DEMOGRAPHIC AND ENVIRONMENTAL FACTORS
AFFECTING FERTILITY DECISIONS IN SWAZILAND

Dissertation for the Degree of Ph. D.
MICHIGAN STATE UNIVERSITY
HELENA EVA WALLENDER
1977

LIBRARY

$233944.

Michigan State
University

 

This is to certify that the

thesis entitled

DEMDGRAPHIC AND ENVIRONMENTAL FACTORS AFFECTING
FERTILITY DECISIONS IN SWAZILAND

presented by

Helena B. wallender

has been accepted towards fulfillment
of the requirements for

Ph.D. Family Ecology

degree in

MWVZK

Major professor

 

 

Date June 30, 1977

 

0-7639

 

ABSTRACT

DEMOGRAPHIC AND ENVIRONMENTAL FACTORS AFFECTING
FERTILITY DECISIONS IN SWAZILAND

By

Helena Eva Wallender

Since rapid population growth became a concern, valuable
time has been lost in too much emphasis in trying to understand
global population trends, while research on the family level has
been neglected. This study attempted to explore factors affecting
fertility decisions in Swaziland, involving both men and women.

The specific purpose of the study was to analyze the relationships
between demographic variables, such as sex, age, number of chil-
dren, and environments to aspects of fertility decisions. The
study focused on such aspects as fertility goals and standards,
attitudes about fertility control, levels of information about
reproduction, hygiene, child care, nutrition, farm practices, use
of community resources and income.

A home management framework was employed in the study. The
reported desired family size was regarded as fertility goals and
the ideal number of children as fertility standards. Measuring
attitudes about fertility control, being of importance in directing
fertility choices, was of special interest. For the measurement of

attitudes a Guttman-type scale was developed havinga reproducibility

Helena Eva Wallender

coefficient of 0.85. In order to measure levels of information,
scores were computed from answers on short fixed alternative ques-
tions. Use of community resources was measured by a ratio of
"utilized" over "available" community resources. An already avail-
able scale was used in measuring farm practices. Three areas of
diversified environmental characteristics were selected for the
implementation of the study. The areas were Velezizweni, Luyengo
and Bhunya. Veleziweni is a rural area with a predominant subsis-
tence economy and many features of traditional Swazi living.
Luyengo can be defined as a semi-rural area of mixed economic and
social systems, and Bhunya as an urbanized area, created in connec-
tion with the establishment of a pulp mill. The total population
within determined boundaries in Velezizweni and Luyengo was investi-
gated, while a random sample was drawn from the total population of
Bhunya. Seventy-three men and ninety-six women were interviewed.
Infbrmation about the use of community resources, farm practices
and income was only collected from men, while information about
hygiene, child care and nutrition was only collected from women.
Sixty-one percent of the sample was matched couples. Couples' com-
posite resourcefulness score and mean attitudes about fertility
control were tested against number of children and environment.
The data were collected in July 1976 by students from the College of
Agriculture at the University of Botswana and Swaziland.

Least squares analysis was used in order to test relation-
ships. Significant results at a .0l alpha level were obtained for

the effect of the environment on a number of variables. 0f greatest

Helena Eva Wallender

magnitude were, in order, population attitude scores, income, farm
practice score and hygiene score. Sex was found to have a signifi-
cant effect on attitudes about fertility control and levels of
reproductive information. Women were more positive about the con-
trol of fertility and had a higher level of reproductive information
than men. Age and number of children were found to have a non-
significant effect on all dependent variables in the study.

The major conclusions drawn from findings in the study were
that traditional fertility values tend to be prevalent in less
advanced stages of social and economic environment. Men have more
traditional fertility values than women and are less motivated to
control fertility. The level of reproductive information tends to
decrease in the transformation to a modern society. Age was not a
determining factor on attitudes about fertility control, fertility
goals, standards nor levels of information. Since the actual number
of children did not influence fertility goals, standards and atti-
tudes about fertility control, it was concluded that fertility
behavior in the three investigated communities was not goal-oriented.

A major implication for home economics program development
in contributing to fertility decline is the necessity of involvement
in reshaping the family environment. Programs should direct them-
selves both to men and women. Training of decision-making skills
should be implemented in a number of home economics subject areas.
Families should be assisted with bringing fertility decisions to a
level of awareness, resulting in rational and conscious decisions

about family size. General policy implications are continuous

Helena Eva Hallender

emphasis on rural development, establishment of rural industries
and decentralization of political decision making.

An in-depth study over a period of time, selecting a few
families from each area, was recommended for gaining a full eco-

logical perspective on fertility decisions.

DEMOGRAPHIC AND ENVIRONMENTAL FACTORS AFFECTING
FERTILITY DECISIONS IN SWAZILAND

By

Helena Eva Wallender

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Family Ecology

1977

© Copyright by
HELENA EVA WALLENDER

I977

ACKNOWLEDGMENTS

It was a feeling of inadequacy as a home economics extension
worker in Africa faced with the overwhelming problems of social and
economic development that brought me to graduate studies at Michigan
State University. A special note of thanks goes to my supervisor
Miriam Kelley, former head of Family Living Education in Michigan,
who assisted me with admission procedures, particularly confusing
for a foreign student.

During nearly three years of course work in East Lansing, I
was given many opportunities to learn about development problems and
activities from extremely qualified teachers and fellow students with
unique experiences. My perspectives were, indeed, widened through
continuous discussions, sharing of ideas and hopes for amore equali-
tarian world.

The person closest to me, my late husband David Wallender,
had the deepest impact on my professional development. His inde-
pendent, creative thinking and sincere devotion for development work
in Africa have given me much inspiration and challenge.

I am deeply grateful for the warm support of faculty members
and students in the Agricultural Economics Department and in the
College of Human Ecology and other friends at the time of Dave's
death. It helped me through a very difficult time and has contribu-

ted much to the completion of my study program.

ii

I was very fortunate to have Dr. Beatrice Paolucci appointed
as my advisor and dissertation director. For her unfailing support,
encouragement and inspiring guidance, I am deeply indebted. She
went far beyond her responsibilities as a major professor. My
appreciation is also extended to the members of my guidance commit-
tee, Dr. Margaret Bubolz, Dr. Linda Nelson, Dr. James Zuiches and
Nancy Axinn, who provided expert advice and constructive criticism.

