive correlation between state examination
results over the past four years and school
self-esteem scores.

12. Schools where students have higher mathematics scores
have done better on state examinations over the past
four years than have other schools.

Hypothesis 120: There is not a significant
positive correlation between state examination
results over the past four years and school
mathematics achievement.

13. Males have higher levels of mathematics achievement
than do females.

Hypothesis 130: There is not a significant
correlation between gender and mathematics
achievement.

14. Males have higher levels of self-esteem than do
females.

Hypothesis 140: There is not a significant

correlation between gender and self-esteem.

Data collecpion instruments

Three major data collection instruments were used in order to
gather information about the above hypotheses: l) the adult version
of the Coopersmith Self-Esteem Inventories (a 25-item questionnaire),
2) an eighth-grade mathematics test derived from the Second Study of
Mathematics of the International Association for the Evaluation of
Educational Achievement (IEA), Population A, and 3) a twelfth-grade
mathematics test derived from the Second Study of Mathematics of the
International Association for the Evaluation of Educational

Achievement, Population B. (For copies of all instruments used, see

75
Appendix A.) A researcher-designed questionnaire was administered
orally to the principals to provide direction to the interviews which
were held. Demographic data were collected from students, teachers,
and principals. Teachers also provided information about which of
the math test items their students had studied that year or in
previous years (referred to in the IEA studies as an
Opportunity-to—Learn questionnaire). In addition, a considerable
amount of information was written down in diary form during the
on-site visits to the various schools. This record contains
information about building quality, observed level of discipline and
various SES indicators, along with a record of the data collection

procedure.

Pilot testing of the instruments was conducted at one school site
within the Butembo district, lasting for three days. This was a site
where the principal was completing his first year and therefore was
outside of the established sampling frame. The site was remote
enough from other sites to prevent accidental contamination of
results. Participating in this pilot testing were 41 eighth-grade
students and 35 twelfth-grade students. The principal and six
teachers also participated. The purpose of the pilot testing was 1)
to test how long it would take students to complete the math and
self-esteem questionnaires; 2) to ensure that all math questions
were appropriate to the curriculum; 3) to ensure that all self-esteem
questions were culturally appropriate; 4) to ensure that the French
used in the math and self-esteem questionnaires was both locally

correct and easily understandable; 5) to test the need for a Swahili

76
version of the self-esteem questionnaire for use with the eighth
grade students; 6) to test the appropriateness of the questions
designed for interviews with the principals; and 7) to ensure that
teachers understood the instrument which they were asked to
complete. Pilot testing resulted in several significant changes

which are described below.

Instruments for collepting data from atudenta

Each student was given a questionnaire consisting of 6
demographic questions, a 25-item self-esteem inventory, and either a
26- or a 30-item math test. (Eighth-grade students completed a
26-item math test while twelfth-grade students completed a 30-item
math test.) These questionnaires underwent several changes as a
result of pilot testing. In addition to minor corrections of French
terminology, pilot testing resulted in a significant reduction in
length. The eighth-grade questionnaire was reduced from 40 to 26
questions during this process. Six questions (41, 49, 51, 58, 67,
and 69) were removed because they dealt with topics not introduced in
the Zairian program until at least the ninth grade; eight questions
(61, 62, 63, 64, 65, 66, 68, and 70) were removed to enable the
students to complete the questionnaire within two hours. The
twelfth-grade questionnaire was reduced from 34 to 30 questions
because two questions (33 and 57) presented material not covered in
the Zairian curriculum, while two others (37 and 46) presented

material included only in the sciences curriculum.

77

The format of these tests also bears consideration. Students in
Zaire are generally not accustomed to a multiple choice format, or to
working with commercially prepared testing forms. This lack of
familiarity may have caused inaccuracies of response, although this
is not believed to have been a serious problem. While five or ten
students in each classroom had difficulty completing the first two or
three items, help was always provided by the researcher and none of
the students indicated any further difficulty. Approximately twenty
questions out of a total of over sixty thousand (for the entire
sample) were filled out in an incorrect manner--two answers given for
a single question or a number selected which was not a possible

answer .

The French version of the self-esteem questionnaire did not
require changes. However, during pilot-testing, many of the
eighth-grade students were found to have difficulty understanding
French and, because of this, a Swahili version of the questionnaire
was developed. This was done in collaboration with the students and
several of the teachers. The Swahili version was tested with both
the eighth-grade and the twelfth-grade students and the decision was
made to read each item of this questionnaire to all students, first
in French and then in Swahili. In addition, several of the questions
were given in Nande. The researcher speaks all three of these
languages fluently and can vouch for the accuracy of the

translations.

78
A short questionnaire asking for demographic data preceded the
mathematics and self-esteem questionnaires. Students were asked to
indicate age, gender, parents' education, the number of years they
had attended their current school, and whether or not they were in

possession of the mathematics textbook being used.

THE COOPERSMITH SELF ESTEEM INVENTORIES

The 25-item self-esteem test translated for use in this study is
known as the adult form, or the short form. In a study of 103
college students (Bedeian et al. (1977), Kuder-Richardson reliability
estimates of .74 for males and .71 for females were reported.

Test-retest reliability estimates of .80 were found for males and .82

for females. Other studies (Kokenes, 1978; Johnson et a1., 1983;
Marsh et a1., 1983) have confirmed the construct validity of the
instrument.

Cautions have been raised about the use of self-concept measures
across ethnic groups. Zirkel & Cable (1977), for instance, find a
lack of reliability and validity data on self-concept measures,
particularly when used with students who are from different ethnic
groups. Rio (1981) found that a majority of studies dealing with
self-concept in ethnic groups other than Anglo found significant
differences in favor of white students. Bachman and O'Malley (1984)
suggest that black-white differences in self-esteem scores could be
affected by response style since blacks have tended to use the
extreme categories on Likert-type scales more than whites. The use

of a two-category scale such as that used in the present study was

79
found to eliminate this problem. Diaz (1984) investigated the
reliability of the Coopersmith Self Esteem Inventories translated
into Puerto Rican Spanish for a sample of 296 Puerto Rican high
schools (96 attending a school in Connecticut and the rest attending
school in Puerto Rico). He found a high degree of reliability (alpha

- .858).

In this study, a Cronbach's alpha reliability coefficient of .49
was found for Grade 12 students and .41 for Grade 8 students. Since
the data are in dichotomous form, Cronbach's alpha is equivalent to

the Kuder-Richardson-ZO reliability coefficient.

THE IEA SECOND INTERNATIONAL STUDY OF MATHEMATICS QUESTIONNAIRES

The 26-item mathematics achievement test used with the Grade 8
students was a translation of the 40-item Population A core test
(with items eliminated according to the procedure described above).
The 30-item mathematics achievement test used with the Grade 12
students was a translation of items selected from the l36-item
Population B test according to their correspondence to the Zaire
mathematics curriculum. In the IEA studies, Grade 8 students

answered 75 questions while Grade 12 students answered 34 questions.

A Cronbach's alpha reliability coefficient of .80 was found for
the sample of Grade 8 students in this study and a Cronbach's alpha

of .85 was found for the sample of Grade 12 students.

80

uments o o c n data rom r nci als

Principals were interviewed for an average of one to two hours.
A 15-item questionnaire was administered orally at the beginning of
this interview. (See Appendix A). The questionnaire was designed to
obtain specific information regarding the emphasis which the
principals placed on instructional leadership. The key items
included for this purpose dealt with the number and nature of
classroom visits, pedagogical committees (faculty meetings during
which methodological issues are discussed and, at times, model
lessons given), written instructional objectives, interactions with
teachers, academic training, purchasing priorities, task priorities,
and perceived role. Where assistant principals were interviewed,
they were asked the same questions. In addition, principals were
asked to provide data from school archives indicating attendance,
retention, pass/fail rates, and state exam results for the previous
three years. The document containing this information about students
is known as the Palmarea and is required of all principals. In most
of the schools, it had been meticulously prepared. The only
exceptions to this were three of the four government schools in the
sample. In one of these, the Balmapes was only available for one of
the past three years--the most recent one had not been prepared and a
second one was lost. In two other government schools, the Paimares
for one of the three years had been lost. In all cases, the
researcher was obliged to hand copy the information from the
£alma;aa--there were no extra copies and no facilities at any of the

schools for making copies. Information regarding state examination

81
results for the 1985-86 school year was obtained by mail subsequent

to its publishing in national newspapers.

The fifteen items listed in the questionnaire for the principals
were intended only to frame the broad areas of interest for the
interview and to bring a certain degree of structure. Within the
general categories defined by these items, many further questions
were asked. These included questions about teacher salaries, working
conditions, building projects and corruption. Several principals
also provided the researcher with copies of school rules (for
teachers as well as students) and an assortment of other documents
ranging from a statement of current needs to forms for classroom

visits.

Based on the interviews with the principals (with the teacher
interviews and, to a lesser degree, personal observations providing
verification for several points) a scale was devised to rate the
emphasis which principals place on instructional leadership. This
scale included the following factors: 1) the number of annual
classroom visits made by the principal (or by an individual delegated
by the principal) for the purpose of evaluation or supervision of
instruction; 2) the number of teachers' meetings held each year for
the purpose of instructional improvement; 3) the degree to which
school goals include explicit instructional objectives arrived at by
consensus with teachers; 4) the principal's familiarity with and
knowledge of the curriculum; 5) the degree to which the principal

places a priority on purchases which will support instructional

82
objectives, as determined in collaboration with teachers; 6) the
degree to which the principal perceives his role as one of
instructional leadership; 7) the existence in the school of an
orderly climate in which students are free to pursue instructional
goals without disruption; and 8) the extent to which the principal
is personally engaged in teaching. (See Appendix D for a more

complete explanation of how this scale was constructed.)

I r me ts for o lectin data om teachers

Formal interviews were conducted only with the math teachers at
the schools visited, although there were many informal discussions
with other teachers as well. The interviews conducted with the math
teachers were short, generally lasting from fifteen to twenty
minutes. Teachers were asked to provide general data on their
educational background, their years of experience, and employment
patterns. When it became apparent part way through the data
collection process that many of these math teachers were giving
courses at more than one school, specific questions regarding this
process were included. In addition to the interview, teachers were
asked to indicate for which of the test items their students had
studied relevant material that year, for which of the items they had
studied relevant material in previous years, and for which of the
items they had never studied relevant material in class. This
questionnaire was based on the similar Opportunity-to-Learn
questionnaires used in the IEA studies. The sole purpose of this
questionnaire was to provide a means to check the validity of the

math questionnaires.

83

papa cpileption procedures
ACCESS TO THE SITE

The first step in the data collection process was to obtain
permission from regional political and education authorities to
conduct the study. This was accomplished in January, 1986, through
correspondence with the responsible authorities. A letter of
permission was prepared for signature and mailed along with an
accompanying letter of explanation to contacts in the area who
visited the authorities indicated and obtained the necessary

signatures (see Appendix B for copies of this correspondence).

Procedures for the data collection process were developed in
collaboration with the Michigan State University Committee for
Research Involving Human Subjects (see Appendix C). They were
designed to protect the anonymity of students, teachers, and
principals participating in this research. In keeping with these
guidelines, an answer sheet was used which made no provision for the
names of respondents; students were specifically requested not to
write their names on either the questionnaires or the answer sheets.
Students were further instructed to place the questionnaires and
answer sheets face down on the teacher's desk upon completion. These
were not to be picked up by the researcher until all tests were
returned. This procedure could not always be implemented, however,
because in some classrooms there was no teacher's desk, while in
others, because of cramped seating arrangements, it was not feasible
for students to leave their position until all others in their row

had also finished. In these cases, the researcher collected the

84
questionnaires and either held them face down or placed them face

down on whatever convenient surface was available.

All students, teachers, and principals were told that they had
the right not to participate in the study although there were none
who availed themselves of this opportunity. Teachers and principals
were guaranteed anonymity. School names were not used on any of the
questionnaires or on any of the large manila envelopes in which
questionnaires and interview notes were kept. Each school was
assigned a number known only to the researcher and all materials for
that school were kept in a separate envelope on which only that
number appeared. These numbers have been changed for use in this

dissertation.

Trapspoptation

Transportation to the field was not easy; from the Zaire border
town of Goma to the town 180 miles away which served as research
headquarters, transportation was secured by standing at the side of
the road and flagging down trucks as they passed. Various mechanical
failures to one of these trucks caused further delays, as did a
number of police checkpoints. From beginning to end, the initial
trip to the research area required twenty-three hours of non-stop
travel, including a good many hours atop a pile of vegetables on a
small pickup, with a steady rain falling. This trip also featured a
breath-taking face-to-face encounter with a pair of lions while going
through a section of the Virunga National Park. (The male was

stretched across the road; as the pickup with nervous researcher

85
clutching desperately to several sacks of potatoes pulled slowly into
the ditch to go around, the lion stood up and looked him straight in
the eye from five feet away. This, thankfully, proved to be the only

encounter with a recalcitrant authority on the entire trip.)

Transportation between research sites in the field, if far from
simple, was nevertheless much less complicated than the process of
getting to the area in the first place. This was due to the use of a
motorcycle loaned by a friend for the duration of the study. A
typical day meant getting on the road about 6:30 in order to reach
the scheduled school for the 8:00 beginning of the school day. Where
possible, arrangements were made in advance for the classroom visits
and the interviews. At several of the more remote locations, this
was not possible, however, and arrangements were made immediately
upon arrival. In no instance was there anything other than the
fullest cooperation offered. Principals, teachers and students alike
took a genuine interest in the study and went out of their way to
ensure its success. At several schools, entire classes remained one
and sometimes two hours beyond the scheduled school closing time when
the questionnaires could not be administered earlier due to
previously scheduled classroom activities. At one school,
twelfth-grade students volunteered to come to school on a Sunday when
no other time was available because of exam scheduling. Every

student in the class was present at this session.

86

PRODECURES FOR COLLECTING DATA FROM STUDENTS

Anticipating average class sizes from forty to sixty in the
eighth grade and from thirty-five to fifty in the twelfth grade, the
researcher arrived in the field with sixty—five eighth-grade
questionnaires and fifty-five twelfth-grade questionnaires. At each
school, principals and teachers requested copies of the math
questionnaire for use in working with students. At the first school,
when such a request was made, it was granted since the researcher did
not expect any more such requests. When it quickly became apparent,
however, that similar requests could be expected at each school,
principals were promised a copy of the questionnaire upon completion
of the study. This was duly accomplished, with the remaining copies
divided up and mailed to all of the schools in the sample on the
researcher's last day in the field. By that time, the researcher was
left with forty-nine eighth-grade questionnaires and forty-one
twelfth-grade questionnaires. This was because the questionnaires
were highly valued by students as well as well as teachers and
principals. Since at the first three or four schools, no careful
safeguards were initiated, several of the students turned in their

answer sheet but kept the questionnaire.

At most schools, the questionnaires, answer sheets and scoring
pencils were distributed one to a student. At four of the schools,
due to the large size of the class, some of the eighth-grade students
were constrained to share mathematics questionnaires. This also
occurred at one school for some of the twelfth-grade students. It is

difficult to say whether this resulted in an increased sharing of

87
answers but this did not appear to be the case. Students were asked
to shield their answers from other students and, in general, were

very careful to do so.

Even in the majority of schools where there were sufficient
questionnaires for all students, conditions in most classrooms were
such that students were commonly sitting four or more to a shared
bench under extremely crowded conditions. Students were asked not to
look at one another's answers and appeared accustomed to shielding
their work. On two occasions, however, virtually identical math
questionnaires were found to have been completed by students sitting
adjacent to one another. This was never found to have occurred where
students were constrained to work from the same questionnaire. In
the last two schools where sharing was necessary, the researcher
removed the staples from the questionnaires and had some of the
students begin with the last questions and others with the first.
This involved only twenty-five or thirty students in the total
sample. There was never any shortage of the self-esteem

questionnaires.

In each classroom, the instructions at the beginning of the
questionnaire were read first and then approximately five minutes
were taken explaining the use of the answer sheets and providing
examples on the blackboard (where blackboards existed, that is). In
almost all cases,' students were using computer-coded answer sheets
for the first time. There was typically very little confusion,

however, and from the 1093 students participating in the pilot

88
testing and the actual study, 1093 answer sheets were returned, all
of which had been completed in a manner consistent with the

instructions given.

A total of 1200 answer sheets and 1200 sharpened scoring pencils
were taken to the field. Pupils were allowed to keep their pencils
upon completion of the questionnaire. The announcement that this
would be done was invariably met with applause and shouts of
approval. This forty-dollar investment proved to be the wisest of
the entire study, securing enthusiastic cooperation and undoubtedly
improving considerably both the rate and the quality of cooperation

with the objectives of the study.

In all but four of the schools, the researcher remained in the
classroom while the questionnaires were completed, spending much of
this time recording observations regarding demographic and school
quality variables. In several schools, scheduling requirements meant
that two classes had to complete the questionnaire at the same time.
In these instances, teachers served as monitors while the researcher
circulated from one room to another. In all instances, all verbal

instructions were given only by the researcher.

The administration of the mathematics questionnaires differed
from that used in the original IEA studies (for this reason and
others, no attempt is made to compare results). Based on time
requirements determined during pilot testing, eighth-grade students

were allowed up to two hours to complete the math questionnaire;

89
twelfth-grade students were allowed up to two-and-a-half hours. (The
IEA studies allowed only fifty minutes for similar Grade 8
questionnaires.) Between five and ten students in the entire sample
of schools rushed through the questionnaires in a perfunctory
manner. No attempt was made to prevent this, as students had been
told they were free to participate as they saw fit. One student was
unable to complete his questionnaire due to an attack of malaria. An
average of from one to two students per classroom were absent on the

day of the testing.

The questions dealing with demographic data were completed
first. They were read to the students and explained, both in French
and in Swahili. The self-esteem questionnaire, which followed, was
also read item by item to the students. Each item was read first in
French and then in Swahili, although students had copies of the
French version only. Several questions were also given in Nande, for
ease of comprehension. An attempt was made to keep questions to a
minimum during this period, but in each classroom, some questions
were asked. All items were read twice and all students were given
time to complete them. The self-esteem questionnaire generally took
between twenty and twenty-five minutes to administer. Students
completed the math questionnaire on their own; very few questions

were asked.

Twelfth-grade students in particular showed great interest in the
study and frequently remained after school hours to discuss the

questionnaire and the answers to questions. It was the rule, rather

90
than the exception, that the researcher spend from one to two hours
fielding math questions, as well as questions about U.S. university

life and so on. On one occasion, four hours were spent in this way.

PRODEDURES FOR COLLECTING DATA FROM TEACHERS AND PRINCIPALS

Completion of the student questionnaires generally required the
entire school day (from 8:00 until 1:30) and commonly an hour or so
more. Most of the teachers also answered questions about the content
validity of the instrument (the Opportunity-To-Learn questionnaires)
during this time. Teacher and principal interviews were conducted
for the most part after school hours. In some cases, teacher and
principal interviews were conducted several weeks after the school
visit. As far as possible, these interviews took place either in the
teacher's classroom or in the principal's office. This was not
always possible, however, both because of scheduling difficulties and
absenteeism. The problem was compounded by the fact that not all of
the teachers and principals who were absent at the time of the
initial visit could be located and interviewed before the school year
ended. In a number of instances, this meant that teachers and
principals had to be located and interviewed in their homes or at
some other site. The local children proved to be very helpful in
locating individuals. Three full days were spent in this way and
ultimately all of the sample principals and 28 of 33 teachers were

located and interviewed.

91

PROTECTION OF DATA

The motorcycle proved a convenient and economical means of
transportation, providing access to many areas where the state of the
road would have rendered access by car very difficult. Travel was
complicated considerably at times by rain, however, and
questionnaires and answer sheets were always wrapped in plastic to
keep them dry. On one occasion, the researcher was the recipient of
a gift of a dozen eggs, several of which broke in his backpack during
a rainstorm. In spite of plastic wrapping, the resultant foamy mass
penetrated almost everything. Fortunately, the day's answer sheets
were still decipherable and the answers were hand copied later to
other answer sheets. The questionnaires were largely undamaged. On
several other occasions, sudden downpours resulted in time spent
drying out around smoky fires in distant villages, but no data were

ever lost.

Air travel to and from central Africa required two days' time
each way. Questionnaires and answer sheets (which weighed a total of
about fifty pounds) were always hand carried, while most clothing and
personal effects were checked through. This resulted in some puzzled
comments from airport security officials, but there were never any
complications. Indeed, the entire data collection process was
remarkably free of complications. A total of only 31 days was
actually spent at the research site (from May 19 to June 19). The
first day was spent paying courtesy visits to the various officals
who had granted permission, collecting data about the schools in the

sample from the Butembo district office, and renewing acquaintances

92
with the inspectors. The next three days were spent pilot testing
the instruments and a day was spent motorcycling from place to place
making initial contacts with a number of the principals. Subsequent
contacts with other principals, where it was possible at all to
contact them prior to the actual data collection visit, were made

while passing by on the way to other schools or in the evenings.

Three weeks were spent collecting data from the schools. An
average of one school was visited each day from Monday through Friday
(the researcher is a Seventh-day Adventist and, although schools were
open Saturday mornings, made no visits at that time). Remarkable
cooperation was offered by the principals, teachers and students.
Since it was near the end of the school year (schools closed
throughout the district on Saturday, June 21, even though the school
year did not officially end until two weeks later), the researcher's
visits began to conflict with school examination schedules during the
last full week during which data were collected from students (June 8
through June 13). Students typically completed their school
examinations before noon, however, and arrangements were made for
them to complete the questionnaires after the day's examinations were
over. This may have resulted in somewhat lower results due to
fatigue for the last three schools. This was almost certainly the
case for twelfth-grade students at School 15, where the
questionnaires were completed Friday afternoon, and for students at
School 13, where examinations lasted until 1:15 and the
questionnaires as a consequence were not completed until about 4:30.

These were the only two schools where students complained about being

93
tired, although even here they were very gracious about participating
in the study. A number of students at School 15 did not complete all

of the mathematics questions, however.

On only one day during the entire three-week period was it
impossible to administer questionnaires to students and this was due
largely to the amount of travel time to the research site (over five
hours). In all, some twelve hundred miles were logged by motorcycle
over this time and another four hundred by other means of transport.
Average travel time was two hours per day. The equator was crossed

exactly forty times during the study.

CHAPTER FOUR

PRESENTATION AND ANALYSIS OF THE DATA

Beautiful, beautiful scenery here. Dramatic
patchwork hillsides of wheat, banana groves waving
in wind. Ruwenzori Mountains clearly visible in

distance, clouds lifting from snow-capped peaks.
Sunny day, cumulus clouds drifting past, shadows
racing up and down valleys and across fields.
Only sounds--cry of a hawk somewhere and
scratching of pencils on paper. Look out school
window, life opens up. Look in, life closes in
around you. Dirt floor. Mud walls in bad
condition--top part crumbling. Benches here so
close together that students in middle get out by
sliding under. Grass growing long inside
classroom along three walls. Good sign. Maybe
growth is possible in here after all.
-field notes, May 29

Introduction

This chapter presents summary statistics of the major variables,
followed by the test results of the hypotheses under investigation.
The hypotheses are restated in statistical form and the statistical
procedures followed are described. Throughout this chapter, all
hypotheses are tested at a significance level of .05. In addition to
the statistical data, the final section of this chapter also presents
qualitative data based on interviews, field notes and other

observations.

Studenp outcpmea
The mathematics achievement and self-esteem scores for the

students at the fifteen schools in the sample are given in Appendix

94

Percentage

Percentage

95

 

100

90"

80"

me-

”an

20‘

10"

 

 

 

7 l I 10

a umnauwu-mnc + Sdhiififl'

Figure 4-1
Mathematics and Self-esteem, Grade 8

0 I I I I— I I I I I I I I
5

He "5 I2

I

13

I
I4

I!