Field research in Africa became possible in connection with
an assignment with the Food and Agriculture Organization of the
United Nations to the College of Agriculture, University of Botswana
and Swaziland. The support by FAO and U85 in carrying out this study
has been much appreciated. My supervisor Caroline Fredrickson, my
colleague Britta Ogle, many other colleagues and friends gave
encouragement and made adjustments in order for me to pursue the
research work. I am grateful to the colleagues in the African
Language Department, who helped with the translation of the ques-
tionnaire.

I owe special gratitude to Professor Alastair McArthur,
Lincoln College, New Zealand, who was on the faculty of UBS I975-
l977. He generously took his time for discussions and gave advice
about the design of the study and the analysis of the data. Valu-
able assistance with the computer analysis was provided by Steven
Norton and Hans van Kesteren at the Data Process Unit at the Live-
stock Project at the Ministry of Agriculture, Mbabane. Many offi-

cers in the Ministry of Agriculture, Health and Education, the

Central Statistical Office and the Census Office gave important
assistance.

The hard work of the students as enumerators and coders was
deeply appreciated. Without their contribution the study would not
have been possible to implement. My sincere thanks are also
extended to all the Swazi families who generously provided the
information.

It is my wish that the findings of this study will contrib-

ute to a sound development of home economics programs in Swaziland.

iv

TABLE OF CONTENTS

LIST OF TABLES . . . . . . . . . .

LIST OF

FIGURES . . . . . . . . .

LIST OF APPENDICES . . . . . . . . .

Chapter
I.

II.

111.

INTRODUCTION . . . .

Statement of the Problem . .
Conceptual Framework . . . .
Assumptions . . . . . . .
Definitions . . . . . . .

RELATED RESEARCH AND LITERATURE . .

Conceptual Framework for Fertility

Studies

Population Studies in Developing Countries
Population-Related Studies in Swaziland

PROCEDURES . . . . . .

Research Design . . . . . .
Research Hypotheses . . . . .
Development of Instruments .

Development of the Population Attitude Score

Farm Practice Scale . . .

Measurement of Variables . .
Desired number of children .
Ideal number of children . .

Attitudes towards fertility control

Determining information scores

Utilization of community resources .

Farm practice score . . .
Income . . . . . . . .

Mean number of children of couples .

Resourcefulness .
Data Collection . . .
The Enumerators . .

Training of Enumerators
Field Work . . .

Page
vii

xi

.d

mum on menu—d

dd

Chapter Page
I Method of Analysis . . . . . . . . . . . 34

Model . . . . . . . . . . 35
Duncan Multiple- Range Test . . . . . . . . 36

IV. DESCRIPTION OF RESEARCH AREAS AND SAMPLE . . . . . 38

Swaziland: Background Information . . . 38
Research Areas . . . . . . . . . . . . 4T
Velezizweni . . . . . . . . . . 42
Luyengo . . . . . . . . . . . . 46
Bhunya . . . . . . . . . . . . . . . 50
Sample . . . . . . . . . 54
Characteristics of the Sample . . . . . . . 56

V. FINDINGS . . . . . . . . . . . . . . . 65

Sex As a Source of Variance . . . .
Age As a Source of Variance . .

Number of Children As a Source of Variance
Environment As a Source of Variance . .
Summary . . . . . . . . . .

VI. DISCUSSION, CONCLUSIONS, AND IMPLICATIONS . . . . 78

00000
0
coo.
V
d

Overview of the Study . . . . . . . .
Limitations of the Study . . . . . .
Discussion and Conclusions . . .

Studying Fertility Behavior . .

The Influence of the Environment

The Influence of Sex . . . .

The Influence of Age .

The Influence of Actual Number of Children
Implications for Home Economics Programs
Implications for Development Policy . .
Implications for Further Research . . .

0000
o
0
0090000000.
0
SO
0

APPENDICES . . . . . . . . . . . . . . . . . 102
BIBLIOGRAPHY . . . . . . . . . . . . . . . . . T49

vi

LIST OF TABLES

Table Page
l. Summary of Tested Relationships . . . . . . . . 2T
2. Respondents: Females, Males, and Couples by Area . . 56
3. Age Distribution by Area . . . . . . . . . . . 56
4. Age Distribution of Females by Area . . . . . . . 57
5. Age Distribution of Males by Area . . . . . . . . 57
6. Highest Level of Education Attained by Sex and Area . . 59
7. Religion by Area . . . . . . . . . . . . . 6O
8. Marital Status of Women by Area . . . . . . . . 6T
9. Marital Status of Married Women by Area . . . . . . 61

lO. Husband Present/Not Present by Area . . . . . . . 62
ll. Marital Status of Men by Area . . . . . . . . . 62
12. Number of Wives by Area . . . . . . . . . . . 63
l3. Number of Children by Area . . . . . . . . . . 63
l4. Mean Number of People in Households by Area . . . . 64

l5. Summary of F-Ratios from Least Squares Analyses with
Sex, Age, and Environment as Sources of Variance . . 66

16. Summary of F-Ratios from Least Squares Analyses with
Sex, Desired Number of Children, and Environment
as Sources of Variance . . . . . . . . . . . 67

l7. Summary of Mean Deviations Based on Sex, for Desired
Number of Children, Ideal Number of Children,
Population Attitude Score, and Reproduction
Score . . . . . . . . . . . . . . . . 68

vii

Table Page

l8. Summary of Mean Deviations Based on Age, for Desired
Number of Children, Ideal Number of Children,
Population Attitude Score, and Reproduction
Information Score . . . . . . . . . . . . 69

19. Summary of Mean Deviations Based on Age of Females
on Hygiene Score, Child Care Score, and
Nutrition Score . . . . . . . . . . . . . 69

20. Summary of Mean Deviations Based on Age of Males on
Farm Practice Score . . . . . . . . . . . . 7O

21. Summary of Mean Deviations Based on Age of Males on
Family Utilization Score and Income . . . . . . 7O

22. Summary of Mean Deviations Based on Actual Number of
Children for Desired Number of Children, Ideal
Number of Children, Population Attitude Score
and Reproduction Score . . . . . . . . . . . 7T

23. Summary of Mean Deviations Based on Actual Number of
Children of Couples for Resourcefulness Score '
and Population Attitude Score . . . . . . . . 72

24. Summary of Mean Deviations Based on Environment for
Desired Number of Children, Ideal Number of
Children, Population Attitude Score, and
Reproduction Score . . . . . . . . . . . . 73

25. Summary of Mean Deviations Based on Environment for
Hygiene Score, Child Care Score, and
Nutrition Score . . . . . . . . . . . . . 74

26. Summary of Mean Deviations Based on Environment for
Farm Practice Score . . . . . . . . . . . . 75’

27. Summary of Mean Deviations Based on Environment for
Utilization Score and Income . . . . . . . . . 76

28. Summary of Mean Deviations Based on Environment on

Couples Resourcefulness Score and Population

Attitude Score . . . . . . . . . . . . . 76
29. Summary of Available Community Resources by Area . . . 87

30. Results of the Analysis of Variance Test for
Desired Number of Children . . . . . . . . . I33

viii

Table Page

31. Results of the Analysis of Variance Test for Desired
Number of Children . . . . . . . . . . . . 134

32. Results of the Analysis of Variance Test for Ideal
Number of Children . . . . . . . . . . . . 135

33. Results of the Analysis of Variance Test for Ideal
Number of Children . . . . . . . . . . . . 136

34. Results of the Analysis of Variance Test for
Population Attitude Score . . . . . . . . . . 137

35. Results of the Analysis of Variance Test for
Population Attitude Score . . . . . . . . . . 138

36. Results of the Analysis of Variance Test for Mean
Population Attitude Score . . . . . . . . . . 139

37. Results of the Analysis of Variance Test for
Reproduction Score . . . . . . . . . . . . 140

38. Results of the Analysis of Variance Test for
Reproduction Score . . . . . . . . . . . . 141

39. Results of the Analysis of Variance Test for
Hygiene Score . . . . . . . . . . . . . . 142

40. Results of the Analysis of Variance Test for
Child Care Score . . . . . . . . . . . . . 143

41. Results of the Analysis of Variance Test for
Nutrition Score . . . . . . . . . . . . . 144

42. Results of the Analysis of Variance Test for
Farm Practice Score . . . . . . . . . . . . 145

43. Results of the Analysis of Variance Test for
Utilization Score . . . . . . . . . . . . 146

44. Results of the Analysis of Variance Test for
Income . . . . . . . . . . . . . . . . 147

45. Results of the Analysis of Variance Test for
Resourcefulness Score . . . . . . . . . . . 148

ix

LIST OF FIGURES

Figure . Page
' 1. Model for Viewing Fertility Behavior . . . . . . . 11
2. Map of Swaziland . . . . . . . . . . . . . 131

LIST OF APPENDICES

Appendix Page
A. Complete Questionnaires . . . . . . . . . . . 103
B. Guidance Questions for Unstructured Interviews . . . 127
C. Map of Swaziland . . . . . . . . . . . . . 130
D

Results of Analysis of Variance Tests . . . . . . 132

xi

CHAPTER I
INTRODUCTION

Statement of the PrOblem

One of the most serious threats to human existence in the
world today is the imbalance between resources and population. One
can argue that this imbalance is created by an unequal distribution
of resources between poor and rich countries, between social
classes in poor countries, and even due to an unequal distribution
of resources between men and women. Estimates of food resources and
other raw materials, services, as well as population, are made on a
worldwide scale every year. They have lately shown very gloomy
perspectives (1973).

Far less is known about the situation on the micro-level,
i.e., resources available and utilized by individual families. This
is particularly true for families in transitional stages in a
developing money economy. In order to achieve family goals, as
well as national development goals in congruence with standards
set, organization of resources is of utmost importance. 1

While fertility behavior in the developed world tends to be
increasingly determined by rational family decisions and social and
economic differentials are decreasing in importance, fertility
behavior in the developing world is still puzzling. Interventions

into the traditionally balanced birth and death rates mainly through

improved public health services, as well as disintegration of tra-
ditional family and economic systems, have resulted in rapid popula-
tion growth. An understanding of motivational forces to control
fertility, of which families' fertility standards and goals are an
integral part, can no longer be neglected. Any family-oriented
program with the ultimate goal of improving the quality of life of
individual families, being agriculture or home economics extension,
nutrition, social welfare or community development programs, needs
to concern itself with the understanding of these processes.

With an annual population increase of 2.7 percent, which
means a doubling of the population in less than 26 years (World
Population Data Sheet, 1977), the government of Swaziland is realiz-
ing the importance of slowing pepulation growth (Second National
Development Plan, 1973-77). An official policy is not yet estab-
lished but is expected to be developed soon. It is hoped that this
study will throw some light on processes within families leading to
a fertility decision, and suggests strategies for program develop-
ment within the area of home economics, with an immediate impact on
the development of the Home Economics Diploma Course at the Univer-
sity of Botswana and Swaziland.

This study was undertaken to test relationships of demoa
graphic variables and environments to aspects of fertility decisions
such as goals, standards, attitudes about control of fertility,

levels of information, use of community resources and income.

The overall research questions posed were:

1. What is the relationship of demographic variables such
as sex, age, and number of children to goals and standards of
family size, attitudes about control of fertility, levels of infor-
mation, use of community resources and income?

2. What is the relationship of environments to goals and
standards of family size, attitudes about control of fertility,

levels of information, use of community resources and income?

Conceptual Framework

The conceptual framework employed in this study was a home-
management framework. Management has been defined as a means of
achieving goals through the use of resources (Nickell, Rice, and
Tucker, 1976: 31-53). Gross, Crandall and Kroll (1973) traced the
short history of home-management through the following chronological
stages: resource-centered, humanrcentered, process-centered, and
values-centered. A recent emphasis has been a systems approach
(Deacon and Firebaugh, 1975), of which the behavioral interactional
(Nickell, Rice, and Tucker, 1976) and the family ecological (Steidl,
1969; Hook and Paolucci, 1970; Paolucci, Hall, and Axinn, 1977) are
two examples. The ecological emphasis, stressing the interrelation-
ship between the household unit and its environments, was the most
relevant for this study.

The purpose of management at all levels, national, familial,
or individual, is the conscious improvement of the quality of life.

National development in itself is the application of the concepts of

management, i.e., human controlled development in the effort of
raising levels of living in a nation. Being a part of this process,
families and individuals will need to participate, to an increasing
degree, in management processes learning to manage family situa-
tions. One of the major arenas of family management is that of
making decisions about the size of family in relation to resources
available for maintaining adequate levels of living and the achieve-
ment of an optimum quality of life.