 

00"

 

 

 

 

”-
IO-I
a I I I I I r I I I I I I I I T I
IZJer‘SO'IIIIOHIZeIZOIJMeMe
Seheele
O mum: + Sent-tenet“
Figure 4-2

Mathematics and Self-esteem, Grade 12

I!

96
E, along with standard deviations and standard error. Figure 4-1
presents mathematics and self-esteem outcomes for the eighth—grade
students. Figure 4-2 presents mathematics achievement and
self-esteem scores for the twelfth-grade students. These are
expressed as percentages of the total possible. Note that at several

of the schools, more than one classroom was visited.

The mean score on the Grade 8 mathematics achievement instrument
was 44.36 percent. The mean score on the self-esteem instrument was
56.88 percent. This compares with a mean score on the Grade 12
mathematics achievement instrument of 40.95 percent for the sciences
option and 32.61 percent for the humanities option. Grade 12

self-esteem scores were 61.19 percent and 61.96 percent.

Principals' emphasis on instructional leadership

The scale used to rate principals on the basis of emphasis on
instructional leadership is presented in Chapter Three. The figures
used to derive the ratings are explained in much greater detail in
Appendix D. Table 4-1 presents the results of this rating
procedure. No ratings are provided at School 4, since the principal
had not been at this school for a long enough period of time for this

to have been meaningful.

97

 

 

 

 

 

 

 

 

 

Table 4-1
Ratings of Principals
Principaln,l 2 3 4 5 6 7 8 9 10 ll 12 13 14 15
Visips 3 3 2 - 2 3 2 l l 3 3 3 l 3 2
Objectives 2 2 2 - 1 2 1 1 1 l l 1 l 2 1
Meetings 2 3 2 - l 2 l l 1 2 2 2 1 3 1
Knowledge 1 l 2 - 2 l 1 2 2 2 3 3 3 3 2
Purchasing 1 2 2 - 3 3 3 3 2 3 3 3 3 3 3
Rpia 2 3 3 - 2 2 2 2 2 2 2 2 2 3 2
Discipline 3 3 3 - 3 3 2 2 3 2 2 3 2 2 2
Iaaching: l 2 l - 2 3 2 3 l 3 3 3 l 2 1
Totals 15 l9 l7 - 17 20 14 15 13 18 19 20 14 21 14

Levaia of parental education

In this study, the level of the mother's education is used as a
proxy for socioeconomic status. Data about both the mother's and the
father's educational level were obtained. A number of references
throughout the literature, however, indicate that the level of the
mother's education, which is usually considerably lower than the
level of the father's education, provides a much more discriminating
measure of socioeconomic status. Appendix F gives both the fathers'
and the mothers' levels of education by school and by grade level.
Figure 4—3 shows the level of education of the mothers of Grade 12

students.

The figures given below actually represent a slightly lower
degree of precision than would appear at first glance. More
precisely, they should be interpreted as follows: a score of 0 means
that the average father or mother has had less than one year of
schooling; a score of 3 indicates an average of two to four years of

schooling; a score of 6, completion of primary school; a score of

98
9, completion of one to three years of secondary school; and a score
of 12 indicates completion of secondary schooling. The figures show
that at some schools, mothers average less than one year of schooling
while at others the typical mother has completed primary school.
Fathers average a considerably higher degree of education than
. mothers. Similarly, the parents of Grade 8 students average a higher
degree of education than do the parents of Grade 12 students. This
probably reflects the rapid expansion of primary education more than
anything else, although with the great number of students who do not
complete secondary schooling, one might have expected that those who
do finish would tend to have higher socioeconomic status than those

who do not.

 

  

MOTHERS' YEARS OF EDUCATION

   
  

 
    

 

 

 

 

a /
:1 g "
W 4
I ’ /
W a R
2"? fig 1 / //H/a //
”C; ”[5 V V /
~flfia a a a 4??
a 6%! E47“: l/I/4I Ig‘jE/‘a pawn I '
Figure 4-3

Level of mothers' education

 

99

Correlates of mathematics achievementgand:§e1f-esteem

The following page presents a list of variables which are
correlated with mathematics achievement and levels of student
self-esteem. The Grade 8 list gives Pearson correlation coefficients
while the Grade 12 list gives partial correlation coefficients which
have been corrected for the differential effect of the option
followed. A much fuller discussion and more precise statistical

treatments are presented throughout the rest of the chapter.

A caveat is in order at this point, however. The list is given
in this fashion for the sake of convenience to the reader but should
not be taken as an endorsement of the "black box" approach (a
discredited static model of school organization whereby discrete
inputs are expected to produce predictable outcomes). The
organizational processes of schooling are far more complex, with the
effect of variables depending frequently on the mediating conditions
which exist. Without an overall interpretive framework, the analysis
of discrete variables will be of little value. An important thrust

of this study has been the development of such a framework.

Table 4-2
Variables correlating with mathematics achievement and self-esteem

 

Mathematics Self-esteem
Variable Grade 8 Grade 12 Grade 8 Grade 12
Principals'emphasis on .16*** .20***
instructional leadership
Classroom SES level .14***
Number of watches .15***
Possession of textbook .15**
Principals' tenure .21*** .31***
Teaching principal .l9*** .lh*** .14**
Teachers' experience .24*** .09*
Teachers' qualifications -.O9** .21*** .10**
Supervision of teachers .20*** .20*** -.15*** .10*
Faculty meetings .21*** .12*** -.ll**
Discipline .14**
Instructional objectives .20*** -.08*
*** p<.001
** p<.01
* p<.05
Da a sis

In this section data are presented following the hypothesis to
which they refer. Analysis of the data is performed for each

hypothesis in order.

Hypothesis 10: There is no significant correlation
between principals' instructional leadership rating and student
mathematics achievement.

Analysis of co-variance was performed on the Grade 12 results,
comparing the rating of the principals to mathematics achievement in
order to determine the general effect of the principals' emphasis on
instructional leadership. The mathematics curriculum followed by the

student was used as a co-variate.

101
Analysis of variance was performed on the Grade 8 results, again
comparing the rating of the principals to mathematics achievement in
order to determine the general effect of the principals' emphasis on
instructional leadership. Since the mathematics curriculum followed

at this level is the same for all schools, no co-variate was used.

The results of the ANCOVA procedure for Grade 12 mathematics
achievement are given as Table 4-3. The results of the ANOVA

procedure for Grade 8 mathematics achievement are given as Table 4-4.

 

 

 

 

 

 

Table 4-3
Grade 12 ANCOVA for Mathematics Achievement
Sogrgg gr Variarion Sum of Squares DF Mean Square F Sigr
Option 3158.314 1 3158.314 32.685 .001
Principal 5794.729 8 724.341 7.496 .001
Explained 8953.043 9 994.783 10.295 .001
Residual 22249.32; 28; 96.629
Total 36202.435 291 124.407
Multiple R squared - .247
Multiple R - .497
Table 4-4
Grade 8 ANOVA for Mathematics Achievement
Sourgg of Variation firm of Sguares DF Mean Square F Sig.
Principal 8028.470 8 1003.559 5.301 .001
Residual ll8882,852 628 189.304
Total 126911.326 636 199.546

Multiple R squared - .063
Multiple R - .252

102

These figures merit careful consideration for several reasons.
First, they show that, for both the Grade 8 and the Grade 12 students
in the sample, there is a statistically significant relationship
between the emphasis which principals place on instructional
leadership and student outcomes (mathematics achievement). The level
of significance found is p-.001 for both the Grade 8 and the Grade
12 students. Second, they show a highly significant effect of the
option followed. (Other data from the study indicate that students
in the sciences option average 8.268 percent higher than students in
the humanities option; this is not particularly surprising since
students in the sciences option average five class periods of
mathematics per week compared to two in the humanities option.)
Third, in addition to statistical significance, there is practical
significance as well. The multiple R squared figure for Grade 12
students (.247) indicates that 24.7 percent of the variation in
mathematics achievement is explained in terms of the variables
included. This is not equivalent to saying that 24.7 percent of the
variation is caused by the variables included, however. First, there
may be a significant non-linear component to this relationship.
Further, there may be a number of confounding variables which have
not yet been considered. For instance, the students where the
principals rate highest may have higher socioeconomic status than the
others and may be doing better for this reason. Again, the higher
ranking principals may have recruited students with better academic
ability and it may be that this is what causes the better results.
Much of the rest of this chapter will be devoted to examining this

question. What can be said for certain at this point is that the

103
data do support the alternative hypothesis, i.e. there is a
significant correlation between principals' instructional leadership

ratings and student mathematics achievement.

The finding of a lower level of practical significance for Grade
8 students (6.3 percent of the variation in mathematics achievement
is explained in terms of the principals' emphasis on instructional
leadership) is also meaningful for several reasons. Principals
generally have much more contact with Grade 12 students than they do
with Grade 8 students. Many principals tend to treat the Grade 7 and
Grade 8 classes as more of a selection process than an instructional
process. Only one of the principals who teach, for instance, teaches
any courses at either the Grade 7 or the Grade 8 level. Grade 8
students have also had less time to be affected by the principals
than have Grade 12 students. The finding of lower significance
levels also tends to argue against the probability that students
recruited at the schools where principals rated higher were
significantly more capable academically than students at other
schools. More will be said about this later. At any rate,

Hypothesis 10 may be rejected at a significance level of .001.

Hypothesis 20: There is no significant correlation between
principals' instructional leadership ratings and student
self-esteem scores.

Since the option followed by the student proved not to make any

significant contribution to the Grade 12 analysis of variance for

self-esteem, ANOVA procedures were conducted for Grade 8 and Grade 12

104
both, showing the percentage of variation in self-esteem explained by

the principals' emphasis on instructional leadership.

It may not, at first glance, seem reasonable to try to determine
the correlation between principals' emphasis on instructional
leadership and the students' self-esteem scores. However, creating a
climate in which students can concentrate on learning is seen as an
important function of principals who emphasize instructional
leadership. A harsh, stifling climate could be expected to result in
lowered self-esteem while a more supportive climate could be expected
to result in a raising of self-esteem scores. Climate is one of the
factors included in the instructional leadership rating. The results
of the Grade 12 ANOVA procedure for self-esteem are given as Table

4-5. The results for Grade 8 are given as Table 4-6.

Table 4-5
Grade 12 ANOVA for Self-esteem

Source of Variation Sum of Squares DF Mean Square F Sig,

 

 

Principals 196.627 8 24.578 3.285 .001
Residual 2162,512 282, 7.483
Total 2359.139 297 7.943

Multiple R squared - .083
Multiple R - .289

105

 

 

Table 4-6
Grade 8 ANOVA for self-esteem
So 0 Variat o um o S uares DF Mean Square F Sig.
Principals 304.028 38.004 5.058 .001
Residual 4270,909 _35 7.513
Total 5074.937 643 7.893

Multiple R squared - .060
Multiple R - .245

These results indicate that, while the emphasis which principals
place on instructional leadership does not have a particularly strong
effect on the development of self-esteem, statistical significance is
obtained. Of more interest than this, however, is the direction of
the results. While the correlation for Grade 12 students is a
positive one, the correlation for Grade 8 students is a negative one.
(When summary statistics for self-esteem are examined, they show that
in schools where principals rate in the top half for instructional
leadership emphasis, Grade 8 self-esteem scores average 55.5
percent. For schools where principals rate in the bottom half for
instructional leadership emphasis, Grade 8 self-esteem scores average
58.0 percent. This trend is reversed for Grade 12.) Since the
correlations found are small and since there may well be other
intervening variables, it would be premature to read a great deal
into this finding but it does indicate that an emphasis on
instruction may have a negative impact on other valued student
outcomes, particularly for those students with limited influence in

the system, such as the Grade 8 students in this study..

Hypothesis 30: There is not a significant positive
correlation between academic intake and principals' instructional
leadership rating.

106

The challenge here is to show that differences in outcomes are
related to differences among the principals. This requires showing
that the academic ability of students enrolling at the schools with
the higher rating principals is not superior to that of students
enrolling at the other schools. The most convincing way to show this
would be with a pretest-posttest design. Students would be measured
at the beginning of some period and then again at the end in order to
determine 1) where the best students are enrolling and 2) the actual
levels of improvement in mathematics achievement for various student

subgroups.

Given the constraints of time and money, this design was not
possible for the present study. Instead, three measures of the
academic intake at each school have been used. First, an expert
panel consisting of three primary school inspectors and one secondary
school inspector was asked to rank all of the secondary schools in
the sample as to academic intake. They stated that there were no
obvious patterns of one school in a particular area attracting all of
the best students, but agreed to provide a tentative ranking. They

stressed that it was tentative. Results of this ranking are given in

Table 4-7.
Table 4-7
Academic Intake of Schools Ranked by Inspectors

Academic Intake Schools

High 1,3,4,8

Medium High 10

Medium 14,15

Medium Low 2,7

Low 5,6,9,11,12,l3

107

When a Pearson correlation coefficient is obtained for academic
intake, it shows a slight negative correlation with Grade 8
mathematics achievement (r--.0912 with p - .011) and no statistical
significance for Grade 12 mathematics achievement. This bolsters the
argument that results are not due to academic intake. However, as
noted, there may be some question as to the validity of this
ranking. It is also found to correlate quite highly with the measure
of socioeconomic status (.2514, p - .001) and may result from a

confounding on the part of the inspectors of these two concepts.

A second measure of academic intake is based on admissions
criteria. A ranking of schools according to the admissions criteria
which they apply was attempted. Again, it is not a simple matter to
differentiate among the schools on this basis because most of the
principals state that they apply similar criteria, i.e., admissions
are usually restricted to Grade 7 and Grade 9 only (referred to as
the recruiting classes) and are based on successful completion of an
entrance examination and presentation of a bulletin from the previous
year with at least a 50 percent average. Two principals indicated a
higher criterion, i.e., presentation of a bulletin from the previous
year with at least a 55 percent average. The following factors were
considered in preparing this second measure of intake: 1) the
percentage required for permission to write the entrance exam; 2)
the standard applied on the exam (percentage required for passing);
3) the number of students applying for a position; 4) whether or not

students are admitted in other than the two recruiting years; and 5)

108
whether or not the decision is made by the principal alone (which

could lead to abuses) or by a committee of teachers.

The rankings obtained in this way are explained in more detail in
Appendix G. They turn out to be surprisingly similar to those
indicated by the inspectors. There is only one school which clearly
has a level of academic intake which surpasses the others. This is
School 3, one of the two schools which require 55 percent of any
student who wishes to write the entrance examination. It is a school
with an excellent reputation and high numbers of candidates each
year. Two principals indicated that this is the school of first
choice for many students. This school has also just begun its own

private primary school following a Belgian curriculum and taught

entirely in French. This project has received government approval
but is supported locally. It is run by the secondary school
principal, as is a combination kindergarten/day care program.

Interestingly, even here, the principal complained that entrance exam
scores frequently turned out to be very poor predictors of subsequent

academic success.

Only one other school was indicated by a number of individuals
(including the .inspectors, another principal and several parents) as
the school of preference in its area. This was School 14. However,
it did not seem nearly as stringent in its admissions criteria and
did not appear to have the same high quality pool of candidates among
which to choose. Many of the other schools were not in competition

for candidates since they were in relatively isolated areas. The

109
school whose students performed the highest in terms of mathematics
achievement at the Grade 8 level appeared to have among the lowest
levels of academic intake in the sample, with the principal
indicating that he was forced to accept students scoring as low as 25
percent on the entrance examinations because of the general low level

of the elementary schools in the area.

When Pearson correlations between mathematics achievement and
academic intake are obtained based on this scale, no statistical
significance is found. There is a high correlation between the two

scales (.6834, p<.001). This scale is shown as Table 4-8.

Table 4-8
Academic Intake Ranked According to Admissions Criteria
Academic intake Schools
High 3
Medium high 1
Medium 2,7,10,14
Medium low 6,8,11
Low 4,5,9,12,l3,15

Another factor influencing admissions decisions is the practice
of matabishi, which is the paying of bribes to principals and
occasionally teachers in order to gain admission or other
advantages. This is a subject about which it was generally difficult
to ask questions but some teachers and principals were surprisingly
open about the practice. One teacher provided a ranking of the
schools at which it cost the most to gain admission. Another teacher
suggested that a general rule of thumb for principals who practise
matabishi is that it will cost a student at least one hundred zaires

per year of secondary school for admission. That is, first year

110
(Grade 7) will require a bribe of one hundred zaires; second year
(Grade 8) will require a bribe of two hundred zaires, and so on.
Another informant indicated that this figure was low and that more
than one thousand zaires could easily be paid. The principal at one
of the schools visited told indignantly of one of their students who
had failed and who had immediately been admitted to a neighboring
school after paying a bribe. Students at this school and at several
others openly said that what they appreciated the most about their
school was that there was no matabishi. One teacher also made the
comment that what he liked most about a certain principal was that he
was honest. On the other hand, a teacher at another school

commented:

The principal here makes his own personal
arrangements when it comes to admissions. There
is no committee. This year, most of the
sixth-year students are new. Some years, there
are only five or ten maximum who pass into sixth
year but when you come back after summer vacation,
you find twenty or thirty of them. You don't
bother to ask where they came from. You just
teach.

In general, it appears clear from the interviews that there are
differences in student intake. The practice of matabishi appears
more prevalent at the schools where principals ranked lower, which
means that.those schools are almost certainly receiving at least some
students with lower academic ability on this basis alone. The
schools in the city have a much larger pool of candidates from which
to choose than do those in the more isolated areas, which again means

that they can be more selective.

lll

Logically, one would expect to find some correlation between
levels of academic intake and the outcome scores for mathematics
achievement and, while this is nominally the case, statistical and
practical significance levels are not obtained. There are a number
of reasons why this may be so. First, it may just be that the
differences in academic intake levels between the different schools
are just not that great. This is the belief of the inspectors who
agreed to attempt a ranking with great reluctance. Other
explanations are possible. Many schools which are not the first
school of choice for the most able students appear to compensate by
admitting great numbers of students in the first year and then
applying more frigid standards for passing. One of the schools began
the school year with 118 students in a single Grade 7 classroom. By
June, there were only 102 left in Grade 7; there were 66 in Grade
8. Almost without exception, the schools with higher ranking
principals had larger upper grade classes than did the others. This
indicates at least that they are not maintaining their high levels of
achievement by high degrees of selectivity. (It also demonstrates
the potential hazards of trying to read too much into between-school
variation without controlling for all of the variables. In this
study, lower correlations between academic intake and student
outcomes may well be the result of model underspecification; lower
correlations between principal ratings and student outcomes may also

occur because of this.)

Hypothesis 40: There is no significant interaction effect

between principals' instructional leadership rating and student
tenure.

112

It was expected that the effect of the principals' leadership
rating would increase the longer the student had been at the school.
However, no significant interaction effect between principals'
ratings and student tenure was found. This appears to result from
the fact that in both Grade 8 and Grade 12, it was the students who
had failed the greatest number of times who had been at the school
the longest. Since student tenure appears to be confounded with the
number of failures, no firm conclusions can be reached and the null

hypothesis is retained.

Hypothesis 50: There is no significant correlation between
possession of a textbook and mathematics achievement.

The total numbers of students in possession of a mathematics
textbook in this study are relatively small. At the Grade 8 level,
41 of 614 (less than 7 percent) students reported being in possession
of a textbook; at the Grade 12 level, this figure rose to 68 of 301
(less than 23 percent). The Pearson product-moment correlation
coefficient (Appendix H) shows that possession of a textbook is not
significantly related to mathematics achievement at the Grade 8
level. This may be because such small numbers of students possess
textbooks that teachers are unable to teach in a way that takes
advantage of the few books that students do possess. One principal
indicated that this is in fact the case--that even those students who
do have textbooks are still expected to copy the same texts, etc.

from the blackboard as the rest of the students.

113

At the Grade 12 level, the Pearson product-moment correlation
coefficient attains significance only for the sciences program. Here
the coefficient is .45 with a significance level of p -.001. For the
humanities program, there is no significance. Again, this may
reflect the teachers' use of these books. The teachers following the
sciences curriculum may make much greater use of the books.
Interestingly, when the possession of a textbook is entered into the
analysis of variance at the Grade 8 level and the analysis of
co-variance at the Grade 12 level, while mere possession of a
textbook remains insignificant, the interaction effect with
principals' rating becomes significant, particularly for the Grade 12
students. This indicates that in schools where principals emphasize
instructional leadership, possession of a textbook is relatively
important. In schools where the principal does not, possession of a
textbook appears to be of little value. This result would appear to
coincide with results reported in a textbook distribution plan in the
Philippines (Heyneman, Jamison & Montenegro,. 1984). The Grade 8

ANOVA is given as Table 4-9; the ANCOVA for Grade 12 as Table 4-10.

 

 

 

 

Table 4-9
ANOVA for Grade 8 (Principals, Textbooks)
Source of Variation Sum of Squares DF Mean Square F Sigr
Principals 6284.791 8 785.599 4.514 .001
Text 62.246 1 62.246 .358 .550
2-way interactions 2404.992 6 400.833 2.303 .033
Explained 8752.034 15 583.469 3.353 .001
Residual 97451,019 569 174.020
Total 106203.053 575 184.701

Multiple R Squared .060
Multiple R .244

114

Table 4-10
ANCOVA for Grade 12 (Principals, Textbooks)

Source of Variation Sum of Sguares DF Mean Square F Sig

Covariates

Option 2522.287 1 2522.287 28.416 .001
Main Effects

Principals 5855.735 8 731.967 8.246 .001

Textbooks 165.979 1 165.979 1.870 .173
2-way interactions 3012,259 8 376.595 4.243 .001
Explained 11556.759 18 642.042 7.233 .001
Residual 228ll,6Z9 .251 88.761
Total 34368.438 275 124.976

Multiple R Squared - .249
Multiple R - .499

Since textbooks of themselves do not account for a significant
degree of variation in mathematics achievement, the null hypothesis
must be retained. However, the data indicate that textbooks have the
potential for a significant impact on mathematics achievement when

principals provide strong instructional leadership.

Hypothesis 60: There is no significant correlation between
a student's socioeconomic status and that student's mathematics
achievement.

Of key importance at this point is to demonstrate that the
relationship found between instructional leadership and student
outcomes is largely independent of the socioeconomic status of the
students. This point will be examined in some detail, for the
argument can and certainly has been advanced that differences in
student outcomes at the better schools result from differences in the
family backgrounds of the students and are not the result of school

inputs and processes. The effective schools movement, of course,

115
received its initial impetus from an attempt to refute the claim made
in the Coleman Report and later in the Jencks report that family
backgrounds accounted for almost all of the differences in
achievement levels. Simmons and Alexander (1978) have suggested that
a similar claim can be made for schooling in the developing
countries, i.e. that differences in achievement are due to family
effects. Heyneman and others have attempted to show, on the other
hand, that schooling in the developing countries has an effect on
learning which is relatively independent of family inputs. The data
examined here support this latter position, although this support is

by no means unequivocal.

The effect of family background, or socioeconomic status can be
measured in a variety of ways. In this study, the level of the
mother‘s education (MOMED) is used as a proxy for SES, when the data
are aggregated at the student level. (The level of the father's
education (DADED) was also recorded but there was less variation than
for MOMED and consequently, less predictive value.) One additional
measure is used when the data are aggregated at the classroom level.
This is the percentage of students in the classroom who wear a
watch. While this measure may be somewhat confounded with
rural/urban distinctions because of the greater availability of
watches in urban areas, it correlates at a high level of statistical
significance with MOMED when this latter variable is aggregated at

the classroom level.