In order to reach goals about family size, resources must
be used. Resources have been classified in different ways--"human
and material, economical and noneconomical, and tangible and
intangible" (Deacon and Firebaugh, 1975: 158). Using any classi-
fication, resources can be regarded as assets consisting of "any
worthwhile possession, personal quality or trait of value, or char-
acteristic of the environment that has use" (Nickell, Rice, and
Tucker, 1976: 34). Consequently, attitudes toward control of
fertility as well as knowledge of hygiene, child care, nutrition,
reproductive biology, modern farm practices, level of education and
income are human resources to be used in decisions about family
size.

The management process consists of several subsystems
requiring decision making: a mental activity leading to the choice
among perceived alternatives best conforming with goals, standards,
and available resources. The managerial process can be broken down
into these four main functions (Nickell, Rice, and Tucker, 1976):

planning to achieve goals, organizing for performance, implementing

the plan, and evaluating the results in light of goals sought. The
management process is complex and continuous, and the functions
interrelated. The outcome of the implementation is fed back to a
new planning phase and long-term goals might be changed, or new goals
might be stated.

The criteria for the evaluation of goals are standards.
Every society has standards which can be traced back to cultural
value based patterns. Standards are dynamic because they stimulate
individuals or groups to action, self-imposed or socially inflicted.
There is a standard set, consciously or unconsciously, for each
kind of activity determined by what is expected to provide maximum
satisfaction.

”Ideal" number of children is an indicator of a family size
standard, while the "desired" family size could be looked upon as
a family size goal.

The home management process is shaped by the environmental
setting in which the family functions. Deacon and Firebaugh (1975)
defined the environment as "a set of elements, conditions, and
properties outside a system that can produce change in the system"
(p. 14). The components of the environment are physical, biological
and social (Steidl, 1964). Using a systems approach, Gross, Cran-
dall and Knoll (1973) concluded about the function of the
environment:

'In general, the transaction between the family and its
environments, whether input or output from the viewpoint

of the family, might be classified as (a) providing
resources or serving as a constraint if appropriate

resources are not available and (b) either motivating or
demanding action. Exchanges might take the form of effort
(either physical, mental, or emotional and either natural
or technical), material objects (animate or inanimate),
men, money, or information (p. 15).

In response to their perception of social and economic
changes in the environment, men and women alter their goals, stan-
dards, and attitudes. In this study the ecological emphasis seemed
appropriate to examine the relationship of environments to aspects

of family size decisions.

Assumptions

 

1. Areas selected represent different stages in
the transformation from a subsistence to a
money economy.

2. Families have a conception of "ideal" and
"desired" family size.

3. Families set goals in order to solve problems,

respond to demand requirements and expecta-
tions.

Definitions

 

Environment: Describes different stages of economic devel-

 

opment from a subsistence to a money economy and accompanying
degrees of traditional Swazi and Westernized life styles.

Family size standard: The number of children reported as

 

"ideal."

Family size goal: The number of children reported as

 

“desired."

Population attitudes: Endorsed or non-endorsed statements

 

of beliefs about family size.

Community resources: Services available to households in

the community.

Family resources: Include the formal level of education,

 

levels of information about hygiene, child care, nutrition, repro-
ductive biology, modern farming methods of males and/or females,
and cash income.

Eamfily; A group of people who make a common arrangement to
provide themselves with food and shelter. A family may occupy more
than one house.

Couple; A man and a woman committed to each other in a

legal or nonlegal union and living together.

CHAPTER II

RELATED RESEARCH AND LITERATURE

The research and related literature will be reviewed under
three major headings--conceptual framework for fertility studies,
population studies in developing countries, and population studies

in Swaziland.

Conceptual Framework for Fertility Studies

Malthus (1803) was a pioneer in studying human populations.
He was the first to point out that the human population cannot grow
indefinitely in geometric progression. The core of his controversial
theory was that poverty and lack of food will check population growth.
He has been criticized for not being able to foresee the technologi—
cal development to come, particularly in relation to food production,
transportation, and birth control methods, and for encouraging
laissez-faire socioeconomic policy. Malthus' arguments did, never-
theless, give rise to the still-continuing polemic about the popu-
lation dilemma.

Since Malthus, fertility behavior has basically been
studied through four main analytical frameworks: the theory of
the demographic transition, the sociological framework, the economic
framework which recently has been related to the fourth and least

explored--the family management framework.

The theory of the demographic transition (Notestein, 1945)
is basically a description of the sequence of events which are
assumed to take place on a macro level when a society moves from
high to low death and birth rates. The theory builds upon the
experience from Europe where the transition started at the time of
the Industrial Revolution and lasted 100-150 years. It does not
actually explain any critical aspects of the problem of the onset of
fertility decline. Generally, the interpretation of the cause of
events has been that decline of death rates, caused by a rise in
living standard as well as advances in such areas as medicine and
sanitation, creates an awareness on the part of individual couples
that fewer children will need to be born to attain desired family
size. Thus, birth rates will fall and at the same time a smaller
family size norm will develop.

Lately this interpretation has been challenged by Coal
(1973) who, after closely studying several European provinces, pro-
poses that fall in birth rate can take place in advance of social
and economic development and in advance of falling death rates.

He found that the main factor for differences in fertility levels
were similarity of cultural and linguistic setting. However, no
particular level of development that triggered a reduction of fer-
tility could be identified.

It is generally assumed that developing countries will go
through the same demographic transition as in the European case.
So far, death rates have fallen drastically from 30-40/l,000 to
15-20/1,000 during a period of 20 years, without a major fall in

TO

birth rates. A slight global drop in birth rates was, however,
recorded in 1977 (Population Reports, Series J, No. 13, Jan. 1977:
J237-J239).

The sociological framework can best be characterized by
the employment of certain variables such as geographic area, urban-
rural residence, socioeconomic status, religion, and ethnic group
in explaining differentials in fertility. Among the white popula-
tion in the USA, fertility differentials have continuously declined
in importance. This is particularly true for the three first-
mentioned variables, while the differentials by ethnic group and
possibly also those by religion have widened in recent years (Kiser,
Wilson, and Campbell: 1968). In developing countries, on the
other hand, the urban-rural differential as well as socioeconomic
status continue to have a strong influence on fertility levels.

Clay and Zuiches (1977) found the identification with the
family of orientation as well as reference groups to be important
intervening variables in shaping family size norms. It appears that
norms and value patterns surrounding family size and childbearing
will increase in importance for future fertility decisions both for
individuals concerned, and consequently for the aggregate growth or
decline of the population.