116

The Pearson correlation coefficients from Appendix H indicate a
significant correlation between MOMED and principal ratings (r-.16
with a significance level of .003). If MOMED can be shown to be
highly correlated with math achievement or self-esteem, it could be
reasonably inferred that the correlations between these variables and
principals ratings are due not to the principals after all, but to
the higher socio-economic status of the students at these schools.
However, when the data are aggregated at the individual student
level, there is no significant correlation between mother's education
and mathematics achievement at either the Grade 12 or the Grade 8
level. The Pearson correlation coefficients found are low and do not
attain statistical significance (again, see Appendix H). When MOMED
is added to the ANOVA and the ANCOVA already run, there is no
additional predictive power and, indeed, a considerable loss of
significance. Socioeconomic status consistently fails to show any
correlation with mathematics achievement, when aggregated at the

student level.

Hypothesis 70: There is no significant correlation between

a student's socioeconomic status and that student's level of

self-esteem.

The correlation matrix in Appendix H shows that there is no
correlation between a student's socioeconomic status and level of
self-esteem. For Grade 12 students, r-.0560 with a significance

level of .157. For Grade 8 students, r--.012l with a significance

level of .377. Hypothesis 70 is retained.

117

Hypothesis 8 : There is no significant correlation between
the socioeconomic level of the school and students' mathematics
achievement.

When the data are aggregated at the classroom level, (see
Appendix F and Figure 4-1) it becomes apparent that there are
significant differences in socioeconomic status among the schools.
Of particular interest here is the statistically significant
relationship between MOMED and the ratings of principals at the Grade
12 level (r=.1602 with a significance level of .003). This would
suggest that either the parents with higher socioeconomic status are
able to get their children admitted to the schools where they
perceive the principals to be the most serious about instruction or,
conversely, that principals give greater attention to instruction
when the students have more educated parents. The important
questions here are: 1) does the socioeconomic status of the
classroom have an effect on the students' levels of mathematics
achievement? and 2) if so, does this account for the correlation
between mathematics achievement and principals' ratings? When
analyses of variance and covariance are calculated for Grade 8 and
Grade 12, it becomes clear that there is a statistically significant,
positive correlation between mathematics achievement and the
socioeconomic status of the classroom. These analyses are given as

Table 4-11 for Grade 8 and Table 4-12 for Grade 12.

118

 

 

 

 

 

Table 4-11
Grade 8 ANOVA for Classroom SES
Source of Variation Sum of Squares DF Mean Square F Sig;
Classroom SES 8716.489 2 4358.245 23.346 .001
Residual 125821,938 §__ 186.679
Total 134538.427 676 199.021
Multiple R Squared - .065
Multiple R - .255
Table 4-12
Grade 12 ANCOVA for Classroom SES
Sourae of Variation Sum of Squares D Mean Square F Sigi
Option 3158.314 3158.314 33.434 .001
Classroom SES £242,501 1121.251 10.402 .001
Explained 5464.731 3 1821.577 16.899 .001
Residual 34277,298 318 107.790
Total 39742.029 321 123.807
Multiple R Squared - .138
Multiple R - .371

119

These figures show that students attending school in higher SES
environments do attain higher levels of mathematics achievement than
students attending school in lower SES environments, at least within
the ranges found in this study. The important question now is this:
‘when, classroom level SES is treated as a covariate, are there still
significant principal effects? The answer, as shown by Table 4-13,
is a resounding 'Yes!‘ Table 4-13 takes the variables already shown
to be important, and adds classroom level SES as a covariate. Note
'that the effects of the principal ratings are still significant at

‘the .001 level.

 

 

 

Table 4-13
Grade 12 ANCOVA for Principals, Classroom SES
Sourca of Variation Sum of Squares DF Mean Square F Sigl
Covariates 3818.853 2 1909.427 20.143 .001
Option 3158.314 1 3158.314 33.317 .001
Classroom SES 660.540 1 660.540 6.968 .009
Principals $651,282 1 807.326 8.517 .001
Explained 9470.135 9 1052.237 11.100 .001
Residual 26732,300 _82 94.795
'Total 36202.435 282 124.407
Multiple R Squared .262
Multiple R .511

Based on these findings, the most accurate explanation for what
is occurring in the sample schools with regard to SES is that parents
‘vith higher socioeconomic status are able to use their influence to
(ibtain admission for their children at the schools where they
13erceive the principals to be the best, and that once in the schools
Eitudents will do slightly better in those classrooms characterized by

lligher SES. The rating of the principal in terms of emphasis on

120
instructional management has a strong effect independent of classroom

level SES, however, and is much greater than the SES effect.

Because of the high degree of debate which has surrounded the
question of SES in developing countries, a second gauge of the
socioeconomic status of the schools was used in order to provide
“verification of the findings in this area. For this purpose, the
Imumber of students wearing watches in each classroom was counted.
Iflatches are highly valued in the Butembo area and give a reasonable
indication of family income. Those who can afford them generally
‘mear them. This count provided the ranking shown in Table 4-14. It
correlates highly with classroom SES figures from Appendix F for
Grade 12 students (r-.4530, p<.001). It does not correlate highly
‘with classroom SES figures from Appendix F for Grade 8 students
(r--.0623, p-.062) The percentage of watches was typically so low
for Grade 8 students that it evidently did not have a great

discriminatory value.

Table 4-14
Percentage of Students Wearing Watches
As an Indicator of SES, by Class and School

Qrade 8 18 02 07 23 12 -- 22 07 39 14 15 10 15 21 45
gagggalg_ 83 50 90 48 67 -- 18 21 67 60 50 41 49 59 84
Sahool 1 2 3 4 5 6 7 8 9 10 ll 12 l3 14 15

The correlation coefficients for these rankings with mathematics
techievement failed to show either statistical or practical

SBignificance both at the Grade 8 and Grade 12 levels, thus

121
strengthening the argument that classroom level SES does not account
for the correlation between principal ratings and student
achievement. Based on the first of these measures, however,
classroom SES does appear to have an effect on mathematics

achievement. Hypothesis 80 is retained.

Hypothesis 90: There is no significant correlation between
the socioeconomic level of the school and students' level of
self-esteem.

Here one may use the two proxies for socioeconomic level of the
school: average level of mothers' education and percentage of
students wearing watches. Both of these proxies show no significant
correlation with students' level of self-esteem for the Grade 12

students. For the Grade 8 students, classes with a higher percentage

of students wearing watches showed a significantly higher level of

self-esteem (r-.1561, p-.001). However, the null hypothesis is
retained.
Hypothesis 100: There is not a significant positive

correlation between mathematics achievement and self-esteem.

For both the Grade 12 and the Grade 8 students, the Pearson
‘product-moment correlation between mathematics achievement and
self-esteem is small and not statistically significant (see Appendix
Ii). The ANCOVAs for both Grade 12 and Grade 8 also fail to show

statistical significance. The null hypothesis is retained.

Hypothesis 110: There is not a significant positive
correlation between state examination results and school
self-esteem scores.

122
The correlation between self-esteem scores and state examination
results for 1986 is -.0386 with a significance level of .259. The
correlation between self-esteem scores and state examination trends
for the past four years is -.0696 with a significance level of .104.

The null hypothesis is retained.

Hypothesis 120: There is not a significant positive
correlation between state examination results and school
mathematics achievement.

Maloba's study of examination results in Zaire (1977) found that
:suecess in individual subjects as measured by the schools was of
'little value for predicting subsequent examination success. In this
study, however, there appears to be a strong correlation between
school mathematics achievement and state examination results, at
least for the current year. The state examination results over the
'past four years are given as Appendix I. Figure 4-4 provides a

summary of these results expressed as the percentage of students

passing.

The correlation between mathematics achievement scores and state
examination results for 1986 is .1809 with a significance level of
.001. However, when trends in examination results for the past four
jyears are examined, they fail to show significance. While the null
lrypothesis is not retained, it should be noted that current
Inathematics achievement scores are not predicted by the trend in
examination results over the past four years. It should further be
Iloted that examination results are aggregated at the school level and

Ilot at the individual level.

123

 

 

 

-~
affflir/ffff/Arf/r

 

‘UN#NNN~N-N‘NNN~N~N

 

Errlrt tr:

 

“““““~““\“‘b
"l”"""’"l"""""4
-.|

 

 

 

““‘\“““~‘

 

 

“‘\‘~‘\‘\““‘\‘~‘~ “““““ “\“.“s

.‘I”””””””"”’ 'I”"" ’l”"""””"""0

   

".--'|-"“.‘.
”“'I”|‘

s“\\““\\“““““““‘\~““““““t

‘."“.‘|‘

             

""”"””l"l”"""””"l
""-'..'|'l"""|
""“I““

A“\“‘~\““~““‘\~\“““\“‘\~
"I””"’ ’l"”"””"' "l' ” ”""""’o
-"'-- ‘-|“

‘\-‘\~\\‘\“-
'llilflilllla
'.|

  

                           

         

  

‘NNNNNNN
lAflA/fff’fffW’V/AVVWV—

.4:

 

 

100

90

.__._
wwwmw

80-
7°—
60-

.Esm .2332 05.8.. 3352...

_
0

14d 14b 15

12¢ 12!: 13

91011

23¢3b4

22 was

was

Schools
B mm was

Figure 4-4

State Examination Results 1983-1986

124

Hypothesis 130: There is not a significant correlation
between gender and mathematics achievement.

The correlation between gender and mathematics achievement is
small and statistically insignificant (see Appendix H). It
approaches significance only for the Grade 12 sciences program, where
females score somewhat better than males. This may be more an
artifact of the sampling frame, however, since the total sample of
those in the sciences program was small, with a disproportionate

‘number at one of the top Catholic schools, where over 77 percent of

the students in the class were female. The null hypothesis is
retained.
Hypothesis l4 : There is not a significant correlation

between gender and self-esteem.

The Kenyan study by Mwaniki (1973) reported no significant gender
differences in self-concept. Similarly, in the overall sample for
this study, the correlation between gender and self-esteem is small
and statistically insignificant (see Appendix H). However, for the
Grade 12 students in the sciences program, there is a correlation
between self-esteem and gender (r-.2276) which is significant at the
.05 level, with girls rating higher than boys. The total numbers
involved are small, however, and it would seem premature to read much
into this finding. Based on the data collected, the null hypothesis

is retained.

§araadipity
It has already been noted that this study was developmental in

tiature. Although many of the findings were anticipated in the

125
research hypotheses, some were not. In particular, the unanticipated
findings relate to trends in the examination results and to teacher
characteristics. Trends in the examination results are explained

first.

It is difficult to handle the examination results statistically
‘because it is virtually impossible to determine with what they should
'be compared. Different subjects have different pass rates. In
(iifferent years, the pass rates appear to vary considerably. From
'region to region. within the country, pass rates vary. Some schools
deliberately keep the size of the graduating class small in order to
improve their passing percentages. There is evidence that this may
be occurring in the Butembo district. The number of candidates
presented at the sample schools declined in each of the past four
years, from 542 to 503 to 475 to 465. This occurred while the number
of students in Grade 7 was going from 1461 to 1571 to 1558 in the
twelve schools for which figures were available. (No Grade 7 figures

were obtained for the current year.)

The last year for which national and provincial figures were
available, 1983, students in the Kivu province did considerably
‘better on the state examinations than did students elsewhere (40
'percent passing as compared to 31 percent overall). 1983 Kivu pass
Irates by option are as follows: math/physics (41 percent), biology/
(:hemistry (52 percent), teaching (32 percent), business (52 percent),
social work (35 percent), literary (66 percent). Rates by

asponsorship for the country in 1983 are: Government (27 percent),

126
Catholic (45 percent), Muslim (10 percent), Protestant (26 percent).
Overall pass rates have been relatively stable since 1979. Figures
from 1979 through 1983 are as follows: 42 percent, 41 percent, 40

percent, 30 percent, 31 percent.

Figures in the sample schools of the present study were
considerably higher than this for 1983 through 1986: 46 percent, 63
percent, 75 percent, 77 percent. These are much higher than the
national averages which were said to be under 40 percent nationally

for each of these years.

The fifteen sample schools were rated on the basis of overall
percentage increases per year and relative rank per year. Those
schools seen as good or improving were given a rating of 3. This
includes schools 1,2,3,4,7, and 8. Those schools for which no
pattern could be determined (i.e. they were either erratic or
average) were given a rating of 2. This includes schools 6,9,10,11,
and 14. The schools which remained low throughout the time period,
or which steadily declined were given a ranking of 1. This includes

schools 5,12,13, and 15.

Based on these ratings, a number of interesting correlations were
noted. Principals' emphasis on instructional leadership was not
shown to be correlated significantly to state examination results
(r-.0424, p- .232). Years of teacher experience and teacher quality
did, however, correlate highly, both at a significance level of .001

(r-.2115 and .5677 respectively). Possession of a textbook also

127
correlated highly (.2246), as did female gender (.3613).
Socioeconomic status proved to be highly significant (.5727 when
aggregated at the classroom level). All of these variables were
measured at a significance level of .001. This would indicate that
the examination system may be slanted in favor of those with higher
socioeconomic status. That is, they may obtain far better results
even though their level of academic achievement is only minimally

'higher.

Teacher characteristics proved to be of considerable importance
in predicting both examination and mathematics achievement results.
While this was not the focus of the study, it fits well within the
overall interpretive framework developed, since the effects of the
principals are seen as mediated through the teachers. It was
expected that teacher characteristics would account for more of the
variation in outcomes than would principal characteristics and this
‘has proven to be the case. However, the principals' effect on
:mathematics achievement remains high and significant even when
teacher characteristics are taken as covariates. This indicates that
‘while principals do affect student outcomes in part by selecting more
‘highly qualified teachers, the other factors indicated are clearly

important.

0 nts ‘n uc ona eadersh'
A number of the specific activities of principals which were used
in the construction of the instructional leadership scale have proven

to be significant predictors of mathematics and state examination

128
success and, to a lesser degree, the levels of student self-esteem.

These variables will be looked at briefly.

SUPERVISION OF INSTRUCTION

The number of classroom visits made by principals for the purpose
of supervising instruction correlates quite highly with mathematics
achievement at both the Grade 8 and the Grade 12 level (r-.20, p<.001
in each case). This variable is also highly correlated with the
number of faculty meetings which a principal will hold (r-.8077,
p<.001). It is negatively correlated with self-esteem (r--.l499,
p<.001). Principals who themselves teach are much more likely to

visit classrooms frequently.

There appears to be at least one other positive side effect of
frequent classroom visitation. At both the Grade 8 and the Grade 12
levels, there are many more textbooks in classrooms where the
principals visit frequently. Principals who are in the classroom
appear more sensitive to instructional needs. This may be as

important a result of supervisory visits as any.

One finding which has equity implications is that principals
appear to visit much more frequently in schools with a higher
socioeconomic level. This may reflect parental expectations. (See
Appendix K for a sample copy of the form used by one principal for

supervisory visits.)

129

FACULTY MEETINGS

The number of meetings held with faculty members is so highly
correlated with the number of classroom visits as to be almost
confounded with that variable. As is true for the number of
classroom visits, the number of faculty meetings is also highly
predictive of mathematics and state examination success. It is not
possible from the data to state with confidence which of these
variables is responsible for the effects measured (or if both are

responsible).

INSTRUCTIONAL OBJECTIVES

There is no school in the study which has developed explicit,
written instructional objectives. Most have rather developed lists
of teaching tasks which will be performed at specific points
throughout the year. Where principals meet regularly with faculty,
the school is much more likely to have given some thought at least to
instructional objectives (there is a correlation of over .7 between
the quality of instructional objectives and the number of faculty
meetings). The quality of instructional objectives also correlates
highly with state examination success (.5788 overall, although the
sample size is small and there may well be other confounding

variables).

TEACHING PRINCIPAL
Principals are much more likely to teach at schools with a lower
socioeconomic level. When the number of students wearing watches is

used to measure SES, the correlation found is —.4665 for Grade 8 and

130
-.5798 for Grade 12. There appears to be no negative effect on
mathematics achievement and may even be a positive one. No
particular effect was found for Grade 8 achievement, which is not
surprising since only one principal teaches any courses at all at
that level, and a positive correlation of .1851 for Grade 12. Any
positive effect may well be due to the increased involvement overall
of the principal with the instructional needs of the Grade level at

which s/he teaches.

SUBJECT MATTER KNOWLEDGE

This variable consistently failed to show significance in the
direction expected. This may have been due to the fact that it was a
very inexact measure (principals were asked to indicate what
mathematics textbook was used and what the general emphases of the
curriculum were). According to the data collected, principals with
greater subject matter knowledge are less likely to supervise
teachers and hold regular faculty meetings and their overall level of

discipline is lower.

PRINCIPALS' VIEW OF ROLE

Principals with greater experience and who meet regularly with
their faculty members are much more likely to view their role in
terms of instructional improvement. Where the principals describe
their roles in terms of instructional improvement, students perform

better on the state examinations (correlation of .5109).

131

DISCIPLINE

The measure of discipline used was highly subjective and based on
several hours of classroom observation only. Nevertheless, this was
enough in many cases to observe broad differences. In some schools,
there was very clearly an attempt to provide an atmosphere in which
students could concentrate on learning. In other schools, this was
clearly not the case. This broad measure showed correlations between
examination trends over the past four years, principal experience
(.6), teacher qualifications (.55), supervision of teaching (.44) and
classroom SES levels (.34). It also correlated positively with
student self-esteem (.14) at the Grade 12 level. There was no

significant correlation with mathematics achievement.

PRIORITY ON PURCHASE OF INSTRUCTIONAL MATERIALS

Principals were asked to indicate what their priorities for
school purchases would be if money were available. A stated priority
for instructional materials (textbooks, maps, chalk, etc.) correlated
negatively with all of the measures of socioeconomic status used. It
also correlated negatively with state examination results and most
measures of principals' emphasis on instructional leadership. It
showed no significant correlation with mathematics achievement. This
may mean that, contrary to expectations, principals who value
improved instruction see a greater need for improved facilities,
etc., rather than for instructional materials per se. The answers
given may also indicate a poorly chosen question, for principals
talking about purchases with unlimited money to spend tended to think

much bigger than books and chalk.

132
General observations
This section looks briefly at qualitative data and overall
impressions gathered during the visits. Observations are placed in
five categories: principals, teachers, students, building quality,
and miscellaneous. Within each of these categories, there is no

attempt at any tight logical ordering.

PRINCIPALS

1. Almost all principals in the sample schools paid a great deal of
attention to classroom visitation. This was universally held to
be an integral part of the principal's job and principals
commonly expressed regret that they did not have more time to
devote to it. Most principals discuss with the teachers what

they have observed. A few only prepare a formal report.

2. Most principals hold teachers' meetings on a very occasional
basis. There is widespread agreement that such meetings are
important. A common component of such meetings is for subject
area teachers to observe and discuss the presentation of a model

lesson.

3. Secondary school principals in the Butembo district tend to know
one another by name but only occasionally get together. Almost
all principals can name the majority of the thirty principals in
the district. Formal meetings organized by a higher level of
authority are rare; the principals themselves had organized one

meeting at a local bar, the last part of which the researcher

133
attended. In addition to three secondary school principals, two

elementary school principals were present.

End-of-year juries provide an opportunity for principals to learn
what their peers consider to be acceptable norms of student
performance. Juries typically consist of an inspector, at least
one visiting principal or teacher, the local principal and
several local teachers. The function of these juries is to grade
students on performance of a practical task--teaching a lesson,

repairing an engine, performing an experiment, etc.

Principals are commonly responsible for branch schools operating
at a distance. They will commonly spend several days a month
away at these branch schools (auccursales). Branch schools may
have well over a hundred students, but will not have all six

grades.

Principals generally are free to hire teachers if they can find
them. They usually rely on informal networking through family
and friends to do this. Most say that they know students at the

university and will contact them to offer jobs.

Principals are named by the sponsoring organization and
frequently belong to an ethnic group other than the Nande.
Principals in the government-sponsored schools may be named by
the local education officer or his superiors, or by the regional

commissioner, who is the top political officer in the area.

10.

ll.

12.

134

Principal turnover is high. Only one has been at a single school
for more than six years; only seven of the thirty in the Butembo
district have been at one school for more than four years. Over
half have been at their present school for less than three years.

The only expatriate principal in the sample, a Catholic sister
from Spain, expressed deep concern about the imbalance between
the curriculum and the milieu. She said, " Teacher salaries
don't allow them the dignity of their profession...classes don't

teach students what most of them will need in their lives."

Several days a month are spent by most principals going back and

forth to Butembo to arrange for the payment of teacher salaries.

Many principals at the teacher training schools have a close
involvement with at least one elementary school in the immediate
vicinity, which serves as a lab school. Some principals have

direct administrative responsibility for this school.

During the course of the interviews, a number of principals

expressed frustration at the systemic impediments to
instructional leadership which they constantly face. The
following comments, separated into general categories, are
typical:

Teacher turnover
"Each year I get a new team of teachers. They use this as a

starting point."

135
“Teachers around here are starting to look for jobs in

Rwanda."

Teacher engagement

"Some teach at neighboring schools but a problem with this is
that their loyalties are divided and they may give less

attention to their work at (this school)."

"I have to keep an eye on the teachers so they don't just sit

there."

"Teachers say, 'Let's eat first. We'll philosophize later.'"

Ema—mien

"Each month when I go for the teachers' pay, I am required to

leave a bribe."

Physical facilities

"We charged each student 30 zaires extra for construction and

we couldn't even finish a small office."

Methodolggy

"It really bothers me the way teachers just have students

copy the text from the board..."

136

BEES—011.1121

"The biggest problem here is finding qualified teachers."

"When it rains, the visiting teachers can't get here, so the

students have to study on their own."

13. Principals keep busy at a wide range of activities, but most have
a hard time defining just what it is they do.

"I don't have a secretary so I'm running in all directions.
I hardly know what job to start with in the morning."
"I combine every kind of work there is: office work,
monitoring students, supervising teachers, riding my bike
back and forth, going to Butembo for instructions, passing
along messages..."
"...church work, visits, committees, travelling..."

TEACHERS

1. Virtually all teachers supplement their income in one way or

another. The most common way of doing this is by gardening. In
many schools, teachers have insisted on a four-day work week to
allow time for this. Many teachers also receive payment from
students in various ways. It is considered acceptable for a
teacher to require students to help in his/her field. Income
from school gardens is frequently distributed among the teachers

as well.

137
Itinerant teachers (referred to as visiteurs) are commonplace.
Some teachers will teach at three different schools during the
week, although officially they are only placed on the salary
charts at one school. Mathematics teachers are particularly in
demand. Arrangements for payment vary. One mathematics teacher
works 22 hours a week at one school, 8 hours a week at a second
and 4 at a third. In addition to his regular monthly salary of
1173 zaires ($19), he receives 350 zaires ($5.65) at the second
school and 272 zaires ($4.40) at the third. He also gets 100
zaires ($1.60) in travel expenses from the third since it

involves a four-hour round trip by bicycle each week. When it

rains, he sleeps over. A teacher at a different school does a
little better. He works 20 hours at his own school and 8 hours
at another. He and another itinerant teacher share a full
salary. The other teacher works 9 hours and gets 9/17 of the

salary. He walks three hours one way.

There is evidence of a great deal of transferring between schools
within the district. Many teachers had taught at two or three

other schools.

The inspectors stated the following impediments to teaching in
the Butembo district: 1) the curriculum followed is too academic
and not adapted to local conditions; 2) textbooks which follow
the official curriculum don't exist; 3) lack of instructional
materials; 4) low morale because of the salary situation; 5)

teaching seen as a jumping off point (the metaphor of bouncing on

138
a trampoline was used); 6) the efforts required of teachers far
outstrip the rewards; 7) the lack of teacher training ("we need a
pedagogical reform"); 8) the lack of supervision given by principals
in terms ~of class visits (this was attributed partially to the
effects of recruitment irregularities: "teachers are chosen to be
principal because they're the brother of this coordinator or that
official or because whoever pays the most gets the job"); and 9)
students are allowed to pass when they should be kept back. The
inspectors speak nostalgically of "teaching like it was when teachers

were paid enough and worked well".