Davis and Blake (1956) worked out a framework for compara-
tive analysis of fertility. It is built around the obvious steps
in the process of reproduction, generally recognized in human cul-
ture: (l) intercourse, (2) conception, and (3) gestation and par-

turition. They labeled the factors which are connected with

ll

cultural conditions influencing fertility, intermediate variables.
The framework is useful in cross-cultural comparisons of fertility
since all these factors are present in every society and have a
positive or negative effect on fertility.

Including Davis' and Blake's intermediate variables, as
well as norms about the same, Friedman (1967) developed a compre-
hensive sociological framework, covering several variables discussed

above. This model can be illustrated as follows:

 

MORTALITY L,
/ RATES INORMS ABOUT

 

v

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

*' FAMILY SIZE '53 INTERMEDIATE
E , VARIABLES 1-—) l FERTILITY]
c2: /
if
>
Ei‘\\a V

SOCIAL &

ECONOMIC NORMS ABOUT

STRUCTURE INTERMEDIATE

7 VARIABLES

 

 

Figure l.--Model for Viewing Fertility Behavior.

While sociologists have included economic aspects in their
analytical frameworks of studying human fertility behavior, econo-
mists have concentrated on effects of income. Their emphasis has
mainly been on the macro level, studying relationships between
such variables as per capita income and percentage of the population
involved in agriculture on the one hand, and fertility levels on
the other. Century-long observations of industrial countries and

more recently of developing countries suggest that economic progress

12

and decline in fertility go hand-in-hand. Much research has
pointed to the fact that there is a "threshold" or "take-off" point
in economic development below which birth rates will remain high.

Leibenstein (1957) used the concept "critical minimum
effort," meaning that economic backwardness can be characterized
by a set of related factors that possess a certain degree of sta-
bility. They can, in fact, be considered equilibrium valves in the
short run. One of these is the population variable, the value of
which will remain the same if the stimulus of all factors is too
small, that is, if it is below the critical minimum required for per-
sistent economic growth.

A fruitful approach, in explaining the incentive or ration-
ale behind the desire to have larger or smaller families, has rather
recently been adopted from microeconomic theory. Assuming that
parents' fertility decisions are rational, Leibenstein (1957) devel-
oped a benefit-cost analysis of fertility behavior. In other words,
as long as parents have more children, they are deriving utility
from an additional child as a consumption good, as a source of
personal pleasure, as a productive agent, or as a source of security.

Also building on microeconomic theory, Becker (1960) devel-
oped his theory of household choice as the underlying mechanism of
fertility behavior. His thesis is that choices come about as the
result of the interaction between preferences on the one hand, and
income and price constraints, on the other. He found in his analy-
sis that different aspects of children are valued in developed com-

pared to developing countries. In the former countries, parents

13

value consumption aspects, while in the latter, production aspects
of children as economic goods are predominant. Becker's findings
further indicated that increase in women's market wages, as well

as increased cost of her time in household production, will cause a
decline in fertility.

De Tray (1972) elaborated on the utility theory and worked
out a model of expected utility analysis for substitution between
quantity and quality of children. Similarly, he found that female
time and female level of education were of great importance for
increasing child quality with implication for substitutability of

the two inputs.

Population Studies in Developing Countries

Except for censuses, administrated in most developing coun-
tries on a regular basis, KAP surveys have been population studies
most commonly undertaken. These are surveys to determine the
extent of family planning knowledge, attitudes, and practice in a
community, region or country. In 1970, in not less than 67 coun-
tries, of which 15 were African, such studies had been completed or
were underway (The Population Council, 1972).

A KAP survey has four major purposes: descriptive in
regard to knowledge and beliefs about fertility, evaluative provid-
ing baseline data for family planning programs, directive for pro-
gram development, and validative for reporting about acceptance
generated by a family planning program. Questions usually cover

vital data, attitudes toward family size and family limitation,

l4

knowledge about reproductive physiology and contraceptive methods,
as well as practice (The Population Council, 1970).

With the attempt of standardizing questions, this impres-
sive body of cross-cultural data has made comparisons on a global
scale possible. Mauldin (1965) compared findings from 44 studies,
and he found strong motivation for limiting family size: in every
society, two-thirds or more of the persons having four or more chil-
dren indicated that they did not want to have more children. In
another worldwide comparison, Berelson (1966) found similarities in
response by similar stages of economic development and was able to
show that various social differentials corresponded with attitudes
to fertility and family planning.

The KAP approach has been developed within the framework of
the theory of the demographic transition. Caldwell (1976) believes
that the failure to update the theory has led to much misunder-
standing and frustration, as well as inadequate research. He does
not see inadequate methodology as the basic problem but "poor
application, especially in application of methods in cultures other
than those for which they were developed" (p. 336). He points at
four specific pitfalls of current fertility research in developing
countries: (1) the neglect and misunderstanding of the magnitude
and direction of wealth (money, goods, services, guarantees) flows
and potential flows; (2) the fact that the "family" of the fertility
survey is often an artifact of the survey, disregarding the decision

power of the extended family; (3) using only the wife as a

15

spokesperson; (4) too much emphasis on modernization in the attempt

of studying change related to fertility transition.

Population-Related Studies in Swaziland

Population censuses have been carried out decennially in
Swaziland since the beginning of the century. The 1966, as well as
the six previous censuses, was undertaken by the British adminis-
tration. An earlier census, in 1898, was administrated by the South
African Republic in connection with tax collection. The last cen-
sus, August 1976, was the first to be implemented by the Swazi
government. While age, marital status, literacy, occupation, and
ethnic groups have been covered in all censuses since the beginning
of the century, only the censes of 1946 and 1966 included fertility
as a special topic.

The final report from the 1976 census was not available
when this was written. Preliminary figures, however, suggest a
total population of 520,297 people (Central Statistical Office,
1977).

The methodology used in the 1976 as well as the 1966 census
was household enumeration by the canvasser method. Data were col-
lected de juro, since estimates of absentees are important, being
a common phenomenon in Swaziland, which supplies the Republic of
South Africa with a substantial number of immigrant workers. In
1966, 5 percent of the total population of Swaziland was reported
to be absent. Other findings from this census (Jones, 1968: 203)

were that the mean household population was 6.9, with the largest

l6

household size 7.4 in the central region. It was further estimated
that the average African woman in Swaziland, if she lived to be 50,
would bear approximately seven live-born children. Of these seven

children, two can be expected to die before reaching the age of 15,
thus leaving five who will survive to adulthood.