5. The teaching profession has been nicknamed '1'anteghambre'--the
waiting room. It is a professional holding tank, providing
contacts and often the opportunity to start out in business on

the side, while receiving at least a minimal income on a regular

basis.

6. A popular song in Zaire currently contains these words: "Don't
let your daughter marry a teacher. She'll surely die of
hunger." One principal reported asking the teachers to help come

up with a list of objectives for the school and the teachers
declined, saying: ”Let's worry about eating first; we'll

philosophize later."

7. Teachers teach an average of twenty to twenty-two hours a week.

There is no school on Sundays and teachers receive a second day

off known as a "journee pedagogigue" (teacher preparation day).

139
In many schools, teachers have insisted on an additional day off
to find ways to supplement their income. This makes scheduling
difficult and as a consequence, students have free periods almost
every day at some schools. In theory, no instructional time is
lost but in practice, teachers often combine two or more class

periods and teach it in less than the allotted amount of time.

8. See Appendix J for a list of the rules which teachers at one
school were expected to follow.

STUDENTS

1. When asked, students all state the same goal--to pass the state
exams (preferably with 60% in order to qualify for university
entrance). They are generally optimistic. A slogan on one
blackboard read: 19 diplomas with omnipotence. (All 19 students
did ultimately pass.) In spite of their optimism, there is a
great deal of gallows humor on the topic of exams and the
possibility of subsequent university attendance.

2. Students are subject to a variety of authoritarian requirements

at many schools. (See Appendix J.) A sign on one door gives a
list of cleaning duties with student names according to date,
then:

"N.B. I insist on the conscientious and regular execution of

these daily responsibilities: sweep, dust the classroom

 

furniture (desks, tables, chairs), remove the apider webs. The

140
classroom monitors will inform the office of any recalcitrant
students.

(Signature, date) The Principal"

Students are keenly aware of both poor quality teaching and poor
quality administration. Comments heard include: "How can anyone
learn anything with this little (i.e. underqualified) teacher?
He's just a D6 (high school graduate)." "Our biggest problem is
that the principal is not a ”debrouillard” (a resourceful

person)."

Students organize themselves into groups to prepare for the state
exams. Occasionally, the principals will also help to organize
these groups. The students are usually left free for several
weeks before these exams but this year was an exception because
the exams had been unexpectedly moved up by several weeks.
Students were concerned that this would not allow them time to
prepare. Invariably, they regarded the mathematics questionnaire
as an excellent form of preparation and asked for copies to study
from. They stayed long after they were done in order to ask

questions and discuss answers.

Girls are much underrepresented, particularly at the Grade 12
level. The justification given at the Catholic schools for
maintaining separate schools for girls was to enable them to have
a fair chance at schooling. Only one of the Catholic girls'

schools still refuses admission to boys, however.

141
A survey conducted by the principal at one of the Catholic girls'
schools asked: What sorts of things would you like to see taught
at this school? Answers included: how to keep a husband, how to

act so our husbands won't beat us.

BUILDING QUALITY

1.

Parents in most areas are responsible for financing new
construction. During an impromptu interview at one school, the
president of a parents' committee complained that the principal
kept all of the money they had raised for construction and used

it himself.

The Catholic schools are generally the highest quality, but there
is tremendous variation. Several schools are occupying buildings

made for other purposes and left behind by the Belgians.

Almost no money is available for maintenance. Even the best
schools need painting; broken windows are commonly replaced, if
at all, by a piece of plywood. One second-floor classroom, where
the buildings date only from 1954 but were abandoned for four
years, has a plank floor with gaping holes where the planks have
rotted away. The ceiling is sagging and broken away in several
large areas. Stains along one side of the ceiling and along one

wall indicate a serious leak.

Many classrooms have no teacher's desk or chair.

142
One classroom had a hole large enough to crawl through in one
wall. Two students at the back of an adjoining class knelt down
and watched the proceedings in the classroom where the researcher
was explaining the use of the questionnaire. They remained

there, undetected by their own teacher, for at least ten minutes.

Lizards are a fixture on the walls of some schools in the warmer
areas. Angola Swallows nest under the eaves of another and
constantly fly back and forth through the room, which has no

window panes.

Almost all schools have a flag; all schools start the day with a

flag ceremony and the singing of the national anthem.

It is uncommon for schools to share facilities but at one site,
three separate schools use the buildings each day: the secondary
school in the morning, a six-grade primary school and a two-grade
secondary school in the afternoon. Several schools have morning
and afternoon shifts, but teachers do not teach both. The

principal is not generally present during the afternoon shift.

Principals scrounge around for materials wherever they can. The
principal at one school sent a letter to a development agency out
with the researcher. The principal at a business school was able
to get several typewriters donated in 1981 and obtained three
more last year through political connections. Church

affiliations are important in the search for instructional

143
materials. The principal at one of the Baptist-sponsored schools
was able to secure a donation of about one hundred books (for
general reading) from a local Baptist businessman. The Catholic
schools have a much better access to textbooks than do the
others. These will not be sold to principals from the
non-Catholic schools so these principals will frequently make
arrangements for principals of Catholic-sponsored schools to make

the purchases for them.

MISCELLANEOUS

1.

Church sponsorship does not necessarily mean that personnel or
students will be predominantly from that church (although this
tends to be so especially for the Protestant schools and, to a
lesser degree, for the Catholic schools). At the only
Moslem-sponsored school in the sample, there were no Moslem
students or staff members. At a Kimbanguist high school (not in
the sample because it does not run a six-year program) which was

visited, only one of the teachers was Kimbanguist.

There has been a move toward decentralization of certain aspects
of the educational process in Zaire over the past few years. The
payment of salaries is the most obvious example of this. Until
1982, salaries were administered from Kinshasa and calculated by
teacher. There was always a delay of at least one year before
new teachers were paid and frequently, the names of individual
teachers would mysteriously disappear from the lists sent out by

the National Ministry, only to reappear several months later.

144
Salaries lost in this way were rarely recuperated. Now, salaries
are administered from the provincial capital, Bukavu, and a total
figure is calculated for each school. Individual teachers will
receive more or less depending on their qualifications and
experience, but all are paid regularly (if somewhat late) each
month. There is a widely expressed sentiment that
decentralization has been a good thing. There has been no
decentralization of curriculum development or the administering

of state exams, however.

There is a strong commitment in the Butembo district to
vocational schooling. A literary section in Beni is being
converted to general mechanics/ electricity. An automechanics
school started recently in Lukanga is being upgraded to six-year
status. Two four-year vocational schools operate in the
district. Parents give strong support to this move. This may be
because the trucking industry is one of the best-paying
industries in the district and a job as a mechanic or as a driver

is seen as a first step in owning a business.

The current mood regarding education in the Butembo district is
somber. Not one of the many individuals formally and informally
interviewed expressed any hope that things would improve at any
time in the foreseeable future. This sense of futility is the
second strongest impression to be gained in visiting with school
personnel in the area. The strongest impression to be gained is

that, for all of the problems, there remains an extraordinary

145
commitment to education. Students see it as the road to the
future; many teachers show a willingness to make almost
unimaginable sacrifices because they believe in what they are

doing; many principals do the same.

CHAPTER FIVE

SUMMARY OF METHODOLOGY, FINDINGS, CONCLUSIONS AND IMPLICATIONS

Periods of educational change make heavy demands

upon leadership. Programs for the education of
secondary school administrators and other
educational leaders are overdue in most African
countries.

-John Hanson (1971), p.22

Introduction

This chapter is divided into six parts: 1) a summary of the
methodology used in the study; 2) a summary of the major findings;
3) limitations of the study; 4) conclusions; 5) an interpretation
of the findings in the context of the literature review; and 6)

practical and research implications.

Summary of methodology

In this study, the emphasis which secondary school principals in
Zaire place on instructional leadership was examined in relation to
two outcomes of schooling: mathematics achievement and the
development of student self-esteem. Questionnaires were administered
to 1017 students; interviews were conducted with 26 teachers, 15
principals and 3 inspectors in the Butembo district of the Kivu
province in northeastern Zaire. A French/Swahili translation of the
Coopersmith Self Esteem Inventories (Short Form) was used to collect
self-esteem data. A French translation of selected items from the
Second International Study of Mathematics (International Association

for the Evaluation of Educational Achievement) was used to measure

146

147
mathematics achievement. Only Grade 8 and Grade 12 students were
included in the sample. Semi-structured interviews were conducted
with the principals, teachers and inspectors. Principals were rated

according to their emphasis on instructional leadership.

Summary of findings

The number system used in the summary of findings corresponds to
the numbering found for the hypotheses listed in Chapters 1, 3 and
4. Additional findings not specifically tied to given hypotheses but
arising from the developmental nature of the study are listed at the

end of this section.

1. The emphasis which principals place on instructional leadership
has a significant effect on student mathematics achievement,

particularly at the Grade 12 level.

2. The emphasis which principals place on instructional leadership
has a mild effect on the development of student self-esteem. A
positive effect was found for Grade 12 students and a negative effect

for Grade 8 students.

3. Principals' effect on mathematics achievement is generally
independent of the academic ability of students enrolling at the

school.

148
4. The emphasis which principals place on instructional leadership
does not have an effect on students which grows significantly
stronger the longer the student is at the school. (This finding may
result in part from an inability to disaggregate confounding

variables such as failure of courses and academic intake levels.)

5. Textbooks are associated with greater mathematics achievement
only for students following the Grade 12 sciences option and for
those students whose principals emphasize instructional leadership.
Fewer than 7 percent of Grade 8 students and 23 percent of Grade 12

students possess a mathematics textbook.

6. The student's socioeconomic status does not affect her/his level

of mathematics achievement.

7. The student's socioeconomic status does not affect his/her level

of self-esteem.

8. The socioeconomic level of the classroom has a small, positive

effect on mathematics achievement.

9. The socioeconomic level of the school does not in general affect

the student's level of self-esteem.

149
10. There is no statistically significant correlation between
mathematics achievement and student self-esteem for the schools

studied.

11. Schools where students have higher self-esteem scores did not
perform any better or any worse on state examinations over the past

four years.

12. Schools with higher mathematics scores achieved greater success

on the 1986 state examinations.

13. No significant correlation between gender and mathematics

achievement was found.

14. Gender differences in self-esteem levels were found only for the

Grade 12 sciences option, where females rate higher than males.

OTHER FINDINGS

a) The effect of principals' instructional leadership, while not
dependent on either the academic intake levels of the schools or the
socioeconomic status of the school, may nevertheless be influenced by
these variables. Principals in schools with either higher academic
intake levels or higher socioeconomic status tend to emphasize

instructional leadership more than do principals in other schools.

150
b) National examination results are significantly correlated with

teacher experience and qualifications.

c) Gender differences are particularly noticeable in terms of access
to schooling, with females much less likely than males to complete

twelve years of schooling.

d) The level of the mother's education is highly predictive of

whether females will remain in school.

e) The number of classroom visits and the number of faculty meetings
correlate highly with mathematics achievement and pass rates for the

Grade 12 national examinations.

f) Schools applying stricter admissions criteria achieve higher

rates of success on the national examinations.

g) The female principals in the study emphasize instructional
leadership much more than do the male principals; the schools run by
female principals consistently outperform the schools run by male
principals both for mathematics achievement and pass rates for the

national examinations.

h) Strategies used by principals to cope with resource scarcity
include: 1) hiring itinerant teachers, 2) granting a four-day work
week to enable teachers to obtain supplementary employment, 3) an

increased teaching load, 4) matabiahi (unauthorized payments

151
solicited from students or parents), 5) increased study time for
students under the supervision of a monitor, 6) solicitation of

financial and material support from the community.

Limitations of the study

The exercise of instructional leadership on the part of
principals is necessarily linked to teacher activities. An important
limitation of this study, however, is that little information was
gathered about teachers and teacher activities. The two teacher
variables about which data were collected, years of teacher
experience and level of teacher qualifications, both proved to be
consistently correlated with mathematics achievement and state
examination pass rates. It can be assumed that closer attention to
teacher activities would have further illuminated the leadership role
of the principal. (High turnover rates and low teacher morale are
examples of two teacher variables which appear of importance to the

topic under study but which were not dealt with at length.)

An important limit to the generalizability of the study is that
the Butembo district appears to be highly atypical, at least insofar
as pass rates on the Zairian national examinations are concerned.
Pass rates in this district have the reputation of being, and were
indeed found to be, much higher than the national average. While
this may lessen the possibility of generalizing to other Zairian
schools, however, it also increases the likelihood of finding and

describing effective management patterns and practices.

152
One further limitation of the current study is the generally low
level of reliability found for the self-esteem questionnaire
(Cronbach's alpha equal to .40 and .49 for Grades 8 and 12
respectively). This may, at least in part, account for the generally
low correlations found between self-esteem levels and a broad range

of independent variables.

A final limitation is the cross-sectional nature of much of the
study, which does not permit causal inferences based on changes over
time. This has particularly restricted the discussion of academic
intake. Also, a lack of time spent observing individual schools and
principals limited knowledge about the degree of actual instructional
leadership exercised by the principals. Because of this, it has only
been possible to measure the emphasis which principals place on
instructional leadership. Placing an emphasis on instructional
leadership and actually exercising it are two separate constructs,

although they may be presumed to be highly correlated.

Emissions
This section has been kept brief by conscious design. The study
was exploratory in nature and it would be presumptive at this point

to draw other than preliminary conclusions.

1. Direct principal/teacher contacts appear to be by far the most
important component of instructional leadership in Zaire. The number
of classroom visits and the number of faculty meetings are both

consistently correlated with important instructional outcomes.

153
2. Effectiveness in promoting instructional goals may not
necessarily lead to the attainment of valued non-cognitive outcomes,
particularly where equity concerns are given low priority. There is
evidence, for instance, that Grade 8 students, who tend not to
receive the particular attention of the principal and for whom
teacher expectations may be lower, actually develop lower self-esteem

at schools where principals emphasize instructional leadership.

3. The effect of out-of—school factors on mathematics achievement
appears to be relatively small; the effect of schooling and school

characteristics appears to be relatively great.

Interpretation

The finding that principals who emphasize instructional
leadership have a significant influence on mathematics achievement is
consistent with the effective schools research. The evidence
suggests that the sorts of bargaining described in much of the
current literature on schools as non-rational organizations are not
characteristic of schools in Zaire, where staff and students alike
are generally in agreement concerning broad instructional goals. The
high dropout rate which is not only tolerated but frequently
encouraged increases the likelihood that those who remain in the
system will form a relatively homogeneous group with respect to

overall educational goals.

The finding that the socioeconomic status of individual students

does not account for a significant degree of the variation in

154
mathematics achievement is consistent with the findings of Heyneman
(1983, 1984) and Currie (1977), but not with the findings of Mukweso
et al. (1979) and Sheline et a1. (1979). These findings are also
counter to the findings of Simmons and Alexander (1978). The finding
that classroom SES levels do appear to affect achievement, on the
other hand, suggests that a more complex multilevel model of school
and family effects may be able to reconcile some of the apparent
differences described in the literature. Even when classroom SES
levels are considered, however, the overall impact on achievement is

relatively low.

The difference between the findings dealing with SES in this
study and the data described by Mukweso and Sheline may be that those
studies were national in scope and would have included a much more
diverse group of students in terms of SES. While variations in SES
certainly do exist in the Butembo district, there is relative
homogeneity, particularly with respect to cultural and economic
opportunities. Socioeconomic status may also have a more highly
differentiated impact in urban environments. It may also be that
there exists a threshold level beyond which socioeconomic factors
become important; this level may not have been reached as yet for
most students in the Butembo district. A final factor rendering the
effects of SES difficult to gauge is that higher socioeconomic status
in some areas of the Butembo district appears to lead to more
employment opportunities without the need for schooling. This may
mean that students with higher socioeconomic status are placing their

energies in other than scholastic endeavors.

155

The finding that academic intake does not appear to significantly
influence academic achievement may be reflective of the generally low
level of academic intake more than anything else. At the one school
where the students were clearly more academically capable coming in,
they were also more academically capable going out. There is
evidence, however, that principals in schools with low academic
intake and low socioeconomic status were able to positively influence
student outcomes. School 6 ranked among the lowest in both of these
constructs, yet the principal had inaugurated the most ambitious
program of any of the schools to emphasize instruction and the Grade

8 students scored higher than any others in the study.

A caveat is in order at this point regarding the interpretation
of many of these findings. Zaire's teachers are among the lowest
paid in the world (averaging $12 to $15 monthly) and morale is almost
universally low. This may mean that under these circumstances, the
level of teaching is so low that a principal who rises "above the
pack" and succeeds in achieving a consensus as to instructional goals
may have a differentially greater impact than in countries where the
level of teaching is already higher. Conversely, the impact of
principals may be obscured by a number of confounding variables which
were not controlled for in this study. This certainly appears to be
the case at School 15, for instance, where the principal received the
highest rating in the sample but where teacher turnover is also
higher than elsewhere in the sample because of relatively attractive

career alternatives.

156

Mime

RESEARCH
l. The overall context of instructional leadership in Zaire
remains unclear. A priority of future research must be to determine

which leadership practices and processes are most associated with
improved student outcomes. The complexity of schooling in Zaire
means that these studies would probably be of more value if the
interactions of inputs and processes are examined rather than an
approach which attempts to identify discrete variables or inputs

which affect student outcomes.

2. Studies employing a pre-test post-test design are needed to
provide a more adequate control of academic intake at the schools and
of the differential impact of instructional leadership on

identifiable subgroups of students.

3. Supervision of instruction is one area which provides great
research potential since it is widely practiced in one form or
another, with principals generally incorporating supervision into
their understanding of what it is that principals "should" do. A
worthwhile area of investigation is the differential impact of
supervision when it is used for control and when it is used for
instructional improvement (or when it is used for both, which may be

more typical).

4. In this study, principals were found to actively screen

student admissions only on the basis of academic achievement. This

157
will certainly mean that students from regions with lower quality
primary schools will be restricted from secondary schooling. It may
also result in lower female participation in schooling. Research is
needed to determine whether the principals' role in promoting or
failing to promote equity is consistent with national objectives of

increasing access to schooling.

5. The relationship between instructional leadership, student
outcomes, and subsequent employment opportunities has not yet been
fully explored in the context of a developing country. If principals
are ”effective" in promoting academic achievement but not "effective"
in promoting subsequent employment, there may be a legitimate
question as to whether this is the sort of effectiveness which is
needed. Again, this points up the need for research which takes into

account the multiple objectives of schooling.

6. The principals' role in recruiting and retaining teachers in
Zaire merits careful research attention. Are there ways in which
principals creatively "bend the rules" to facilitate teachers'
collaboration in the attainment of instructional objectives? The
four-day work week, the hiring of itinerant teachers and the
rearranging of schedules at the larger schools to provide a salary
for a vice-principal of instruction are three ways which come to
mind. What are the positive and negative effects of these

practices? Does the good outweigh the bad?

158
7. What is the role of matabishi in secondary school
administration in Zaire? Research into this sensitive area will need
to investigate the impact of natabishi on both equity and the quality
of teaching. Are there only negative impacts? Does it serve as a
form of unofficial privatization? What sorts of implements do policy

makers control in order to change current practices?

8. The cost-effectiveness of pre-service and in-service

training programs for principals merits serious research attention.

9. Research is needed into how principals can manage innovative
approaches to instructional improvement in times of extreme resource
scarcity and increased demographic pressure. Facility-sharing,
multiple shifts, every-other-day instruction, and cross-age peer
tutoring are several approaches which may be worthy of research

attention.

10. More carefully refined measures are needed to shed light on
how schooling contributes to the development of self-esteem and other

non-cognitive outcomes.

11. Ethnographic studies of outlier schools are potentially of
considerable value. In the present study, for instance, Schools 3
and 4 have shown consistent stability of academic excellence over the
past four years. One of these schools is low SES and one is high
SES, providing an opportunity for meaningful contrasts and a deeper

understanding of instructional leadership under varying conditions.

159
The principal of School 6 has shown a remarkable commitment to
instructional innovation with apparent success. Long-term results
are well worth researching. Also, the pilot school for this study
has demonstrated high self-esteem levels and high mathematics
achievement, along with consistently high national examination pass
rates. The explanatory potential of ethnographic studies at these
and other schools could lead to a greatly enhanced understanding of

instructional leadership and effective schooling in Zaire.

Pr ical m catio

1. It would seem reasonable to begin programs at the tertiary
level to train principals in Zaire. In a situation of intense
resource scarcity, it may not be possible to give all teachers an
optimal level of training, but it may be much more cost-effective to
train a principal and give him/her responsibility for the

instructional organization of the school.

2. Increasing the opportunity for sharing knowledge and ideas
through workshops and in-service programs is a potentially
cost-effective means of improving student outcomes by improving
principal capabilities. Even the publication and diffusion of a
simple newsletter dealing with effective instructional leadership

would be helpful.

3. Virtually all principals have access to radios--regular
radio programs could serve to update principals in terms of trends in

supervision and instruction.

160
4. A strengthening of the collaborative role of the inspectors
would be one means of improving instruction and providing the
principals with guidance and feedback. Inspectors could also

organize seminars for principals and teachers, etc.

5. A decentralization of the curriculum would afford
principals, teachers and inspectors an opportunity to design a
program adapted to the varying realities of life in the context in
which students study. This would also provide for greater parental

input.

6. A focus on the contact points between principals and
teachers may hold the promise of further improving instruction. The
forms currently used widely for instructional supervision, for
instance, could be re-evaluated and re-designed in the light of
research findings to provide greater assistance and support for

teachers.

7. Women are very much underrepresented at all levels of
education in the Butembo district and, indeed, throughout Zaire, yet
their contribution far outweighs their numbers. They represent a
vastly under-utilized resource in an area where resources are
scarce. A greater concern for equity here may lead to quality

payoffs as well.

APPENDICES

APPENDIX A

Instruments used for data collection

Demographic questionnaire (students)

Coopersmith Self-Esteem Inventories (French, Swahili)
Mathematics questionnaires

Principals' interview schedule

Teachers' interview schedule

161

DEMOGRAPHIC QUESTIONNAIRE (Students)

The demographic questionnaire, the self-esteem questionnaire, and the
mathematics questionnaire were stapled together and completed by the
students at one sitting. Note that the numbering system is
sequential and that the last of the demographic questions followed
the mathematics questionnaire. It was numbered 71 on the Grade 8
questionnaire and 65 on the Grade 12 questionnaire. Individual
questions appear here exactly as on the questionnaires filled out by
the students, but there are differences in pagination.

INTRODUCTION

Le questionnaire qui suit est en deux parties. I1 vous faudra
environ une heure pour le compléter. La premiEre partie vous posera
des questions a propos de vous-meme--qu'est-ce qui vous plait,
qu'est-ce qui ne vous plait pas, etc. La deuxieme partie est un
examen de mathematiques. Puisque cette partie est tirée d'un examen
international qui a eté donné en plus de 20 pays, il peut y avoir des
questions sur des sujets que vous n'avez pas etudiés. Ceci est tout
a fait normal. Veuillez continuer avec les questions auxquelles vous
pouvez repondre. “

Votre participation est volontaire et vous avez 1e droit d3 ne pas
continuer. Ce test ne constitue pas une partie du cours regulier et
n'aura aucune influence sur les c3tes. Si vous ne voulez pas
continuer, il n'y aura aucune peine. Si vous acceptez de continuer,
vous avez 1e garanti de l'anonymité. Veuillez ne pas mettre votre
nom sur le questionnaire ou sur la feuille des reponses.

Les resultats de cette etude contribueront a une meilleure
comprehension des ecoles secondaires au Zaire. Merci d'avance pour
votre cooperation. Quand vous aurez termine, vous pouvez gardet'le
crayon en guise de remerciement. Veuillez commencer maintenant avec
le numero 1.

1. Indiquer votre sexe
0. feminin 1. masculin

2. Indiquer votre age

17 ans ou moins
18 ou 19 ans

20 ou 21 ans

22 ou 23 ans

24 ou 25 ans

. plus de 25 ans

U|§UNH°

162

3. Indiquer 1'éducation du pEre
0. 1 an ou moins

2 a 4 ans

5 cu 6 ans

7 a 9 ans

10 a 12 ans

plus de 12 ans

U'IbUNI‘

r l'éducation de la mere
1 an on moins
2 a 4 ans
5 ou 6 ans
7 a 9 ans
10 a 12 ans
plus de 12 ans

4. Indiq

PFPPEPE

5. A la fin de juin, cela fera combien d'années que vous aurez
etudie a cette école?

O. 1 an 4. 5 ans

1. 2 ans 5. 6 ans

2. 3 ans 6. plus de 6 ans
3 4 ans

71. Est-ce que vous avez le texte de mathematiques?

0. oui
1. non

PLEASE NOTE:

COpyrighted materials in this document
have not been filmed at the request of
the author. They are available for
consultation, however, in the author’s
university library.

These consist of pages:

Coopersmith Self-Esteem Inventories ( French translation)

 

P. 163-180

 

 

 

 

 

University
Mic rofilms

lnte rnational
300 N Zeeb Rd., Ann Arbor, MI 48106 (313) 761-4700

163

COOPERSMITH SELF-ESTEEM INVENTORIES (French translation)

Following is the French-language translation of the Coopersmith Self-Esteem
Inventories used by all students in the study. It is followed by the Swahili
version which was used orally with the same students to improve comprehension.

Coopersmith Self-Esteem Inventory
Author: Stanley Coopersmith
Copyright date: 1981
Publisher: Consulting Psychologists Press, Inc.
Translator: R. Prouty
t date: 86

Pour les prochaines 25 questions, mettez '0' si la phrase vous decrit et '1' si
la phrase ne vous décrit pas. C'est-a-dire, 0 31 vous dites, 'Oui, je suis
comme cela' et 1 si vous dites, 'Non, je ne suis pas comme cela'.

6. D'habitude, les choses ne me gEnent pas facilement.

O. oui 1. non

7. C'est difficile pour moi de parler devant toute la classe.
0. oui 1. non

8. J'ai beaucoup de traits que je changerai‘si possible.
0. oui 1. non

9. Je peux faire des decisions facilement.
0. oui 1. non
10. Les gens aiment Stre avec moi.

0. oui 1. non ‘
11. Je suis souvent mecontent(e) chez moi (a la maison).
0. oui 1. non
12. Je prends beaucoup de temps avant de m'habituer 5 ce qui est nouveau.
0. oui 1. non
13. Je suis tres populaire avec ceux qui ont 1e mEme age que moi.
0. oui 1. non
14. D'habitude, ma famille prend mes sentiments en consideration.
0. oui 1. nod

15. Je cede trEs facilement aux autres.
0. oui 1. non
l6. Mes parents s'attendent a trop quant 5 ce que je ferai de ma vie.

O. oui 1. non

l7. Vivre comme moi, c'est tres difficile.
0. oui 1. non

18. Les choses sont trés troublées dans ma vie.
0. oui 1. non

19. D'habitude, les gens suivent mes idées.
0. oui 1. non “

20. Je n'ai pas une bonne opinion de moi-meme.
0. oui 1. non

21. 11 y a beaucoup de fois que j'ai voulu quitter la maison pour du bon.
0. oui 1. non

164

22. Je suis souvent malheureux (malheureuse) E l'ébole.

0. oui 1. non
23. Je suis mains beau (belle) que la plupart des autres.
0. oui 1. non
24. 8'11 y.a quelquechose que je veux dire, d'habitude je le dis sans
hesiter.
0. oui 1. non
25. Mes parents me comprennent.
O. oui 1. non
26. La plupart des gens ont plus d'amis (amies) que moi.
0. oui 1. non
27. Je sens souvent que ma famille me pousse trop a réussir.
0 . oui l . [non ’ \ I
28. Je suis souvent decourage(e) a l'ecole.
0. oui 1. non I .
29. Je pense souvent que ce serait mieux si j'etais quelqu'un
d'autre.
0. oui 1. non
30. On ne peut pas avoir confiance en moi.
0. oui 1. non

COOPERSMITH SELF-ESTEEM INVENTORIES (Swahili translation)

Following is a translation of the Coopersmith Self-Esteem
Inventories into Kingwana, the dialect of Swahili used in eastern
Zaire. Because of the relatively low level of French comprehension
of the Grade 8 students in particular, each question was read in
Kingwana after being read in French. Students did not have printed
copies of the Kingwana translation. For the three questions where
further explanations in Nande were given, these follow the Kingwana
version in square brackets.

Kwa maswali 25 inayofuata, muweke '0' mukiona kama swali ile ni kweli
kwako. Mukiona kama swali ile si kweli kwako, muweke '1'. Maana
yake, muweke '0' mukijisemea, 'Ndiyo, ninajiona vile' na muweke '1'
ikiwa munajisemea, 'Hapana, haiko vile kwangu.‘

6. Zaidi, mambo ya maisha haiwezi kunihuzunika sana.
[ebindu by'eribyaho sibyangatoka erinyitsandya obuli ndambi]

7. Kusema mbele ya wanafunzi wote, ni nguvu kwangu.

8. Runs mambo mingi kwangu (maana yake, kwa maisha yangu) ningegeuka
ningaliweza.

9. Ninaweza kukata shauri (au kuchagua) bila shida.
10. Watu wanapenda kuwa pamoja na miye.

11. Mara mingi, ninajisikia mubaya mu nyumba (maana yake, sifurahi.)

12.
13.
14.
15.

l6.

17.
18.

19.

20.

21.

22.
23.

24.

25.

26.

27.

28.

29.

30.

165
Inanifaa muda murefu mbele ya kuzoea vitu vipya au watu wapya.
Ninapendezwa na wale wenye umri yangu.
Zaidi, jamaa yangu inaweza kuuliza mawazo yangu.
Wengine wakiuliza kitu fulani, ninaweza kukubali tu.

Wazazi wangu wanakuwa na mawazo juu zaidi kama nitafanya nini na
maisha yangu.

Kuishi saa miye, iko nguvu sana.
Mambo ya maisha yangu, iko nguvu sana.

Zaidi, watu wanafuata mawazo yangu. [abandu bakakwama
amalengekanya wage]

Ninapojaribu kufikiri juu yangu, au juu ya maisha yangu, ninawaza
kama si vizuri sana.

Kuna mara mingi nilipotaka kutoka mu nyumba na kuwenda tu, bila
kurudia tena.

Mara mingi, sifurahi mu masomo.
wengine wananipita kwa uzuri wa uso.

Ikiwa kuna kitu fulani ambacho ninataka kukisema, basi, niko
naisema tu, bila kusitika.

wazazi wangu wananisikia.

Zaidi, wengine wako na rafiki mingi kupita miye. [abandi bawite
abira bangi kwilaba ingye]

Ninasikia mara mingi kama jamaa yangu inanishukuma muno ili
nipate kuweza.

Mara mingi, mu masomo, ninasikia kama siwezi.

Mara mingi, ninawaza kama ingekuwa vizuri zaidi niwe mutu
mwengine, maana yake, nisiwe mimi mwenyewe.

Watu hawawezi kuniamini.

166

MATHEMATICS QUESTIONNAIRE (Grade 8)

“MATHEMATIQUES A

31. 2 matres + 3 millimetres -

2,0003 metres
2,003 metres
2,03 metres
2,3 metres

5 metres

#UNHO

32. 1/5 est égal h

0. 0,20%
l. 2%

2. 5%

3. 20%
4. 25%

33. Si 5x + 4 - 4x - 31, alors x est egal a

0. -35
l. -27
2. 3

3. 27
4. 35

34. Une mire a mis quatre bols de riz sur la table. Chaque bol
contenait 1 litre. Quand la famille a termine son repas, il restait
1 bol vide, 2 bols remplis a moitié'et un bol rempli a trois-quarts.
Combien de litres de riz la famille avait-elle mange?

0. 3 3/4

1. 2 3/4

2. 2 1/2

3. 1 3/4 I

4. Aucune de ces reponses.

35.

 

Laquelle des réponses suivantes donne 1a meilleure approximation
de 1'aire du rectangle?

0. 48 m2
l. 54 m2
2. 56 m2
3. 53 m2
4. 72 m2

167

36.

 

L'aire de la figure indiquée en noir est

23 carrés
20 carrés
18 carrés
15 carres
12 carrés

p o R U V

waHO

 

37.

 

 

 

 

Milo X"

 

 

 

z T
Si 1a figure indiquée est pliée afin de former un cube, quels
deux coins toucheront 1e coin P?

0. les coins Q at S
l. les coins T et Y
2. les coins W at Y
3. les coins T at V
4. les coins U at Y
A a

38. l——*.
inmum -

p . Q

__l

 

‘

La longueur de XE est 1 unite. Quelkest la longueur de P6?

0. 2 unites
l. 6 unites
2. 10 unites
3. 14 unite}.
4. l8 unites

O

 

39‘ illllllllllllllllllllllllllllllll’llillllllllllllllllllll

Sur l'échelle en haut, 1a fleche indique un point entre

 

 

0. 51 et 52
1. 57 et 58
2. 60 et 62
3. 62 et 64
4. 64 et 66

168

40. Un cube solide en plastique pese l gramme. Chaque cbté de ce
cube mesure 1 centimetre de longueur. Combien pesera un cube solide
en plastique semblable si chaque cote mesure 2 centimetres de
longueur?

0. 8 grammes
1; 4 grammes
2. 3 grammes
3. 2 grammes
4.1 gramme

41. Sur une droite, deux points A et B sont donnes. La coordonnee
de A est -3 et la coordonnee de fi_est +7. Quelle est la coordonnee
du point C, si B est le mi- -point du segment AC?

0. -13
1. -1/2
2. +2
3. +12
4. +17

42. Un peintre doit melanger les couleurs vert et jaune dans la
proportion de 4 a 7 afin d'obtenir la couleur voulue. 3' il a 28
litres de peinture verte, combien de litres de peinture jaune
devra-t-il ajouter?

0. ll
1. l6
2. 28
3. 49
4. 196

43. 51 P-LW et 31 P- 12 at L- 3, alors West egal A

0. 3/4
1. 3
2. 4
3. 12
4. 36

44. Un bateau construit en miniature est 1/10 de la longueur‘du
bateau original. Si 1a largeur du bateau original est de 4 metres,
la largeur du bateau en miniature doit etre

0. 0,1 metre
l. 0,4 metre
2. 1 mbtre
3. 4 metres
4. 40 metres

45.

46.

47.

48.

49.

169

La valeur de 0,2131 x 0,02958 est environ

l-‘le-‘O

00000

0000‘»
cm
0‘

006
0006
<-2> x (-3) -
0. -6
l. -5
2. -l
3. 5
4. 6
Lequel des angles suivants est aigu?
. 2. \<:
O. @<

3.

Si 23 - 0, alors x est egal a

 

12

0. 0
l. 3
2. 8
3. 12
4. 16

R
I“
8

La circonference du cercle avec centre 0 est 24, et la longueur

de 1'arc RS est 4.Que11e est la mesure en degres de l'angle ROS ?

50.

O. 24
l. 30
2. 45
3. 60
4. 90

Dans une competition, quelqu' un a jeté une pierre 61,60 metres.

Quelqu' un d'autre a jete une pierre 59,72 metres. Combien plus loin
est- -ce que la premiere pierre etait jetee?

1,12 metres
1,88 metres
1,92 metres
2.12 metres
121,32 metres

§UNH°

170

51.

«r
I 3 G I
Dans 1e diagramme en haut, les triangles ABC et DEF sont des

ho‘

 

equivalents, avec BC - EF. Quelle est la mesure de l'angle EGG?
0. 20°
1. 40°
2. 60°
3. 80°
4. 100°
52.

 

I \
x est egal a

 

 

 

 

 

 

 

 

 

0. 75
l. 70 :0
2. 65 ;, ..; 32.,r.:
3. 60 ..'.'...ee a....:. e‘
4 40 ~..’.". "..'.'.-J
53. . .e.e..h.fl . '0. 9.: .0"... 15 .
Un carre a eta enleve du fectangle.ci- haut. Quelle est 1'aire

qui reste?

o. 315 m2
1. 300 m2
2. 284 m2
3. 80 m2
4. 16 m2

54. Le tissu est vendu par le metre carre. Si 6 metres carres de
tissu coutent 480 zaires, alors le prix de 16 metres carres sera

#taJNi-‘o

55.
degres.

1.280
1.440
2.880
5.280

zaires
zaires
zaires
zaires

12.800 zaires

La temperature de 1' air an pied d'une montagne est de 31
En haut de la montagne la temperature est .7 degres.

Combien plus chaud est 1' air an pied de la montagne?

#th-‘O

~38 degres
~24 degres
7 degres
24 degres
38 degres

171

56. 0,40 x 6,38 -

0. 0,2552
1. 2,452
2. 2,552
3. 24,52
4. 25,52

57. Un commerpant possede x kilos de the. 11 vend 15 kilos et puis
il repoit 2y kilos. Combien de kilos possede-t-il maintenant?

 

 

0. x - 15 - 2y
1. x + 15 + 2y
2. x - 15 + 2y
3. x + 15 - 2y I
4. Aucune de ces reponses O/’, ’f/ /
58. .22258
//

 

 

 

 

 

Dans la figure, les petits carrés sont tous de la meme grandeur
et 1'aire du rectangle entier est egal a 1. L'aire des carres noirs
est égal a:

0. 2/15
1. 1/3
2. 2/5
3. 3/3
a. 1/2

59. D'habitude, 1a distance entre deux villes est mesurée en:

0. millimetres
l. centimetres
2. decimetres
3. metres

4. kilometres

60. Le tableau en bas compare 1a distance qu'une balle tombe (d) et
la distance qu'elle rebondit (r).

 

d 50 80 100 150
.r 25 40 50, 75

 

 

 

 

Quelle equation decrit cette relation?

_d2
2d
d/2
d+25
d - 25

#uNi-‘O
HHNHH

172

61. 2/5 + 3/8 -

5/13
5/40
6/40
16/15
31/40

bullet-‘0

52. 7 3/20 est egal 5

bunt-Io
\J\J\J\J\J
d‘uanahICD

63. Dans une ecole de 800 eleves, 300 sont des garpons. Le nombre
de garpons par rapport au nombre de filles est

0. 3 : 8
l. S : 8
2. 3 : 11
3. 5 : 3
4. 3 : 5

64. 20 est quel pourcent de 80?

0. 4% l. 20% 2. 25% 3. 40% 4. Aucune de ces reponses

65. La phrase " un numéro x diminué par 6 est moins de 12” peut etre
ecrite comme l'inégalité suivante:

0. x - 6 > 12
l. x - 6 2 12
2. x - 6 < 12
3. 6 - x 2 l2
4. 6 - x < 12

66. 30 est 75% de quel numero?
0. 40 1. 90 2. 105 3. 225 4. 2250

67. Lequel des points A, B, C, D, E sur la droite correspond a 5/8?

A 8 c 0 E 1
AA 1 1 A 1 AA 1 I. l 1 l 1..
v V U V I Y I I V’-

l L
' V'Y Viv

0 l

0. point A 1. point B 2. point C 3. point D 4. point E

173
68. 20% de 125 est egal 3

0. 6,25 1. 12,50 2. 15 3. 25 4. 50

69.

1
T-
4!-

Quelles sont les coordonnées du point P?
0. (-3,4) l. (-4,-3) 2. (3,4) 3. (4,-3) 4. (-4,3)

70.

 

3 .. - .9

Les triangles PQR et STU sont semblables. Quelle est la
longueur de SU?

0. 5 l. 10 2. 12,5 3. 15 4. 25

FIN

174

MATHEMATICS QUESTIONNAIRE (Grade 12);

MATHEMATIQUES B

31. a - (b + (c - d))

- +
I

I
O‘U‘O‘
I
O O 0 0
90-0-0-

+ o + I

- b + I
aucune de ces reponses

$‘h’hih‘C’
9 b 9 9
I

32. La courbe y - 3x(x - 2)(2x + 1) intersecte l'axe des ordonnees
(x) uniquement aux points suivants:

('2. 0) et (1/2. 0)
(29 0) 8t (’1/29 0)
(3, 0) et (-2, O) at (1/2, 0)
(3, 0) et (2, 0) et (-l/2, 0)
(0, O) et (2, 0) at (-1/2, 0)

F‘UahJF‘CD

33. Quelle est la transformation complexe associée a la similitude s
directe telle que: B - s(A) et 0 - s(B)

avec A(0,1) B(-l,0) C(0,-3)

z' - (l + 21) z + 1 - i
z' - (l - 21) z + 1 + i
z‘ - (1 - i) z + 1 + 21
z' - (1 + 21) z

.' aucune de ces reponses

C‘U’hDP‘CD

34. Les fonctions f et g sont dgfinies par f(x) - x - 1 2
et g(x) - (x + 3)

g(f(x)) est egal L

(x - 1)£x + 3)2
(x + 3) - 1
(2x - 252
(§+2)

x + 8

fi'h’hJF‘CD

175

35. Le diagramme montre 1e graphique de la fonction cubique f.

(-102)
y FL
.] 0
(19-2)

La seule equation possible qui decrit cette fonction est la
suivante.

 

0
1.
2. x33 - 3x
3 3§ - x
4 x + 3x2

36. Quelle est la somme de la serie geometrique infinie qui suit?
1 - 1/2 + 1/4 - 1/8 + ...

5/8
2/3
3/5
3/2
an:

kwNHO

' D“ y-lx est Egal a

37. -
(lx‘olr

0 -2

1. -l

2. 2

3 1n 2

4 3 1n 3

38. Equation A y - 6x - x2
Equation B y - 2x I
Si 0 < x < 5, pour quelles valeurs de x est-ce que l'equation B

sera plus grande que l'équation A?

0. 0 < x < 4
l. 0 < x < 5
2. 3 < x < 5
3. 3 < x < 4
4. 4 < x < 5

176

39. Un cote d'un triangle equilateral est sur l'axe des ordonnees.
La somme des pentes des trois cotes est.

0
-1
1

2:2]?

40. Pour l'equation x2 - 5x + 6 + 0

#le-‘O

il n'y a aucune solution

11 y a exactement une solution

i1 y a exactement deux solutions
11 y a exactement trois solutions
11 y a plus de trois solutions

9WNHO

41. Le symbols P/lQ represents l'intersection des ensembles P et Q,
et 1e symbole PtlQ represents 1'union des ensembles P at Q. Lequel
des suivants représente ce qui est en noir ici-bas?

 

 

 

 

 

 

0. (PnQ)UR
l. PU(QnR)
2. PfHQUR)
3- (PAQM R
4- (PUQNIR

42. P est un polynome an x de degre m, et Q est un polynome en x de
degre n, avec n < m. Le degre du polynome (P + Q) (P - Q) est

0. 2

1. m2

2.

3.29
4. m2 - n2

43. Quelle est l'equation de la droite qui passe par la point (0,
-5) et qui est parallels a la droite representee par l'equation y -
2x + 3?

x+2y+5-_0
2x - y - 5 - 0
2x + 3 - -5

2x - 5y + 3 - 0
2x + y + 5 - 0

”UNI-‘0

177

44. Lequel des suivants est negatif?

0. sin 51'";
1. sin 5y!
2. tg {jPflb
3. tg :gfikq
4. cos 5-

/‘2
45. Lesquels des suivagts, (x - 1), (x - 2), (x + 2), (x - 4) sont
des facteurs de x3 - 4x - x + 4?

(x - l) seulement
(x - 1) at (x + 2)
(x - 2) at (x + 2)
(x + 2) at (x - 4)
(x - 1) et (x - 4)

¢~uah>hdc>

46. Le nombre complexe (l + i)2 est egal a

i
+ i
+ 2

$‘uah)r*C>
NHNNO

i

47. Si cosy - 1/2, alors cos 2 flest e’gal 2

0. -1/2
1. 1/2
2. - 3/2
3. 3/2
4. 1
. + + r ‘
48 x1,1.0":111-233—-}|-}—-(-"—-J'-)-3x - 2 est egal a

0. -1/2 1. 2/3 2. 1 3. 5 a. (>49
49. La pente de la droite qui passe par les points (-1, 3) et (4, -7)

CSC
0. -1/2 1. -3/4 2. -4/3 3. -2 a. -10/3

50. La courbe ici 3 cat; represente
1aquelle des equations suivantes?

 

0. y - (1 - x) (x - 2)
1. y - (l - x) (2 - x)
2. y - (1 - x) (2 - x)2
3. y - (1 - x)2(x - 2)
a. y - (1 - x)2(2 - x)

o.

178

51. Un précepteur doit arranger pour qu'un certain nombre de garcons
occupe un dortoir. S'il y a deux gar ons dans chaque chambre, alors
11 y aura deux gargons sans chambre. S'il y a trois-garcons dans
chaque chambre, alors 11 y aura deux chambres qui restent vides.
Gombiens de chambres y a-t-il dans la dortoir?

0. 6 l. 8 2. 10 3. 12 4. 14

5%. Lequel des suivants represente 1e graphique de y - x4 -
x

 

53. 501: logb 2 - 1/3, alors 10gb 32 est egal L

0. 2 1. 5 2. -3/5 3. 5/3 a. -———1—-—
logz 32

54. Soient x et y des nombres réels, pour quels x peut-on dafinir y
par

Y 7
9 - x2

Tout x sauf x - 3

Tout x sauf x - 3 et x - -3
x < -3 ou x > 3

-3 < x < 3

x < 3

l-‘UNHO

2

55. A quel point est-ce que y - 3x - x a un minimum

local?
0. (2,4) l. (3,0) 2. (1,2) 3. (0,3) 4. (0,0)

56. De combien de manieres peut-on ranger sur une etagere de
bibliotheque, 5 gros livres, 4 livres moyens et 3 livres minces,
sachant que les livres de meme grosseur restent voisins?

5! 4! 3! 3! - 103680
5! 4! 3! - 17280
(5! 4! 3!) x 3 - 51840

5 § 4 x 3 x 3 - 180

21 x 3 -12288

L‘th-‘O

179

57. Soitéun space affine de dimension 3 et (0, (13,16) un repere
orthonorme de . Soit f 1'app1ication deg dansé qui a tout point M
de coordonnées (x,y,z) associe 1e point M' de coordonnézs qr}, y',
z') définies par:

x' - z + 1

y' - x + 1

z' - y + 1
Laquelle des propositions suivantes est vraie?

f est une translation

. f est une symétrie orthogonale par rapport
f est une rotation

f est un vissage ‘
f est une symetrie orthogonale par rapport a une droite

DI

un plan

#UNHO

58. Une fonction f est paire et dErivable en 0. Quelle propriete'
doit satisfaire f'(0)?

0. f'(0) - l
l. f'(0) > O
2. f'(0) < 0
3. f'(0) - 0
4. f'(0) peut prendre n'importe quelle valeur.

59. Si log N - n, alors log N2 est égal h

0. n + 2
1. n2

2. n/2
3. 2n

4. n - 2

60. Sur une droite, deux points A at B sont donnes. La coordonnee
de A est -3 et la coordonnée de B est +7. Quelle est la coordonnee
du point G, 31 B est le mi-point du segment AC?

#UNHO
4.
M

+17 M

61.

N-‘r‘

     

canfiméfre:

        
     

States; at:

 

i 2 a b 9 3 ’ ' ’ ” " u."

Dans la figure, on indique combien de centimetres de pluie sont
tombées pendant 13 semaines. La moyenne par semaine est environ

0. 2 cm 1. 4 cm 2. 6 cm 3. 8 cm 4. 10 cm

180

62. Si sy - 1 at x est superieur A 0, lequel des suivants est vrai?

0.

l.
2.
3
4

Quand x est superieur

1, y est negatif.