The crude birth rate was estimated at 47-48 per thousand,
the crude death rate at 20-21 per thousand, infant mortality at 172
per thousand live births, and life expectancy at birth to 44 years
(Jones, 1968: 211, 213, 215).

In 1975 the first demographic survey (Gani, 1975) was under-
taken in Swaziland, with the main objective to ascertain whether
birth and death rates were changing significantly from the findings
of the 1966 census. The survey was also undertaken with the pur-
pose of providing a check on the system of vital registration which
was also simultaneously contemplated to be introduced on a compul-
sory, nationwide basis. The findings have, at this writing, not
officially been released.

Although agricultural statistics were registered in the
1966 census, a special "Agricultural Sample Census" (Central Statis-
tical Office, 1973), concentrating on agriculture in the subsistence
sector, was carried out in 1971-72. In addition to commonly
reported population characteristics, characteristics such as "active
in agriculture," "paid employment in agriculture," and "main occu-
pation of population active in agriculture" were registered in a

special section of farm population.

17

A large nationwide survey was undertaken in collaboration
with the Swaziland administration and the University of Natal in
1960 in order to determine development potential and policy in
Swaziland (Holleman, 1964). A multi-disciplinary research approach
was used in covering social, anthropological, demographic, economic,
and agricultural aspects. The survey resulted in a wealth of tabu-
lated information and a statistical instrument of "sufficient sta-
bility to allow comparative analyses to be made from time to time
in order to assess the trends and measure development, and of suffi-
cient flexibility to permit the assessment of local differences,
present and future" (p. 22).

Two studies were undertaken in connection with the estab-
lishment of the Home Economics Diploma Course at the University of
Botswana and Swaziland; one was a survey (Ministry of Agriculture,
1970) in which approximately 150 women were contacted for the pur-
pose of providing baseline data for home economic program activi-
ties, and the other was an investigation resulting in recommendations
for curriculum for the same course (FAO, 1971).

A focus on population attitudes and fertility history is a
part of Allen's in-depth study of "Dimensions of Swazi Households
in Rural and Urban Areas" (1973), as well as Koza's descriptive
"Proposed Guidelines and Models for a Home Economics Family Planning
Program in Swaziland" (1974). Allen found the medians of family
size norms to be three to four children in the urban area and five
to six children in the rural area. She also found that 82.6 percent

of women in the rural area she studied believed that children die

18

more frequently in the current situation than they did in the past.
The figure for the urban area was 65.9 percent.

Another finding from the same study was that the great
majority of women were very vocal in their dislike for polygyny,
which was said to be a source of much quarrel, tension and favor-
itism. Moreover, an overwhelming majority of women in both the
rural and urban area felt that it was really God's will that deter-
mined the number of children women had. They also felt that women
were having more children now than in the past. However, 82 per-
cent in the rural area and 65.9 percent in the urban area wanted
smaller families, and expressed a strong desire to limit their num-
ber of children in order to provide food, clothing and schooling.
Allen (1973), nevertheless, concludes that:

However interested many women are, they have often been
thwarted in their attempts to obtain and/or use birth
control methods. Resistance and antagonism of husbands,
difficulties in obtaining information, expense of the
method, fear for their health--all combine to make the

numbers of women using birth control methods small
(p. 363).

CHAPTER III

PROCEDURES

The overall design of the study can be defined as survey
research with both descriptive and explanatory components. Struc-
tured interviews with men and women were used as a method of gath-
ering data. Sixty-one percent of the people interviewed were
husband and wife. The questions were, with a few exceptions, of
fixed-alternative type. Attitudes towards control of fertility

and farming practices were measured by Guttman type scales.

Research Design

 

The study was undertaken to test relationships between demo-
graphic variables and environments as they relate to fertility
decisions: family size goals, standards, attitudes about fertility
control, and levels of infbrmation about reproduction, hygiene,

child care, nutrition and farm practices.

Research Hypotheses
The following research hypotheses were stated:
Ho 1: There is no effect of sex on . . .
a. desired number of children
b. ideal number of children

c. attitudes about fertility control
d. knowledge of reproduction

19

20

There is no effect of age on . . .

a. desired number of children

b. ideal number of children

c. attitudes about fertility control

d. knowledge of reproduction

There is no effect of age of females on . . .
a. knowledge of hygiene

b. knowledge of child care

c. knowledge of nutrition

There is no effect of age of males on . . .
a. farming practices

There is no effect of age of males on family . . .

a. use of community resources
b. income

There is no effect of actual number of children on . . .
a. desired number of children

b. ideal number of children

c. attitudes about fertility control

d. knowedge about reproduction

There is no effect of actual number of children of couples
on . . .

a. resourcefulness
b. attitudes about fertility control

There is no effect of environment on . . .

a. desired number of children

b. ideal number of children

c. attitudes about fertility control

d. knowledge about reproduction

There is no effect of environment on female . . .
a. knowledge of hygiene

b. knowledge of child care

c. knowledge of nutrition

There is no effect of environment on male . . .

a. farming practices

21

11: There is no effect of environment on family . . .

a. use of community resources
b. income

H 12: There is no effect of environment on couples' . . .
a. resourcefulness

b. attitude about fertility control

TABLE l.--Summary of Tested Relationships.

 

Independent Variables

 

Dependent Variables

Actual No.
Sex Age of Children Environment

 

Desired number of children * * * *
Ideal number of children * * * *
P0pulation attitude score * * * *
Reproduction score * * * *
Hygiene score * *
Child care score * *
Nutrition score * a
Farm practice score * *
Resource utilization score * *
Income * *
Resourcefulness score * *

 

*Indicates tested relationships.

22

Development of Instruments

Two instruments, an A and a B questionnaire for interview-
ing males and females (see Appendix A), were used in collecting
data, both covering, with a few exceptions, identical sections of
questions about family size, marriage, reproduction information,
and the population attitude scale. In addition, the A question-
naire, used for interviewing women, included sections about hygiene,
child care, and nutrition information. In the B questionnaire,
used for interviewing men, sections regarding general household
information, utilization of community resources, cash family income,
and the farm practice scale were similarly added. This dissertation
is based only on a part of the data collected.