\
a

Quand x est superieur a l, y est superieur a 1.
a

Quand x est inférieur

1, y est inférieur a l.

Quand x augmente, y augmente.
Quand x augmente, y diminue.

53. 51 2x2 - 12x + 9 - 2 (x - a)2 + b, alors

§UNHO

64. La courbe
minimum relatif a (2,3). Les valeurs de a et de b sont

«L‘le-‘O

- -3 at b - -9
- 3 at b - -9
- -3 et b - 9
- -3 et b - 27
- 3 et b - 27

definie par y - x3 - ax + b contient un point

- 8 1/2 et b - 12
- 12 et b - 8 1/2
- 12 at b - l9
- 19 at b - l2
- 19 et b - 32

FIN

181

INTERVIEW SCHEDULE (Principals)

The 15 question groups listed were used to frame the interviews with
the principals. They represent general topics which were then often
discussed in considerably greater detail. They are all given first
in French, followed by an English translation. The questions were
not necessarily worded exactly as given here. Often there was a
short conversational lead-in to soften the impact of the question,
particularly if the question could be seen as sensitive.

l. Est-ce que vous-mame vous vous occupez des visites de classe?
Combien de visites faites-vous chaque période? Est-ce que c'est 1e
méme pour les anciens que pour les nouveaux? Y a-t-il des visites
officielles et non-officielles?

2. Comment est-ce que vous arrangez 1a question des admissions? Y
a-t-il beaucoup de candidats? Quel est 1e pourcentage requis pour
etre admis? Qui decide les cas difficiles? Est-ce que chaque
professeur a droit d'admettre deux enfants?

3. Est-ce que l'école a des objectifs écrits quelque part? (Si oui)
Comment est-ce que ces objectifs étaient élaborés? (Si non)
Qu'est-ce que vous’voyez comme 1e but principal ou bien les buts
principaux de cet ecole?

4. Combien de fois est-ce qu'on fait des reunions pédagogiques?
Quelle sorte de chose est-ce qu'on y discute?

5. Quand les professeurs sont absents ou bien en retard, quelles
sortes d' action est-ce que vous prenez? Y a-t-il des réglements
pour cela? Est-ce que c'est un probléme ou bien ce n'est pas un
probléme?

6. Quel est l'effectif actuel de l'ecole? Combien d'élEves y a-t-il
dans chaque classe?

7. Donnez-vous cours? Quelle matiere? Combien d'heures? Est-ce
I
que vous preferez donner ou ne pas donner cours?

8. Quel texte de mathématiques est-ce qu'on emploie en Zeme? en
6éme?

9. Pour la question de l'achat du materiel pédagogique, comment
est-ce qu'on decide les priorités? Si vous aviez suffisamment
d'argent, qu'est-ce que vous acheteriez? Est-ce que les professeurs
vous indiquent ce dont ils ont besoin?

A
10. Vous etes prefet depuis combien d'annees? Quelle est votre
formation? Est-ce que vous avez pris des cours avancés en
mathématiques?

182

11. Comment est-ce que vous partagez votre horaire? Combien souvent
ates-vous obliges d'etre absent de l'école? Pour chercher les
salaires, cela prend combien de jours chaque mois? En general,
qu'est-ce qui vous occupe 1e plus? La discipline? Les visites en
classe? Les comités? Le travail de bureau? Les parents?

12. Est-ce que vous introduisez de la methodologie aux professeurs?

l3. Est-ce que vous examinez regulierement les plans de lecons des
professeurs?

l4. Est-ce que vous faites une preparation particuliere pour
l'exétat? (organisation des eleves, recherche des items, etc.?)

15. Pour vous, quel est le r61e le plus important du préfet?
Pourquoi?

INTERVIEW SCHEDULE (Principals)-Translation

(As noted above, only the general topics of the interview are
listed. The questions as they appear here were reworded by the
researcher and broached in a generally more cautious manner than
would appear from the rather blunt form in which they appear here.)

1. Do you do the classroom supervision of teachers yourself or does
someone help you with this? How many visits are made each quarter?
Is this the same for new teachers and experienced teachers? Are
there both official and unofficial visits?

2. How is the question of student admissions handled? Are there
many candidates? What percentage is required for admission? Who
decides the difficult cases? Does each teacher have the right to
admit two students?

3. Does the school have written objectives somewhere? (If yes) How
were these objectives elaborated? (If no) What do you see as the
principal goal or the principal goals of this school?

4. How often are faculty meetings held? What sorts of topics are
discussed?

5. When teachers are absent or late, what sort of action is taken?
Are there rules covering this? Is this generally a problem or is it
generally not a problem?

6. How many students attend this school? What is the breakdown by
class?

7. Do you teach? What subject? How many hours? Given the choice,
would you prefer to teach or not to teach?

8. What mathematics textbook is used in Grade 8? In Grade 12?

183

9. As far as purchasing instructional materials is concerned, how
are priorities decided? If you had the money, what would you buy?
Do the teachers tell you what they need most?

10. How long have you been principal here? What educational
background do you have? Have you taken advanced courses in
mathematics?

11. How is your daily program broken down? How often are you obliged
to be absent from the school? How many days are required each month
to take care of picking up the salaries? In general, what keeps you
busiest? Discipline? Classroom visits? Committees? Office work?
Parents?

12. Do you introduce new teaching methods to the teachers?
13. Do you regularly examine the teachers' lesson plans?

14. Do you make any special preparation for the state exams? (by
organizing the students, tracking down multiple choice questions used
in previous years, etc.?)

15. For you, what is the most important role of the principal? Why?
INTERVIEW SCHEDULE (Teachers)

(The comments made regarding the nature of the questions for the
principals also apply to the questions for the teachers.) In
addition to the questions listed here, teachers indicated for each
mathematics question whether or not it had been covered in class
during the current school year, during a previous school year, or not
at all. As with the principals' interview schedule, the English
translation follows.

1. Depuis quand donnez-vous cours ici? Ailleurs? Quelle est votre
formation?

2. Combien de fois est-ce que le prefet (ou bien la proviseur) a
visité votre classe cette année? Comment est-ce que le préfet vous
informe de ce qu'il voit pendant ces visites?

3. Avez-vous change de méthodologie suite 5 ces visites?

4. Combien souvent y a-t-il des reunions pedagogiques? Quel est 1e
but de ces reunions? Quelle sorte de choses est-ce qu'on traite?

5. Etes-vous membre de comités pédagogiques?

6. Est-ce que le prefet s'occupe beaucoup de l'enseignement ou
est-ce qu'il s'occupe plutot d'autres affaires?

184

INTERVIEW SCHEDULE (Teachers) -Translation

1. How long have you taught at this school? Elsewhere? What is
your educational background?

2. How often has the principal (or an individual given this
responsibility) visited your classroom this year for purposes of
supervision? How does the principal inform you about what he has
observed?

3. Have you changed your teaching in any way because of the
principal's visits?

4. How often are faculty meetings held? What is the nature of these
meetings?

5. Are you a member of any curriculum committees, etc.?

6. Does the principal get very involved in the teaching process or
does he tend to be occupied mainly with other sorts of things?

APPENDIX B

Site access correspondence

185

LETTRE DE PERMISSION

Je sous-signs, , préfet des etudes de

l'Institut , donne a Robert Prouty 1a
permission de faire remplir ses questionnaires aux eleves pour
lesquels je suis responsable. J'ai recu une copie des questionnaires
et je comprends qu'il s'agit de 25 questions sur l'estime de soi et
de 34 a 40 questions a propos des mathématiques. Je comprends que
ces questionnaires demandent environ une heure a remplir. J'ai
arrange ceci avec le titulaire des classes en question et j'accepte
toute responsabilite in loco parentis.

Fait a
le 1986

 

 

Préfet des études

186

LETTRE DE CONSENTEMENT

Cher Inspecteur:

Je suis un candidat au doctorat en Administration de l'Education
a Michi an State University aux Etats- Unis. Comme partie intégrale
de mes etudes, je suis en train de faire un recherche- terrain dans
plusieurs ecoles secondaires de la Zone de Lubero. Le but de cette
recherche est de mieux connaitre les activites des prefets des etudes
et d'etudier la relation entre ces activités et les resultats obtenus
chez les eleves.

Si vous acceptez de participer dans cette recherche, je vous prie
de signer 1e formulaire en bas. Veuillez comprendre que vous n'avez
aucune obligation et que vous etes libres a quitter la recherche
n 'importe quand sans aucune difficulte. Si vous acceptez de
participer, sachez que tous les resultats seront gardes dans la plus
stricte confidence et que 1' anonymité de chacun est garanti. Si
cependant, vous voulez une copie des resultats avec ces restrictions,
je serai tres content de vous les pourvoir.

Cet interview prendra entre une et deux heures. Je vous remercie
d'avance pour votre comprehension.

\
Sincerement,

R. Prouty
Ancien prefet des etudes
Institut de Lukanga

Je sous- signe, , atteste
par la presente que cette recherche m' a ete expliquee, que je 1a
comprends, et que je consens librement a y participer.

Fait a
le 1986

 

 

Inspecteur

187

LETTRE DE CONSENTEMENT

Cher Préfet:

Je suis un candidat au doctorat en Administration de l'Education
a Michigan State University aux Etats-Unis. Comme partie integrale
de mes études, je suis en train de faire un recherche-terrain dans
plusieurs ecoles secondaires de la Zone de Lubero. Le but de cette
recherche est de mieux connaitre les activites des prefets des etudes
et d'etudier la relation entre ces activites et les résultats obtenus
chez les éléves.

Si vous acceptez de participer dans cette recherche, je vous prie
de signer le formulaire en bas. Veuillez comprendre que vous n'avez
aucune obligation et que vous étes libre d quitter la recherche
n'importe quand sans aucune difficulte. Si vous acceptez de
participer, sachez que tous les resultats seront gardés dans la plus
stricte confidence et que l'anonymité de chacun est garanti. Si
cependant, vous voulez une copie des résultats avec ces restrictions,
je serai tres content de vous les pourvoir.

Cet interview prendra environ une heure de temps. Je vous
remercie d'avance pour votre comprehension.

\
Sincerement,

R. Prouty
Ancien prefet des etudes
Institut de Lukanga

Je’sous-signé, , atteste par la
presente que cette recherche m'a été expliquee, que je 1a comprends,
et que je consens librement a y participer.

Fait a le 1986

 

(Signature)

188
LETTRE DE CONSENTEMENT

Cher Professeur:

\ Je suis un candidat au doctorat en Administration de l'Education
a Michigan State University aux Etats-Unis. Comme partie integrale
de mes etudes, je suis en train de faire un recherche-terrain dans
plusieurs ecoles secondaires de la Zone de Lubero. Le but de cette
recherche est de mieux connaitre les activites des prefets des etudes
et d'etudier la relation entre ces activites et les resultats obtenus
chez les eleves.

Si vous acceptez de participer dans cette recherche, je vous prie
de signer le formulaire en bas. Veuillez comprendre que vous n'avez
aucune obligation et que vous etes libres é quitter la recherche
n'importe quand sans aucune difficulte. Si vous acceptez de
participer, sachez que tous les resultats seront gardes dans la plus
stricte confidence et que l'anonymité de chacun est garanti. Si
cependant, vous voulez une copie des résultats avec ces restrictions,
je serai trés content de vous les pourvoir.

Ce questionnaire vous demandera environ quinze minutes. Je vous
remercie d'avance pour votre comprehension.

Sincérement,

R. Prouty I
Ancien prefet des etudes
Institut de Lukanga

I
Je'sous-signe, , atteste par la
presente que cette recherche m'a été'expliquée, que je la comprends,
\
et que je consens librement a y participer.

Fait : 1e 1986

 

(Signature)

189

A QUI DE DROIT

Je sous-signe' KAKURU “WASH“: , Commissaire du Zone de

Lubero, atteste par la presente que Robert Prouty, candidat au
doctorat a l'Universite de l'Etat de Michigan at ancien prefet
des etudes de l'Institut de Lukanga, a man accord oour faire sa
recherche-terrain dans le territoire du Zone de Lubero.

 

Je prie aux autoritég tant civiles que militaires de lui
' aoporter l'aide dont il aura besoin.

Fait a Lubero 1e 26/Février/1986

 

 

’K/Axuau MUHASHANI HUBUNGA

190

A QUI DE DROIT

Je sous-signe, Sfllié '- REG’EQ , atteste par la presente que

Robert Prouty, candidat au doctorat a l'Université de I'Etat de
Michigan et ancien préfet des études de l'Institut de Lukanga,
a man accord pour faire sa recherche-terrain dans des eboles
secondaires du Zone de Lubero.

Je prie aux préfets des études de lui accorder 1'aide dont 11
aura besoin.

Fait a Butembo 1e
‘ (26 [041986

 

 

191

 

APPENDIX C

Letters of Permission

University Committee on Research

Involving Human Subjects

Second International Mathematics Study

Consulting Psychologists Press

192

MICHIGAN STATE UNIVERSITY

 

UNIVERSITY comma-z 0N RESEARCH INVOLVING I-:Asr LANSING . MICHIGAN . mums
IquAN suajrcrs (nouns)

m ADMINIsnIAnON BUlLDING

Innsnaus May 13, 1986

Mr. Bob Prouty
620 Lexington Avenue
East Lansing, Michigan 48823

Dear Mr. Prouty:

Subject: Proposal Entitled, "Principals' Instructional
Management Emphasis and Two Outcomes of Secondary
Schooling in Zaire"

I am pleased to advise that I concur with your evaluation that this
project is exempt from full UCRIHS review, and approval is herewith
granted for conduct of this project.

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 prior to May 13, 1987.

Any changes in procedures involving human subjects must be reviewed by the
UCRIHS prior to initiation of the change. UCRIHS must also be notified
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,
Leia/c,
Henry E. Bredeck
Chairman, UCRIHS
HEB/jms

cc: Dr. Samuel A. Moore II

193

Will 1‘ Second International Mathematics Study

<{g{$§3}. Deuxiéme Etude lntemationale en Mathématiques

N3, ,7 A 341 Amory. umw at Illinois. Champion. Illinois 61820. u.s.A.
Mm: (217) 33187“! ‘-

May 12. 1986

Mr. Bob Prouty
620 Lexington Avenue
East Lansing, MI 48823

Dear Mr. Prouty:
I am glad that the materials were helpful to you.

With respect to the French instruments, I suggest that you request
copies from the Ontario Centre, since they tested a bilingual population.
My understanding is that they, in turn, obtained the instruments from
France (or French-speaking Belgium).

I should note, however, that some questions have been raised about
the adequacy of the French translations. You might ask, as well, if
they have any comments along those lines. Feel free to get back if
you run into problems.

Contact person: Dr. D. Raphael
Educational Evaluation Centre
252 Bloor Street West
Toronto, Ontario M55 1V6
CANADA -

I see no reason why you should not go ahead and do your testing in
Zaire--and you may regard this letter as permission to do so, as far as
I am concerned. But I would much appreciate receiving a report of your
findings.

Sincerely,

.‘ ' .--.‘O

\L'Q.Y\ K/

Kenneth J. Travers
Professor of Mathematics Education
KJT:del

194

CONSULTING PSYCHOLOGISTS PRESS, INC.
577 College Ave. (P.O. Box 60070), Palo Alto, California 94306 (415) 857-1444

 

 

Dear Robert: S. Prouty:
We appreciate your interest in our test. wraith Self-Esteem Inventogz

 

 

by Stanle Coopersmith and are responding to your letter ol
1 Jamar? l 6 in which you request permission to translate the test
into FIench for your use in research. .

 

We would be willing to authorize you to make this translation provided you
agree to thmlollowing conditions: '

1. You will lniorm us of the subject ct your research and how many copies oi
the translation you expect to use. It you lntend to make more than 200
cc ies. lease noti us re ardin lurthe rrnis ion an a ossible tee.

Ap fee eat .08c pefir’ copygis av r the 5809 copfes used. 0

2. You will send us three copies of your translation and all rights to the trans -

lation will be assigned by you to Consulting Psychologists Press, Inc.

3. The translation will be used for your own research project only. and you will
not sell or give away any copies for others to use. Upon completion of the
project you agree to destroy the cepies printed except for the tow you need
for your records. '

4. Each copy of your translation will have on it a credit line to indicate. in
English. the test name. author. copyright date. publisher. translators name.
and the date at the translation.

it you agree to these conditions and intend to proceed with the translation. sign
the enclosed copy oi this tom and return it to the Permissions Department at
Consulting Psychologists Press. I: you decide not ioproceed. return this letter
to the Permissions Department and indicate that you are not proceeding with
the translation.

Sincerely ' Agreed to by Era—vi x9 - grout"
9%?" (gZMD Date: (nameAIpru, 33" l‘N’é

‘S‘dli'a Perntz U
Permissions Editor

Date: 21. Janaury 1989

 

   
 

pervising proiessor

   

(name-3i supervrsing rolessor)

Datezjlq 4.1 {1

APPENDIX D

Explanation of principal ratings

195

Ratings for principals represented a composite score based on
eight components of instructional leadership. These are: 1) number
of classroom visits, 2) existence of explicit instructional
objectives, 3) number of meetings with teachers for instructional
purposes, 4) principal's knowledge of mathematics curriculum,
5) principal's emphasis on instructional materials when stating
purchasing priorities, 6) principal's emphasis on instruction when
stating perceived role of principal, 7) principal's stated emphasis
on creating an orderly, yet open learning environment and 8) the
teaching load of the principal. In each of these categories,

principals were assigned a rating of l, 2 or 3 (low, average or

high).

In the category of classroom visits, schools where principals had
visited the two mathematics teachers a combined total of two times or
less during the course of the year were given a low rating.
Principals who had visited a combined total of between three and
seven times were ranked average and principals who had visited more
than seven times were ranked high. Where there were vice-principals
whose duty it was to visit classes, these visits were considered in
the same way as visits by the principals. However, visits by
vice-principals did not generally appear as consistently associated

with higher achievement as did visits by the principals themselves.

In the category of teachers' meetings, principals who had held

fewer than three were rated low; principals holding between three and

196
five were rated average and the two principals who had held eight and

fifteen respectively were rated high.

None of the principals rated high on instructional objectives,
since none of the schools had formulated explicit written objectives
in instructional terms. Principals who indicated that teachers had
met to discuss general instructional objectives were rated average
and those who indicated no discussion of instructional objectives

were rated low.

Principals were asked to name the mathematics books used in their
Grade 8 and Grade 12 classes and to briefly describe the curriculum.
Those who could not name the books were rated low in the category of
curriculum knowledge. Those who could name the books but who either
openly stated they knew nothing about the curriculum or who showed no
knowledge of the curriculum were rated average. Those who appeared

to know something about the curriculum were rated high.

Two questions were collapsed into one to rank principals on the
basis of the priority which they assigned to the purchase of
instructional materials. Principals were asked to list their
purchasing priorities and whether or not teachers were given input
into purchasing. Principals who gave teachers input received at
least a rank of average in this category. Principals who gave
priority to the purchase of textbooks, maps, and other instructional
materials were rated either average or high (depending on whether or

not teachers were given input). Principals who gave priority to

197
other sorts of purchases or who showed no particular interest in

instructional materials were ranked average or low.

Principals were asked to state what sorts of activities occupied
them the most, what reasons motivated most of their absences from
school, and what they saw the role of the principal to be. The
emphasis on instruction in each of these answers provided the

categorization of high, average, or low.

Principals were ranked according to the observed level of
discipline at the school, i.e., whether or not there appeared to
exist an orderly climate in which students could learn without an

excessive number of disruptions.

Finally, principals were ranked according to their own teaching
load, i.e., the number of hours personally spent in the classroom
each week. Principals teaching more than four hours were ranked
high, principals teaching from one to four hours were ranked average
and those who do not teach were ranked low. Principals' teaching
hours were as follows:

0 hours--Schools l,3,4,9,13,15
h hours--Schools 2,5,7,14

6 hours--School 6

7 hours--Schools 12

8 hours--School ll
10 hours--School 8
12 hours-~School 10

APPENDIX E

Mathematics achievement and self-esteem scores

198

Table A-1
Grade 8 Mathematics and Self-Esteem Scores

School N Mathach St.Dev- St.Err. N Self-Esteem St.DevList.Err.

 

 

l 40 44.808 9.158 1.448 40 50.184 9.516 1.504
2 51 41.629 10.426 1.460 55 52.032 9.920 1.336
3 42 49.817 14.539 2.243 42 61.272 9.044 1.396
4 40 34.519 9.521 1.505 40 56.556 12.444 1.968
5 42 41.575 14.093 2.175 42 58.380 12.252 1.892
6 43 61.091 9.759 1.488 43 53.676 12.504 1.908
7 41 45.122 10.534 1.645 41 54.892 10.932 1.708
8 67 37.773 11.459 1.400 67 56.048 12.400 1.516
9 30 43.462 17.086 3.119 31 62.956 12.432 2.232
10 64 46.635 15.423 1.928 66 52.268 10.188 1.256
11a 28 41.484 12.836 2.426 28 57.440 10.312 1.948
11b 33 52.797 11.347 1.975 33 58.796 9.132 1.588
12 41 38.743 9.476 1.480 41 57.444 9.392 1.468
13 27 48.575 16.006 3.080 27 62.816 12.832 2.468
14 38 47.874 17.217 2.793 38 61.768 7.728 1.252
15 50 39.154 12.618 1.784 50 60.448 9.664 1.368
Table A-2
Grade 12 Mathematics and Self-Esteem Scores

Sghool fl Mathach St.Dev. St.Err. N Self-Egtgem St.Dev. St,ErrI
1 18 28.333 9.093 2.143 18 57.800 12.980 3.060
2 40 36.667 10.487 1.658 40 62.200 11.844 1.872
3a 22 32.727 9.739 2.076 22 64.364 12.276 2.616
*3b 18 38.333 8.024 1.891 18 66.667 8.784 2.072
4 30 32.667 10.947 1.999 31 53.020 14.008 2.516
*5 9 28.519 12.705 4.235 9 60.444 11.568 3.856
6 11 27.273 7.575 2.284 11 64.000 10.584 3.192
7 15 31.333 6.146 1.587 17 62.932 11.744 2.848
8 29 30.115 9.063 1.683 29 64.000 8.000 1.484
*9 12 42.222 9.779 2.823 12 63.268 14.452 4.172
10 15 45.333 9.984 2.578 15 66.836 8.460 2.184
*11 10 40.000 6.479 2.049 10 45.816 10.864 3.436
12a 22 33.788 8.184 1.745 21 71.436 12.004 2.620
*12b 7 62.857 12.387 4.682 7 66.448 7.360 2.780
13 34 27.451 8.802 1.510 35 60.980 9.724 1.644
14a 13 30.513 6.213 1.723 13 61.296 8.028 2.228
14b 2 40.000 14.142 10.000 2 58.000 14.144 10.000
15 15 34.667 9.241 2.386 19 59.788 7.240 1.660

APPENDIX F

Parents' level of education

School MOMED

y..-
OOCDVO‘U‘DWNH

11a
11b
12
13
14
15

0‘03

‘OQNO‘UIvPWU-DNH

F'P‘ F‘P‘F‘F‘P‘ ham) P‘P‘F‘F‘H‘P‘F‘ P‘hJF*FJ

F‘BJF‘

.700
.400
.333
.350
.571
.698
.550
.439
.323
.556
.429
.121
.195
.556
.128
.320

.444
.100
.545
.944
.258
.333
.700
.118
.857
.250
.000
.300
.864
.143
.200
.692
.500
.105

199

Table A-3
Parents' Level of Education—-Grade 8

DADED
.949
.255
.286
.500
.643
.372
.244
.343
.710
.919
.714
.091
.077
.593
.205
.140

NWHNNHNNNNI—‘l—‘NWNN

Table A-4
Parents' Level of Education--Grade 12

.611
.800
.500
.056
.226
.556
.636
.529
.655
.917
.400
.700
.955
.571
.371
.750
.000
.158

NHHNHHNUNl—‘N

nahahahihihi

APPENDIX C

Academic Intake Rankings

200

ACADEMIC ADMISSIONS SCALE

The purpose of this scale is to determine the selectiveness of
schools in admitting students. It is designed to provide a further
measure of differences in academic intake at the schools and is
calculated on the basis of five elements: 1) the percentage with
which the student must have passed at the school from which s/he is
coming; 2) the percentage which the admitting school requires on its
own admissions exams; 3) the number of students applying for
admission; 4) whether or not admissions are open to all grades or
only to Grades 7 and 9; and 5) whether or not the principal has broad
discretionary power in admitting students (if s/he does, the rating
on item 5 is 0). Tighter admissions procedures on any of these items
will give the school a score of 1 for that item (or up to 1.5 if very
stringent criteria are applied by the school). Average or below
average admissions procedures for an item give a score of 0 or .5 for
that item. Where incomplete data were available, scores were

averaged.