Development of the Population
Attitude Scale

Guttman's method of scalogram analysis (Guttman, 1950) was
chosen for measurement of attitudes toward fertility control. The
Guttman scale has been proven to be a useful research tool for
measurement of cumulative change, particularly when examining small
shifts in changes in attitudes.

Items on the Guttman scale have the properties of being
ordinal and cumulative. Both the items and the respondents are dis-
criminated and placed in their order of magnitude. A certain score
obtained on'a scale will indicate exactly which scale item the
respondent has endorsed and which has not been endorsed--with a
certain percent error for the sample as a whole. The degree of

errors or deviation from the perfect scale pattern is commonly

23

expressed in reproducibility coefficient, computed from the follow-
ing formula:

no. of errors

R = 1 - no. of responses

 

The development of the Population Attitude Scale took part
in the following steps:

1. Unstructured interviews, December 1975
Derivation of attitude statement, December 1975
Selection of attitude statement, December 1975
Pilot study, January 1976

Scale analysis, February 1976

01 U" h (.0 N
O O O O

. Reliability testing, March-June 1976

About 60 unstructured interviews with women in different
age groups were first carried out and recorded in different environ-
ments, similar to the selected research areas. (For guidance ques-
tions, see Appendix B). The interviews were conducted in siSwati
by female students from the College of Agriculture, UBS. From the
recorded material, a pool of attitude statements was derived. In
order to select a manageable number of statements for scalogram
analysis the same students were instructed to judge individually
the percentage--0, 20, 40, 60, 80, or 100--of the interviewed women
who would agree to the statements. Statements obtaining an average
percentage ranging from 0-20 and 80-100 were omitted for having a
low discriminatory value. After this procedure, 21 statements

remained.

24

In order to facilitate a scalogram analysis, 80 different
women, again in areas of similar characteristics to the selected
research areas, were interviewed and agreement/disagreement to the
items were recorded. Respondents were assigned a score of one for
each "correct" and a score of zero for each "incorrect" agreement/
disagreement. Statements which had received a high degree of "don't
know" were first cancelled. After scalogram analysis, 14 statements
were finally selected, having a reproducibility coefficient of
0.845, which was considered to be acceptable. Thus the maximum
score of 14 indicates an extremely positive attitude, while a score
of 0 indicates an extremely negative attitude towards the control of
fertility.

The statements with answers adding up to a maximum score
were grouped under headings for agreement and disagreement (see page
25). In the reliability test, as well as in the final instruments,
the items were listed in a random order.

A reliability test of test/retest type was carried out on
30 respondents with about a two-month interval. The correlation

coefficient between the answers was 0.81.

Farm Practice Scale

In the B questionnaire another Guttman scale measuring
farm practices was included (see Appendix A, page 112). This scale
was developed during the maize storage research carried out by the
University in Velezizweni. The scale has a reproducibility coeffi-

cient of 0.91. It was used in all three research areas. Even in

10.

11.

12.

13.

14.

25

POPULATION ATTITUDE SCALE

In order to decide how many children one
wants, husband and wife must cooperate.

It is better to decide on the number of
children one wants, because by doing so
one is able to manage them properly.

Most mothers want fewer children because
then each can get the best in life.

Most mothers want fewer children because
men fail to support them nowadays.

It would be desirable to be able to
decide how many children one gets.

Nowadays there are means to decide how
many children one wants.

In the next generation no woman should
have more than five children.

Parents should have as many children as
they like as long as they can support
them.

One cannot decide how many children one
would like to have.

There is no limit to the number of
children a person should have.

A woman should get many children in
case some die.

More children should be born in Swazi
land.

There is no way of preventing preg-
nancies.

Most women want to have a lot of
children.

Agree

Disagree

26

Bhunya, the selected urban area, men participated somewhere along
the line in the farm management process on the home farm. Ques-
tions about farm practice were therefore not irrelevant. Further,
most of their home farms can be assumed to be located in an area

with similar characters to Velezizweni, the selected rural area.

Measurement of Variables

The variables, except for population attitudes and farm
practices, were measured by short fixed-alternative questions of
multiple choice or yes/no type with follow-up probing questions of
the type "If yes/no, why?" Short open-ended questions were also
used (see Appendix A). Several questions from the Home Economics
Survey in Swaziland (Ministry of Agriculture, 1970) were adopted and
included in the study.

Separate sheets, listing available environmental resources,
were developed for each research area (see Appendix A, pages 108-
111). Based on the information recorded, ratios were developed to
measure use of community resources.

Composite scores were computed to measure levels of informa-
tion and resourcefulness. Below are given maximum scores as well as
mean scores and standard deviations for all of the measures employed
in the study. Figures are based on data from the largest sample of

this study.

Desired number of children.--Desired number of children is
the number answered on the question: "If you would have exactly the

number of children you want, what would that number be?"

27

Ideal number of children.--Ideal number of children is the
number answered on the question: "What do you think would be the
ideal number of children for Swazi families, if they could control

the number as they wish?"

Attitudes towards fertility control.--Attitudes about con- //

trol of fertility were measured by an attitude scale of Guttman type.
The maximum score was 14, the mean, 8.89, and the standard devi-
ation, 2.91. Couples' attitude about fertility control was the mean

of the men's and women's attitude scores.

Determining information scores.--Leve1 of information of
reproduction, hygiene, child care, and nutrition was measured by
computing the composite scores on selected questions in the respec-
tive sections of the questionnaires. Questions were of multiple
choice and yes/no types, indicating level of information and prac-
tices. "Correct" answers are indicated by an asterisk (*) in
Appendix A for respective questions.

In determining reproductive biology score, questions number
1, 2, 3, 4, 5, 6 and 7 from the section "Reproduction" (see Appen-
dix A, page 122) were used. One point was given for each correct
answer. The maximum score was 7, the mean was 2.74, and the stan-
dard deviation, 1.02.

In the same way, questions 1, 2, 3, 5, 6, 7, and 8 from the
section “Hygiene" were used to measure the level of hygiene informa-
tion (page 120). Maximum score was 8, the mean, 4.16, and the

standard deviation, 1.76.

28

In measuring the level of information of child care, ques-
tions 1, 3, 4, 5, 6A, 68, 7, and 8 in the section "Child Care"
were used (page 124). The maximum score was 8, the mean, 5.31, and
the standard deviation, 1.25.