Schools with an overall score of 0 are ranked low, 1 is medium
low, 2 is medium, 2.5 is medium high and 3 or above is high. The
exact cutoff points for these classifications is admittedly somewhat
arbitrary. Statistics based on this variable use an order ranking
which has the effect of creating values of 1 through 5 for the five

categories. Scores are given below.

201

S 01 S_cale Total. 8512122
.1. 2. .3. .‘1 2

1. O 1.5 0 1 O 2.5 medium high

2. 1 0 0 0 1 2 medium

3. 1 0 l - - 3.3 high

4. 0 0 0 0 0 0 low

5. 0 0 0 0 0 0 low

6. 0 0 l 0 - 1.25 medium low

7. 0 1.5 0 0 0 1.5 medium

8. 0 0 0 1 O 1 medium low

9. 0 O 0 0 .5 .5 low

10. 0 .5 0 - 1 1.9 medium

11. 0 0 0 - 1 1 medium low

12. O 0 0 0 0 0 low

13. 0 0 - 0 0 0 low

14. O - - 1 1 2 medium

15. 0 0 - 0 0 0 low

APPENDIX H

Pearson Product-Moment Correlation Coefficient Matrices

202

The following variables are compared in this appendix:

MATHACH
SESTEEM
PRINCIP
MOMED
DADED
ADMISS
INSTAKE
TEACHX
TEACHQ
STAB
AGE
XAMPALL
TEXT
VIS

SEX

Mathematics achievement

Self-esteem

Principals' ratings for emphasis on instructional leadership
Mothers' level of education

Fathers' level of education

Admissions selection criteria

Academic intake ranked by inspectors

Years of teacher experience

Teacher qualifications

Years student has been at same school (Student tenure)
Student age

Trends in pass rate over four years

Possession of a textbook (0 - YES, 1 - NO)

Number of classroom visits for purposes of supervision

Gender (0 - FEMALE, 1 - MALE)

PRINCIP

SESTEEM

MOMED

DADED

ADMISS

INSTAKE

STAB

AGE

TEXT

TEACHQ

TEACHX

VIS

SEX

MATHACH

.2071
(292)
P-.001

.0140
(320)
P-.402

-.0549
(320)
P-.164

-.0285
(321)
P-.305

.0303
(322)
P-.294

-.0183
(322)
P-.372

.0892
(319)
P-.056

-.1656
(322)
P-.001

-.1472
(303)
P-.005

.2479
(322)
P-.001

.1908
(290)
P-.001

.1942
(322)
P-.001

-.0184
(322)
P-.37l

SESTEEM

.0902

(298)
P-.06O

.0553
(327)

P-.160

.0951
(328)

P-.043

.0906
(329)

P-.051

.1413
(329)

P-.005

.0673
(326)

P-.1l3

.0656
(329)

P-.118

-.0385

(309)

P-.250

.0382
(329)

P-.245

.0996
(298)

P-.043

.0970
(329)

P-.039

-.0690

(329)

P-.106

GRADE 12

203

XAMPALL

.0577
(301)
P-.159

-.0681
(329)
P-.109

.2313
(330)
P-.001

.1416
(331)
P-.005

.6208
(332)
P-.001

.4113
(332)
P-.001

-.1645
(329)
P-.001

-.1735
(332)
P-.001

-.2232
(312)
P-.001

.5451
(332)
P-.001

.2170
(300)
P-.001

.0086
(332)
P-.438

-.3614
(332)
P-.001

PRINCIP

SESTEEM

MOMED

DADED

ADMISS

INSTAKE

STAB

AGE

TEXT

TEACHQ

TEACHX

VIS

SEX

MATHACH

.1631
(637)
P-.001

.0474
(676)
P-.109

-.0757
(672)
P-.O25

-.0310
(670)
P-.211

.1631
(677)
P-.001

-.0173
(677)
P-.327

.0517
(675)
P-.O90

-.0926
(674)
P-.008

.0076
(616)
P-.426

-.1629
(677)
P-.001

-.0413
(677)
P-.142

.2037
(677)
P-.001

.0164
(677)
P-.335

GRADE EIGHT

SESTEEM

-.0525
(644)
P-.092

-.0121
(679)
P-.377

.0168
(677)
P-.33l

-.1150
(684)
P-.001

-.0755
(684)
P-.024

-.0550
(682)
P-.076

-.0327
(681)
P-.197

-.0385
(623)
P-.169

-.1013
(684)
P-.004

.0430
(684)
P-.l31

-.l499
(684)
P-.001

-.0087
(684)
P-.410

APPENDIX I

State examination results, 1983-1986

205

Table A-5
State Examination Pass Rates 1983-1986

School Section W83 P83 %83 W84 P84 %84 W85 P85,%85 W86 P86 886

1 Teach. 37 4 11 29 15 52 22 22 100 18 12 67
(Math/Ph. 35 12 34 31 27 87 23 20 87 29 27 93)

2 Teach. 35 33 94 28 27 96 37 31 84 45 45 100
3 Bio/Ch. 31 4 13 24 23 96 20 18 90 18 17 94
Teach. 24 18 75 30 14 47 40 39 98 22 22 100

4 (Math/Ph. 38 22 58 38 22 58 48 44 92 35 34 97)
Bus. 27 12 44 33 6 18 22 8 36 - — -

5 Math/Ph. - - - 6 3 50 15 5 33 9 1 11
6 Social 9 7 78 8 5 63 13 13 100 14 5 36
7 Teach. 28 25 89 29 29 100 18 11 61 19 19 100
8 Teach. 18 2 11 27 22 81 21 18 86 29 27 93
9 Bio/Ch. 21 9 43 11 8 73 10 6 60 14 11 79
10 Bus. 28 11 39 27 15 56 27 27 100 15 12 80
ll Bio/Ch. 21 7 33 17 6 35 14 3 21 11 11 100
12 Teach. 46 16 35 47 43 91 18 9 50 28 11 39
Math/Ph. 13 6 46 13 6 46 8 3 38 7 5 71

13 Teach. 27 l 4 37 17 46 39 12 31 42 21 50
14 Lit. 28 19 68 20 18 90 19 15 79 - - -
Bus. 39 29 74 36 8 22 41 41 100 - - -

15 Teach. 27 9 33 12 8 67 20 9 45 20 9 45
Total 532 246 46 503 319 63 475 354 75 375 289 77

Below are the ranks of each school for each of the past four years.
Based on these ranks, trends over the past four years are placed into
one of three categories and assigned a score which is used for
calculating examination pass rate trends (XAMPALL).

Table A-6
Annual State Examination Ranks

Sghogl 83 84 85 86 Status (Rank score)

1 12 6 5 7 good or improving (3)

2 l 2 7 1 good or improving (3)

3 7 7 3 4 good or improving (3)

4 5 l4 8 5 good or improving (3)

5 - 11 13 14 poort or deteriorating (1)
6 3 9 2 13 no clear trend (2)

7 2 1 9 2 good or improving (3)

8 13 4 6 6 good or improving (3)

9 6 5 10 9 no clear trend (2)

10 8 10 l 8 no clear trend (2)

11 11 15 15 3 no clear trend (2)

12 9 3 ll 11 poor or deteriorating (l)
13 14 13 14 10 poor or deteriorating (1)
l4 4 12 4 - no clear trend (2)

H
U‘
...:
O
co
H
N
H
N

poor or deteriorating (1)

APPENDIX J

Rules governing teacher and student behavior

at two schools in the sample

206

RULES FOR TEACHERS

1. The behavior of the teacher must serve as an example for the
students. The teacher will not permit himself to do what he forbids
his students to do.

2. It is strictly forbidden for teachers to ask students to pay a
tip in order to receive a favor.

3. It is forbidden for teachers to engage in scandalous conduct
(drunkenness, smoking tobacco, immorality, etc.).

4. The school does not authorize any private meetings between
teachers and students, even for a missed quiz. These must be taken
care of at the school and not at the teacher's home.

5. It is strictly forbidden to allow visitors into the classroom,
except for cases of absolute necessity.

6. The respect for and the swift execution of directives from the
principal is recommended for all teachers at this school.

7. No teacher is authorized to enter the classroom and begin
teaching without informing the administration of his presence. Upon
arriving and leaving, each teacher must sign the register which is
located in the principal's office.

8. It is forbidden for a teacher to send a student on an errand
during class hours. The student is at the school to study, and not
to play the role of an errand boy.

9. Clandestine meetings of a revolutionary character are forbidden
and may lead to an investigation.

10. All requests must be made in writing, with courtesy, on good
writing paper.

11. Corporal punishment is forbidden.

12. Teachers do not have the right to exclude students or to
confiscate anything belonging to a student without the approval of
the administration.

13. It is forbidden to punish a student at one's home for a fault
committed at school.

14. Punishments must be proportional to the fault committed.

15. Teachers must keep their lesson plans with them and present them
when requested to do so by an authority.

SLIGHT

1.

207
STUDENT INFRACTIONS: CLASSIFICATION
(The list given here for students does not come from

the same school as do the above rules for teachers.)

tardiness, lack of cleanliness, wearing of a hat or

non-prescription glasses

SERIOUS

O
OWNO‘U‘IPUNHE @VO‘U’lpUJNH

talking in class, at flag raising, or while entering class
failure to wear school uniform

failure to bring tools for intellectual or manual labor
extravagant dress, beard, straightened hair, makeup
smoking

speaking in a language other than French or English

hand games

lack of respect for peers

repeated offenses in the above categories

refusal to accept a punishment

cheating

neglect of manual labor, sports and cultural activities
fighting

insolence toward a member of the staff

unexcused absence

plundering

failure to keep up school documents

VERY GRAVE

l.
2.

O‘U‘I-pw

repeated GRAVE infractions

lack of respect, insolence, arrogance toward school authority
or third person

false documents

theft at the school or at the exterior

revolt or attempted revolt and any subversive activity
presence at places of ill repute (bars, brothels), drunkenness,
libertinage

pornographic behavior (writings, conversation, reading, etc.)

PUNISHMENTS

1.

E9; slight infractions

-individual reprimand in front of class or parents
-physical or intellectual labor to perform
-exclusion from one to three hours of class

-a grade of 'fair‘ on the report card

Fo se us nf actions
-reprimand in front of the entire student body
-no more than one day of physical labor
-no more than a one-day suspension, with parents being
called to the school
-a grade of 'mediocre' on the report card

208

3. For grave infractions

-no more than three days of manual labor

-three-day suspension with a severe warning in the presence
of parents

-a grade of 'bad' on the report card

4. For very grave infractions
-definitive expulsion after meeting of the Discipline

Committee

APPENDIX K

Instructional supervision form

209

ppflrCSSI-Un: eeeeeeeeeeeoeeeoeeooeeeoo
I'll-HE DE pnprnATIDN r:'eeoeoooeoeooe

hetiercs : ......1.......Classe : ...........DATE:..........Heurc:......
1. s-Ilulciot d9 1” leg-On: 0.0.0.0000...0................OODOO......OOOOOOOOO

2. FUt d' In 10:3": ................OCOOOOOOOOOOOOOOOO0......00.0.0.0...

3. Réréfnncc:eoeoooooooo00000000000.00000000000000000.000000000000000coo
‘. “83621131 didactiquc ( a .‘ZBDDOILCI Ch ClCSSC):eooeeoeoeoeoeoeooeeeeeee

s. rntiéerp1s“10:0":00.00.00...0.00....0............OOOOOOOOOOOOOOOO

nl

PPEAMBULE (liaison avec la legon precedente):..........l...........

.-.-‘

B) pn’lfi’q nriHCioP-Ux:eeeeeeeeeeeeeeeeooeeeeeee00000000000000.0000...

..... ......OCOOOOOOCOOOOOOO0.0...0..........OOOOOOOO..0.0.0.000.......

-4.- ..---

.c) svnthaSE: ......OOOOOOO.......-0..........OOOOOOOO0.00.00.00.00...

6° Travail impose aux eleveSC Sujet d'Intcrrogntion: ...................

' c—o-
. -..- - -- -- .
... e—v- "'
.. ..-

7' Critique do la lcgonlfivcntucllcl': 2.77?:;...........................

 

:C'O'CQQOOOOOOOOOCOOOO...‘C.COOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO000......

.-----
. o¢u—¢—---'- I-‘
.‘

o - I
-----------.---------—-- - - - -----—-----

BIBLIOGRAPHY

BIBLIOGRAPHY

Apumbi-Lomema (1984). Les langues dans l'enseignement au Zaire.

Educareur, 8, 33-45.

Armour, D. (1972). Equality of educational opportunity: The basic
findings reconsidered. In Mosteller, F. & Moynihan, D. (Eds.) Qg_

ggggliry gr edggariongl opportugiry, New York: Random House.

Averch, H. et a1. (1972). How effective is schooling? A critical

rgvigw gng gynrhggis oi research iindings, Santa Monica, CA:
The Rand Corporation.

Bachman, J. & O'Malley, P. (1984). Black-white differences in
self-esteem: Are they affected by response styles? American

Journal of Sociology, 28(3), 624-639.

Bedeian, A. et a1. (1977). Test-retest reliability and internal
consistency of the short form of Coopersmith's Self-Esteem

Inventory. Psychgiogical Reports, 81, 1041-1042.

Behrman, J. & Birdsall, N. (1983). The quality of schooling:

Quantity alone is misleading. The American Ecggomig Review,
18(5), 928-946.

Berman, P. & McLaughlin, M. (1978). Federa ro rams su ortin
ggugationai change, 8. Santa Monica, CA: Rand Corporation.

Bidwell, C. (1965). The school as a formal organization. In J.
March (Ed.), and 00 o o nizat ons. Chicago: Rand McNally,
972-1022.

Bidwell, C. (1977). The school as a formal organization: Some new
thoughts. In G. Immegart & W. Boyd, Eds., Problem-finding in
educational administration: Trends in research and theory.
Lexington, MA: Heath.

Bloom, B. (1980). The new direction in educational research:
Alterable variables. Phi Qelta Kappan, 8;, 382-385.

Bossert, S. et a1. (1982). The instructional management role of the
principal. Educgtional Administratiog Quarterly, i8, 34-64.

Bowles, S. & Levin, H. (1968). The determinants of scholastic
achievement: An appraisal of some recent evidence. Journal of

figmgn Resogrggs, 8, 3-24.

Brookover, W. & Lezotte, L. (1977). Changes in school

ghgrggreristics coinciggnr wirn changgs in studegt achigvement.
East Lansing: Michigan State University College of Urban
Development.

Brookover, W. et al. (1979). §chool sociai systems and student
achievement; Schools cag make g difference. New York: Praeger.

210

211

Buschmann, M. (1984). The use of research knowledge in teacher

education and teaching. merican Jour a of Educatio , 28(4),
421-439.

Cain, G. & Watts, H. (1970). Problems in making policy inferences

from the Coleman Report. American Sociological Review, 88,
228-242.

California State Department of Education, Office of Program
Evaluation and Research. (1977). School effectiveness study;
Ine iirst year. Sacramento, CA: California State Department of
Education.

Chang, L. & Ruzicka, J. (1985). Second international mathematics
study (United States Technical Report 1). Champaign, IL:
Stipes.

Chapman, D. & Windham, D. (1985). Academic program "failures" and
the vocational school "fallacy": Policy issues in secondary

education in Somalia. International Journal of Educational
Development, 8(4), 269-282.

Cohen, M. (1981). Effective schools: What the research says.
Ioday's Education, April/May, 48G.

Cohen, M. (1983). Instructional, management, and social conditions
in effective schools. In Odden, A. & Webb, L. (Eds.) §cnooi
ina e d SC 00 m ovement' Linka es 0 he 980's.
Cambridge, MA: Ballinger.

Cohen, M. et a1. (1972). A garbage can model of organizational
choice. Administrativg science Quarterly, i1, 1-25.

Colclough, C. (1980). e o c ve ment' A
review pf the eviggnpg. (World Bank Staff Working Paper No. 399)

Washington, DC: World Bank.

Coleman, J. et a1. (1966). Equality of educational opportunity.
Washington, DC: U.S. Office of Education.

Coleman, J. et a1. (1982). High school achievement. New York:
Basic.

Cooksey, B. (1981). Social class and academic performance: A

Cameroon case study. gonparative Education Review, 28, 403-418.

Coopersmith, S. (1967). The pntecedents pf self-esteem. Palo Alto,
CA: Consulting Psychologists Press.

Corwin, R. (1974). Models of educational organizations. In F.
Kerlinger & J. Carroll, eds. ev ew o Resear in ucation
V91, 2. Itasca, IL: Peacock.

212

Crook, et a1. (1984). The linkage of work achievement to
self-esteem, career maturity and college achievement. Journgl of

chationnl Behavior, 28(1), 70-79.

Cuban, L. (1983). Effective schools: A friendly but cautionary
note. Ehl Delta Kappgn, 83(10), 695-696.

Cuban, L. (1984). Transforming the frog into a prince: Effective
schools research policy and practice at the district level.

flnryprd Educnrionni Review, 88(2), 129-151.

Currie, J. (1977). The occupational attainment process in Uganda:
Effects of family background and academic achievement on
occupational status among Ugandan secondary school graduates.
Comparativg Education Review, 2;, 14-28.

Currie, J. (1979). Family background, academic achievement and
occupational status in Uganda. In Heyneman, S. & Currie, J.
(Eds.) choolin cademic erformance nd occu ationa
nrrninment in a non-industriglized society. Washington, DC:
University Press of America, 80-93.

Dede, C. & Freiberg, H. (1986). The long-term evolution of
effective schools. Ihg_Egnnnripnnl_£prnn, 81(1), 65-79.

Diaz, J. (1984). A cross-cultural study of the reliability of the

CooPersmith self-esteem inventory. Educational and Psychological
neasnrempnr, 35(3), 575-581.

Dove, L. (1980). The role of the community school in rural
transformation in developing countries. Comparative Education,
18(1), 67-79.

Edmonds, R. (1978). d u 510 f e ter ur nd sues
rglprgd ro gffgptivg sphopling. A paper prepared for the

National Conference on Urban Education. St. Louis, Missouri,
July 10-14.

Eisemon, T. (1985). o - co omi onse uence f chool ex ansion
in Kenyg. Unpublished manuscript. McGill University, Montreal.

Elmore, R. (1978). Organizational models of social program
implementation. Enbiic Policy, 88(2), 185-228.

Elmore, R. (1983). Social policymaking as strategic intervention.

In E. Seidman (Ed.) flandbook of sppial intervention. Beverly
Hills, CA: Sage, 212-236.

Elmore, R. (1984). Forward and backward mapping: Reversible logic
in the analysis of public policy. In Hans, K. & Toonen, T.

(Eds.), Eclipy inpigmentntion in federni and nnitary systems.
Boston: Nijhoff, 33-70.

213

Erickson, F. (1984). School literacy, reasoning, and civility: An
anthropologist's perspective. Review Educa ona Research,

55(4), 525-546.

Farnsworth, C. (1986, November 17). Washington watch: Plea by
Zaire on debt seen. Ihfi New York Iingg, p.22.

Firestone, W. & Herriott, R. (1982). Prescriptions for effective
elementary schools don't fit secondary schools. Educational

Lgndership, 38, 51-53.

Firestone, W. & Wilson, B. (1985). Using burearcratic and cultural
linkages to improve instruction: The principal's contribution.

Educational Administration Quarterly, 28(2), 7-30.

Foster, P. (1965). The vocational school fallacy in development
planning. In C. Anderson & M. Bowman (Eds.) Education and

Eponomic Qeveiopmenr. London: Frank Cass, 142-166.

Gamoran, A. & Dreeben, R. (1985, August). goupling and control in
gducational organizatipns. A paper prepared for the annual

meeting of the American Sociological Association.

Glasser, W. (1969). S o withou a re. New York: Harper &
Row.

Good, T. & Brophy, J. (1985). School effects. In Wittrock, M.

 

(Ed.) Handbook of researpn pn reaching, 8r8 gdirion. New York:
Macmillan.

Goodlad, J. (1983). ca 1e choo ' as t he
inrnrg. New York: McGraw-Hill.

Gould, D. & Amara-Reyes, J. (1983). ef ec c u ti on

s v e ' u t t s o evelo 1

countries (Staff Working Paper No. 580.) Washington, DC: The
World Bank.

Gran, G. (Ed.) (1979). a e' The 011 ica economy of
nnderdevelopment. New York: Praeger.

Griffiths, D. (1958). min ativg theory in gdugntion.
Chicago: University of Chicago.

Hallinger, P. & Murphy, J. (1985a). Instructionni effeptiveness and

- c u 0 or t s ood o
rhg gander? A paper presented at the annual meeting of the
American Educational Research Association, Chicago.

Hansford, B. & Hattie, J. (1982). The relationship between self and

achievement/performance measures. Revigw pf Educatipnni
8mm. 52(1). 123-142.

214

Hanson, J. (1971). nhanc n the contribution of formal educgtion

in Africa. Primnry sgnopis, secondary schoois 8nd teacher
training institutiong. Washington, DC: American Council on

Education.

Hanson, M. (1975). The modern bureaucracy and the process of

change. Educationai Administration Quarterly, ii, 21-36.

Hanson, M. (1976). Beyond the bureaucratic model: A study of power
and autonomy in educational decision-making. Interchange, 1,
27-38.

Hanushek, E. & Kain, J. (1972). On the value of equality of
educational opportunity as a guide to public policy. In

Mosteller, F. & Moynihan, D. (Eds.) On equality of educational
ppportunity. New York: Random House.

Heyneman, S. (1976). Influences on academic achievement: A
comparison of results from Uganda and more industrialized

societies. §ocioiogy of Education, 88, 200-211.

Heyneman, S. (1980). Differences between developed and developing
countries: Comments on Simmons & Alexander's 'Determinants of

school achievement'. Economic Development and Cultural Change,
28(2), 403-408.

Heyneman, S. (1983). Improving the quality of education in
developing countries. Einance and Deveiopmenr, 28(1), 18-21.

Heyneman, S. (1984). Research on education in the developing
countries. nte at one Journa o Educ io a ev o ment.

Heyneman, S. & Loxley, W. (1982). Influences on academic achievement
across high and low income countries: A re-analysis of IEA

data. Socioipgy of Educarion, 88, 13-21.

Heyneman, S. & Loxley, W. (1983). The impact of primary school
quality on academic achievement across 29 high- and low-income
countries. erica ournal o Sociolo , 88, 1162-1194.

Heyneman, S. et a1. (1984). Textbooks in the Philippines:
Evaluation of the pedagogical impact of a nationwide investment.

Ennggrionnl Evnlnnripn an8 Policy Analysis, 8(2), 139-150.
High, R. & Achilles, C. (1986). Principal influence in
truct onal cho . A paper presented at the

annual meeting of the American Educational Research Association,
San Francisco.

 

Huff, S. et a1. (1982). i ci a ff enceS' E cellence in school
Lendersnip nng nanaggmenr. Boston: McBer.

Hurst, P. (1981). Some issues in improving the quality of
education. 80mparntivg Egucation, 11, 185-193.

215

Irvine, D. (1979). Factors associated with school effectiveness.

Edupptipnal Tegnnplogy, 22(5), 53-55.

Jencks, C. et a1. (1972). lneguality; A reassessmenr of rhe effect
pf inmily gun pchpoling in Anerica, New York: Basic.

Johnson, B. et a1. (1983). The Coopersmith self-esteem inventory: A

construct validation study. Educarional nnd P§ycnological
Beeeerement. 43. 907-913.

Kagia, R. (1986). The impact of educational research on policy:
Trends and ppssipilirigs in Africa. A paper presented at the

annual meeting of the Comparative and International Educational
Society, Toronto.

 

Kaplan, 1. (Ed.) (1979). Zaire; A cpuntry study. Washington, DC:

Government Printing Office.

Klitgaard, R. & Hall, G. (1974). Are there unusually effective
schools? Journal of Human Resources, 18, 90-106.

Kokenes, B. (1978). A factor analytic study of the Coopersmith
Self-Esteem Inventory. Agplpgngnpg, 18, 149-155.

Lanzas, A. & Kingston, P. (1981). English achievement in Zaire: The
effects of family status and residential disruption. Conparative

Edeeetien.£eziee. 25. 431-441.
Lezotte, L. (1986). Ephool effectiveness; Reflectiong and future

e on . A paper presented at the annual meeting of the
American Educational Research Association, San Francisco.

Lezotte, L. et a1. (1980). 00 ear n imate d tudent
aphigygngnr. Institute for Research on Teaching. East Lansing,
MI: Michigan State University.

Madden, J. et a1. (1976). ect ess st ° tat
Enlifprnin. A paper presented at the annual meeting of the
American Educational Research Association, San Francisco.

Maloba, N. (1977). Predicting schools' performance in the state
examinations: A model for Shaba province in Zaire. Dissertation

Abeereete_lntereetiene1. 38. 5383A.

Manasse, A. (1985). Improving conditions for principal
effectiveness. Elementary School Journal, 88(3), 439-463.

March, J. & Olsen, J. (1979). Anniguity and choice in organizations,
28 gdition. Bergen, Norway: Universitesforlaget.

Marsh, H. et a1. (1983). Self-concept: Reliability, stability,
dimensionality, validity, and the measurement of change. Journal

e£_Eeeeatiene1_£e1ehelesx. 12(5). 772-790.

216

Martin, W. & Willower, D. (1981). The managerial behavior of high

school principals. Egnpnripnnl Adminigrrnrion Quarterly, l1,
69-90.

McAndrew, G. (1981). The high-school principal: Man in the
middle. Daedalus, lTQ, 105-118.

McLaughlin, M. & Marsh, D. (1978). School management and
organization in successful schools. Sopiology of Education, 8T,
211-226.

Meunier, R. & Ketelslegers, M. (1985). Definition des objectifs
educationnels. Educnteur, 8, 22-32.

Meyer, J. & Rowan, B. (1978). The structure of educational
organization. In M. Meyer (Ed.) Environments and
Qrganizatipns. San Francisco: Jossey-Bass, 78-109.

Meyer, J. & Rowan, B. (1983). The structure of educational
organizations. In J. Meyer and W. Scott (Eds.) Qrganizational
EnvironmentS' R t a t a t . London: Sage.

Morse, W. (1964). Self Concept Data. NASSP Eulletin 48 23-27.

’ —,

Mukweso, M. et a1. (1984). Education and occupational attainment
from generation to generation: The case of Zaire. Comparative
Education Review, 28(1), 52-68.

Mwaniki, M. (1973). The relationship between self-concept and
academic achievement in Kenyan pupils. D sertat o bstra ts

International, 88, 1138A.

New York State Department of Education. (1974). 8cnopl factorg
v e ' ud o tw inne - i
pchopls. Albany, NY: Office of Education Performance Review.

Nimnicht, G. et a1. (1969). w u e choo . New York:
General Learning.

Parsons, T. (1947). Introduction. In M. Weber, T88 Theory of

c a a Economi a . New York: Free, 3-86.
Powell, A. et a1. (1985). 3 ma hi h choo ' Winners
nnn losers in the educgtional markerplape. Boston: Houghton
Mifflin.
Psacharopoulos, G. (1973). et r duc o ' An ter ational
ppnpnrinpn. Amsterdam: Elsevier.

Psacharopoulos, G. (1986). To vocationalizg or not to
vopationalizez The; is rhe curriculnn question. A paper

presented at the annual meeting of the American Educational
Research Association, San Francisco.

217

Psacharopoulos, G. & Woodhall, M. (1985). Education for
Envelopment; Analysis of investment choices. New York: Oxford

University Press.

 

Purkey, S. & Smith, M. (1982). Too soon to cheer? Synthesis of
research on effective schools. Educational Leadership, 82,
64-69.

Purkey, S. & Smith, M. (1983). Effective schools: A review.
Elsmsntary §chool Journsl, 88(4), 427-452.

Purkey, S. & Smith, M. (1985). School reform: The district policy
implications of the effective schools literature. Elementary
Eshool Journal, 88(3), 353-389.

Ralph, J. & Fennessey, J. (1983). Science or reform: Some questions
about the effective schools model. Phi Delta Kappan, 88,
689-694.

Raudenbush, S. & Bryk, A. (1986). A hierarchical model for studying
school effects. Sociology of Education, 82, 1-17.

Rio, A. (1979). Defensiveness, self criticism and self concept in a
sample of Black, Mexican, and White American adolescents
(Doctoral dissertation, Michigan State University, 1979).

Disssrtation Abstracts international,

Rowan, B. (1982). Instructional management in historical
perspective: Evidence on differentiation in school districts.

Enngsrionsl Administrsripn Qnarterly, 18(1), 43-59.

Rowan, B. & Denk, C. (1984). Management succession, school
socioeconomic context, and basic skills achievement. American

Educational Research Journal, 2l(3), 517-537.

Rowan, B. et al. (1982). hethodological considerations inistudies of

sffectivs principals. A paper presented at the annual meeting of
the American Educational Research Association, New York.

Rowan, B. et a1. (1983). Research on effective schools: A

cautionary note. WW. 12(4). 24-31.

Saunders, M. & Vulliamy, G. (1983). The implementation of
curricular reform: Tanzania and Papua New Guinea. Comparative

Edupsrion Eeview, 21(3), 351-373.

Sedlak, M. et a1. (1986). Selling students:short: Classroom
a n cademi e o e e ca hi h scho . New
York: Teachers' College Press.

Simon, H. (1971). Decision making and organizational design. In D.

Pugh (Ed.) Orgsnigsrion Theory. New York: Penguin.

218

Spady, W. (1982). Outcome-based instructional management: A
sociological perspective. Aus ralia ournal ucat o , 28
10-29.

Squires, D. (1980). Qharncteristics of effective schools; Ths
importance of school processes. Washington, DC: National

Institute of Education. (ERIC Document Reproduction Service No.
ED 197 486)

Steers, R. (1975). Problems in the measurement of organizational
effectiveness. Administrativs Sciencs:Qusrterly, 28, 546-558.

Thompson, J. (1967). Organisations in Action. New York:
McGraw-Hill.

Tyler, W. (1985). The organizational structure of the school.

Annual Eeview of Sociology, ll, 49-73.

Udy, S. (1959). Bureaucracy and rationality in Weber's organization

theory: An empirical study. erica Sociolo a eview, 28,
791-795.

Urevbu, A. (1984). Vocational education in Nigeria: A preliminary
appraisal. e a 0 al u n du at ona eve o ment,
8(3), 223-229.

Vansina, J. (1982). Mwasi's trials. Daedalus, ill, 49-70.

Varon, B. (1979). Znire; Current economic situation snd
ponstraints. Washington, DC: The World Bank.

Weber, G. (1971). n - it h dre e au ht to ea ' Four
sucpessful schools. Washington, DC: Council for Basic
Education.

Weber, M. (1947). The essentials of bureaucratic organization: An
ideal-type construction. In R. Merton, et al. (Eds.), Readsr in
hnrsnupracy. Glencoe, Il: Free Press, 18-27.

Weick, K. (1976). Educational organizations as loosely coupled

 

systems. Administrative Scisnce Quarterly, 2i, 1-19.
Wellisch, G. et a1. (1978). School management and organization in
successful schools. Eociplogy pf Educaripn, 8T, 211-226.
Wholeben, B. (1985). eci o imulat o modelin or educational
pgninisrrnripn. Unpublished manuscript, University of Texas at

El Paso, Department of Educational Leadership and Counseling, El
Paso, Texas.

Williams, P. (1984). Africsn educaripn un8er siege. Paper

presented at the Conference on Education Priorities in
Sub-Saharan Africa, Windsor, England.

219

 

Windham, D. (1985). te a ef cie the Afr can 3 hool.
Unpublished manuscript, Universite de Dijon, Dijon, France.
Wise, A. (1979). Lsgislated lsarning; Ths bureaucratization of the

e an ass . Berkeley, CA: University of California
Press.
Wylie, R. (1979). e - on e ° 0 and resea ch
selected topics, Vol, 2. Lincoln, NB: University of Nebraska
Press.

Youdi, R. (1972). An exploratory study of achievement and attitudes
of high school students in the Congo: An aspect of socialization
for national development. b ac a iona .
Ann Arbor, MI: University Microfilms.

Young, R. (1986). ' nal of e o in
th Louis ublic schools--findin s rom the e lication

implementation fielg test, June, 1984, A paper presented at the
annual meeting of the American Educational Research Association,
San Francisco.

Zirkel, P. & Gable, R. (1977). Reliability and validity of various
measures of self-concept among ethnically different adolescents.

WWW. 19(4). 48-54.

GENERAL REFERENCES

Alexander, K. et a1. (1981). Measure for measure: On the use of
endogenous ability data in school-process research. Anerican

fiopiplpgicnl Rsview, 88, 619-631.
Assie-Lumumba, N. (1983). §ocial inequality and access go schooling
d Wo d' r'ca c . Revised version of a paper

presented at the annual conference of the Comparative and
International Education Society, Atlanta. (ERIC Document
Reproduction Service No. ED 236 788)

Austin, G. (1979). Exemplary schools and the search for
effectiveness. Educational Leadership, 81(1), 10-14.

Blank, R. et a1. (1985). District and schoo oli ies and mana ement
ract es eadin o e fec ve secondar schOOIS' eli inar

drafr for non intrusive sites. A paper presented at the annual

meeting of the American Educational Research Association,

Chicago.
Blase, J. (1986). mensi n effe ve sch aders ' e
rsnchsr's psrspestive. A paper presented at the annual meeting

of the American Educational Research Association, San Francisco.

Bolman, L. & Deal, T. (1984). Modern approaphes to nnderstanding
sng managing prganizatipns. San Francisco: Jossey-Bass.

Bor, W. van den (1985). Problematic aspects of expatriate
educational research with and in developing countries.
nternational Journa f Educational evelo ment 8(3), 167-181.

Boyer, E. (1983). schoo ' re ort o secondar educatio in
Anerica. New York: Harper & Row.

Bridges, E. (1982). Research on the school administrator: The state

of the art, 1967-1980. Edupational Administration Quarrsrly,
18(3), 12-33.

Clift, R. & Waxman, H. (1986). Some neglected elements of effective
schools research: A review of literature. Journal of Classroom

lnteractipn, 28(2), 2-11.

Cohn, E. & Rossmiller, R. (1985). Res a ch ef e ve s h0018°
1 ca n or ess-d ve 0 ed countries. Washington, DC:
World Bank.

Coleman, J. (1984). Professorial training and institution building
in the Third World: Two Rockefeller Foundation experiences.

Domparative Education Review, 28(2), 180-202.
Coleman, J. & Ndolamb, N. (1983). Zaire: The state and the

university. In Thomas, R. (Ed.) Politics and education.
Oxford: Pergamon Press.

220

221

Coopersmith, S. (1981). Self-esteem inventories. Palo Alto, CA:
Consulting Psychologists Press.

Crahay, M. (forthcoming). Search for school effects in developing
countries: 1966-1986. In Crahay, M. (Ed.), L'Art s; 1a §cience

8e l'Enseignement; Eonnage s Gilberr de Landsheere. Paris:
Nathan.

Crandall, D. et a1. (1983). Psople, polipies and practices;

Enamining the chain of school improvement. Andover, MA: The
NETWORK.

Cusick, P. (1983). The sgalirarian ideal and the American high
sphool. New York: Longman.

De Bevoise, W. (1984). Synthesis of research on the principal as
instructional leader. Educatipnsl Leadership, 81(5), 14-20.

Donmoyer, R. (1985). Cognitive anthropology and research on

effective principals. Educational Administration Quarterly, 21,
31-57.

Dwyer, D. et a1. (1983). F ve r ci ls in sction: Perspectives on

insrrngripnsl_nnnsgsnsnr. San Francisco: Far West Laboratory

for Educational Research and Development.

Edmonds, R. (1979). Effective schools for the urban poor.
Edeeetienel_Leedershie. 31. 15-24.

Elmore, R. & McLaughlin, W. (1982). Strategic choice in federal
education policy: The compliance - assistance trade-off. In

Lieberman, A. & McLaughlin, M. (Eds.), Pplipynaking in
sfingsripn. Chicago: University of Chicago Press, 159-194.

Etzioni, A. (1964). Mpgsrn_8rggniznripns. Englewood Cliffs, NJ:
Prentice-Hall.

Fallon, B. (1979). Principals--the instructional leaders--hit or
myth? NA§§P Eulletin, 88, (423).

Farrar, E. et al. (1983). fect v hools ro rams i h

ho 3' m catio s o oli r ice and esea ch.
Cambridge, MA: Huron Institute.

Farrell, J. (1977). The IEA studies: Factors that affect

achievement in six subjects in twenty-one countries. ea e s'
Qellese_3eeere. 12. 289-297.
Fuller, B. (1985). 001 uali n develo in countrie3°
at ves ts b ea nin (Report No. EDT7.) Washington,

DC: The World Bank.

222

Furtwengler, W. (1986). Penching success through involvement:

lnplementaripn srrsregy fipr greeting and maintaining effective
sghnpls. A paper presented at the annual meeting of the American
Educational Research Association, San Francisco.
Gauthier, W. (1982). e e e ves n i truc iona
ec school ' A mode nd cess. A paper presented at

the annual meeting of the American Educational Research
Association, New York.

Gray, J. (1981). School effectiveness research: Key issues.
Edncsripnal Eesenrph, 28(1), 49-54.

Greenfield, W. (1982). ca es a ci a ‘ e
srste-Qf-the-arr. A paper presented at the annual meeting of the

American Educational Research Association, New York.

Hallinger, P. & Murphy, J. (1985b). Assessing the instructional

management behavior of principals. e e a l,
88,(2), 217-247.

Hanson, E. (1981). Organizational control in educational systems:
A case study of governance in schools. In Bacharach, S. (Ed.)
a o v c o n d t . New
York: Praeger.

Harbison, R. et a1. (forthcoming). an veme
incterel_Breeil1_JReeelte_ef_inieiel_eeelxeie. Washington. DC:
World Bank.

Hawkridge, D. et a1. (1982). In-service teacher education in
Kenya. In Perraton, H. (Ed.) a v o t 0 £0
sgncarion. Baltimore: Johns Hopkins University Press.

Heyneman, S. (1978). Tsnrhpohs snn schisvsment; Ehst do ws know?

 

Washington, DC: World Bank.

Heyneman, S. & Jamison, D. (1980). Student learning in Uganda:
Textbook availability and other factors. 88mparativs Education
Review, 28(2), 206-220.

Heyneman, S. & White, D. (Eds.) (1986). The quality of education and

on dev . Washington, DC: The World Bank.
Johnson, H. (1978). r c als ake a difference e
e o s e a - a v a u

pn;sgmss_in_1§E_sgh881s. Washington, DC: National Institute of
Education. (ERIC Document Reproduction Service No. ED 168 181)

Kabuya, L. (1980). Assessment of reading abilities of Zairean

students at ages 12 and 18 years. Disssrtation Absrracrs
Internetienel. 41. 2529A.

223

Reyes, E. (1975). The problem of wastage in Zairian elementary

schools. WWW. 3a. 6614.
Kokenes, B. (1974). Grade level differences in factors of
self-esteem. stslppmsnrsl Psychplpgy, 18, 954-958.

Kotter, J. (1982). What effective general managers really do.
finryarg Dusiness Revisw, 88(6), 156-167.

Leithwood, K. & Montgomery, D. (1982). The role of the elementary

school principal in program improvement. Review of Educational
Researph, 82, 309-339.

Lezotte, L. & Bancroft, B. (1985a). School improvement based on
effective schools research: A promising approach for
economically disadvantaged and minority students. Jpnrnsl of

Negro Educarion, Summer Yearbook.

Lezotte, L. & Bancroft, B. (1985b). Growing use of the effective
schools model for school improvement. Educarional Leadership,

82(6), 23-27.

Licata, J. et a1. (1978). The school and environmental robustness:
An initial inquiry. Jonrnsl pf Erpsrimental Education, 81,
28-34.

Linn, R. (1983). eas r sch o ef ct vene53° ow evement

t a a no used. A paper presented at the annual
meeting of the American Educational Research Association,
Montreal.

Lipham, J. (1981). Eiisprive principal, sffective school. Reston,
VA: American Association of Secondary School Principals.

Madaus, G. et a1. (1979). The sensitivity of measures of school

effectiveness. flarysrg Egusntional Review, 82, 207-230.
Madaus, G. et al. (1980). cho effect veneSS' A reassessment of

the evi8ence. New York: McGraw-Hill.

Mandeville, G. (1986). e valuation of a "count down" procsdurs_
de e ect v SC 00 . A paper presented at the
annual meeting of the American Educational Research Association,
San Francisco.

Mandeville, G. & Anderson, L. (1986). ud o the tab 0

o ff c ve e meas e a ade a sub e reas. A

paper presented at the annual meeting of the National Council on
Measurement in Education, San Francisco.

Masiala, M. (1984). La problematique de la langue nationale dans
l'enseignement. Ennnnrsnr, 8, 25-32.

224

Matthews, K. & Brown, C. (1976). The principal's influence on
student achievement. NASSP Bulletin, 88(402), 1-15.

Maxwell, R. (1968). Leader behavior of principals: A study of ten
inner-city elementary schools of Flint, Michigan. Dissertation
Abstracts. 2.8. 2950A.

McDonnel, L. (1985). Implementing low-cost school improvement
strategies. ement ch our a , 88, 424,438.

Meyer, et a1. (1978). Institutional dissensus and institutional
consequences in schools. In M. Meyer (Ed.) Environments snd
Qrganizatipns. San Francisco: Jossey-Bass.

Miskel, C. (1977). Principals' perceived effectiveness, innovation
effort, and the school situation. Educational Administration
Quarterly, 18, 31-46.

Murnane, R. (1981). Interpreting the evidence on school

effectiveness. Tsachers' Dpllege Record, 88(1), 19-35.

Nagel, T. a Scholar, M. (1986). A on itudinal ud s em tic
0 ts t a se sta dar d achievement tes score
actor ho e f ct veness esearc . A paper presented at

the annual meeting of the American Educational Research
Association, San Francisco.

Niles, F. (1981). Social class and academic achievement: A Third

World reinterpretation. 88mpararivs Education Review, 28,
419-430.

Odden, A. & Odden, E. (1984). Education reform, school improvement,
and state policy. Educsrional Leadership, 82(2), 13-19.

O'Malley, P. & Bachman, J. (1983). Self-esteem: Change and
stability between ages 13 and 23. stelopmenral Psychplogy, 12,

257-268.
Peters, T. & Waterman, R. (1982). s arch o e ce e ce° ssons
ro e i ' - u om e . Cambridge, MA: Harper & Row.

Phi Delta Kappa (1980). flhy d8 some urban schools succeed? Study of

srceptional urhan slementary sghools. Bloomington, IN: Phi
Delta Kappa.

Pitner, N. & Hocevar, D. (1986). An empirical investigation of the
l i-d me ona at i ci a1 leader . A paper

presented at the annual meeting of the American Educational
Research Association, San Francisco.

Porter, J. & Washington, R. (1979). Black identity and
self-esteem: A review of studies of black self-concept,

1968-1978. Annual Revisw pf Eopiology, 8, 53-74.

225

Psacharopoulos, G. (1983). Education as an investment. In Habte, A.

et al. (Eds.) Educarion and development: Views from the World
Dank. Washington, DC: World Bank, 11-14.

Rampaul, W. et al. (1984). The relationship between academic
achievement, self-concept, creativity, and teacher expectations
among native children in a northern Manitoba school. Alberts

ourna f Educa iona esearc 8(3), 213- 225.

Reidel, J. (1980). Self-esteem, achievement scores and IQ scores
among students of three ethnic groups in grades seven and eight.

Disssrtatipn Abstrsgps international, 82, 996A.

Rosenholtz, S. (1985). Effective schools: Interpreting the
evidence. Anerican Journal of Edusation, 93, 352-388.

Rutter, M. et al. (1979). Pifteen thousand hours; §econdary
sshools and thsir sffsprs 8n philgrsn. Cambridge, MA: Harvard

University Press.

Salmi, J. (1984). Educational research on the Third World or with
the Third World: A view from the South. IDS Dullerin, 18(4),
9-11)

Sheline, Y. (1979). Education and national development: The
effects of school sponsorship and other factors on academic
achievement in Zaire. e bstract e na . Ann
Arbor, MI: University Microfilms.

Sheline, Y. et a1. (1984). The effect of school sponsorship on
academic achievement: A comparison of Catholic, Protestant and

government secondary schools in Zaire. Comparative Education,
28(2), 223-236.

Simmons, J. 8 Alexander, L. (1978). The determinants of school
achievement in developing countries: A review of research.

Economip stelopmenr nnd intural Change, 28, 341-358.

Rutter, M. (1983). School effects on pupil progress: Research
findings and policy implications. In Shulman, L. 8 Sykes, G.
(Eds.) andbook o e chin nd olic . New York: Longman.

Saidi, R. (1979). An emerging problem for educational planners in
Zaire: Unemployment among the educated; causes and tentative

solution. Dissertatipn Abstracrs International, 82, 4605A.

Sergiovanni, T. 8 Corbally, J. (Eds. ) (1984). Leadership and

prganiga arional pnlrure; New perspectives on ndministrative
rhso ory sn n8 practise. Urbana, IL. University of Chicago Press.

Shoemaker, J. 8 Fraser, H. (1981). What principals can do: Some
implications from studies of effective schooling. Phi Delta
Rnppnn, November, 178-182.

226

Simmons, R. et a1. (1978-79). Self-esteem and achievement of black

and white adolescents. Sppial Prohlems, 28, 88-96.

Sirotnik, K. 8 Burstein, L. (1983). Methodological issues in

tud n he fec v n schoolin ' Recen develo ments and
lingering concerns. Los Angeles: University of California,
unpublished.

Sizer, T. (1984). Rorace's compromise; The dilemma of the American_
high school. Boston, MA: Houghton Mifflin.

 

Somerset, T. (1984, December). A note on the effects of head
reachers and of superyisers on school Quality. A paper presented

at the Conference on Education Priorities in Sub-Saharan Africa,
Windsor, England.

Sparks, G. et al. (1985). School improvement through staff
development. Educational Leadership, 82(6), 59-61.

Stewart, J. (1978). Bureaucratic structure, interpersonal climate
and student achievement. Dissertation Abstracts International,
82, 6457A-6458A.

Weick, K. (1982). Administering education in loosely coupled
schools. Phi Delrn Rsppan, 88, 673-676.

Young, B. (1980). Principals can be promoters of teaching
effectiveness. Thrusr for Educnripnal Leagership, 2, 11-12.