Nutrition information was measured by questions 1, 2, 3, 4,
5, 6, 7, and 8 in the section "Nutrition" (page 125). The maximum

scre was 8, the mean, 3.37, and the standard deviation, 1.42.

Utilization of community resources.--The degree of utiliza-
tion of community resources was the ratio between utilized resources
by the family, versus available resources, as reported by husbands.

The mean was 48.1538 and the standard deviation, 2.12.

Farm practice score.--Farm practice score was measured by a

 

Guttman type scale (see Appendix A, page 112). The maximum score

was 10, the mean, 3.78, and the standard deviation, 2.95.

Income.--Income was the cash family income during the pre-
vious year as reported by husbands. Figures were coded in 10

classes.

Mean number of children of couples.--Mean number of children
of couples was the mean number of children of the man's and woman's

children.

Resourcefulness.--Coup1es' resourcefulness was the sum of

 

the scores on the farm practice scale, hygiene, child care,

29

nutrition, and reproductive information. The total maximum score

was 41, the mean, 23.70, and the standard deviation, 4.76.

A calendar of events recently updated for use in the census
was used as an estimation when age was not known.

At an early stage questions were distributed to specialists
in areas covered in the questionnaires, asking for their reaction
and comments. For example, questions dealing with nutrition knowl-
edge and practice were given to both national and expatriate nutri-
tionists in the Ministry of Agriculture and at the University.
Questions on child care, hygiene, health and family planning were
in the same manner discussed with experts in the public health units
of the Ministry of Health. Census staff were contacted about the
census type questions. The complete first drafts of the question-
naires were further discussed with a number of Home Economic Exten-
sion Agents, particularly with those working in the research area.

After revision of the first draft questionnaires, second
drafts were tested in 25 interviews with both women and men. The
major finding during these interviews was that the sequence of the
sections needed to be changed in such a way that questions demanding
more thought and consideration by the respondents needed to be at
the beginning of the interview, while easier questions were left to
the end.

After preliminary tabulation, revision, and editing the
final questionnaires of the English version took shape. They were

translated into siSwati by the African Language Department at the

30

University, and later on back-translated by staff who had not pre-
viously been involved. On several points, problems of linguistic
equivalence were thus brought up and discussed. This procedure

was again followed through by the enumerators, some of whom had not

earlier participated in the preparatory phase of the study.

Data Collection

About one year before the actual data collection took place,
authorities on a national, district, and also local level were
informed about the planning of the study, and their approval was
obtained. Governmental expertise, already involved in the develop-
ment of the instrument, was kept up-to-date with the implementation.
The district commissioner of Manzini district, in which all three
research areas, Velezizweni, Luyengo and Bhunya, are located, was
contacted again and enumerators introduced to him just before the
start of the actual data collecting period. Agricultural and home
economics personnel were similarly contacted. The census office
was informed about the exact dates for carrying out interviews in
different areas to help prevent any confusion the study might create
considering the implementation of the 1976 census about a month
later.

In Swaziland it is of special importance to have the
approval of the local chief before indulging in an endeavor such
as data collection. Without his sanction his subjects will not par-

ticipate. 0n the other hand, if the right procedure has been

31

followed and the chief has given his approval, almost any informa-
tion can be asked for.

In Velezizweni, where University staff were already well-
known, and a brother of the chief was engaged as a part-time
enumerator and interpreter, and the fact that the author had visited
the area several times during the previous year, some short-cuts
could be taken. No meeting needed to be held at the chief's kraal;
introduction by his brother of the enumerators to individual fami-
lies was considered to be enough. It was important, however, that
the first interviews had been held at the chief's kraal. His senior
wife was interviewed and some additional household members. Unfor-
tunately the chief, himself, was away during the time for data
collection in Velezizweni.

In Luyengo, in the same manner, an elder, appointed by the
chief and employed as an agricultural laborer, accompanied the
enumerators. This was particularly important in Luyengo, since
resentment towards the University still lingered. Already, at the
meeting with the chief a year earlier, the old man was appointed
for this special task. During regular meetings with elders in the
Luyengo chieftainship, the chief also infbrmed others about the
planned study.

The situation in Bhunya, not a traditional chieftainship,
was different. After approval by the company and the town clerk, a
notice was placed in the weekly local newspaper of Bhunya, explain-

ing the purpose of the study and asking for the people's

32

cooperation. The enumerators were apprehensive at the beginning
about interviewing people in this community, but only one refusal
was recorded.

In all three areas a male and female enumerator approached
the families together and, if possible, individually interviewed
husband and wife. The male enumerator interviewed the man, while
the female enumerator interviewed the woman. If only one of them
was at home he/she was interviewed if this was convenient, and an
appointment was made with the other partner for a later occasion.
However, the most common procedure was that appointments were made
with everyone a day in advance. It was surprisingly easy to carry
out interviews without either spouse or other household members

standing around.

The Enumerators

The enumerators were four female students enrolled in the
Home Economics Diploma Course, and three male students, one enrolled
in the General Agriculture Diploma Course and two in the Animal
Health Course at the University, Luyengo campus. All had completed
their first year of training. The male students were specially
employed by the home economics department, while the data gathering
experience was considered as part of the required "extension prac-
ticals" for the female students. In Velezizweni, after the data
gathering period, they also carried out food and nutrition demonstra-
tions. This was planned to take place in all the three research

areas, but due to unexpected circumstances it could not be

33

implemented in Luyengo and Bhunya in connection with the actual

data gathering period.

Training of Enumerators

A week was allocated for the training of the enumerators
before the field work in order for them to be well informed about
the overall objectives of the study, to be familiar with the
lengthy and, in their opinion, difficult questionnaires, and to
practice skills in interviewing. The female students had previously
been involved in the development of the questionnaires and were
familiar with the study. They had also completed a course in basic
research methods with the focus on family studies. The introduction
as well as transitional and explanatory statements before each new
section of the questionnaires were also worked out together with the

students during this period.

Field Work

During the field work an excellent team spirit developed
among the students. Their attitudes towards their work, the fami-
lies they interviewed, and other people contacted were the best pos-
sible. In brief, they were eager, cooperative, and sensitive.

A general observation was that a lot of learning took place
during this period. Contributing to this ideal situation were also
the two drivers who were in charge of the four-wheel drive vehicle
used throughout this period. Usually, after a reporting and plan-
ning session in the morning, the team stayed out in the field the

whole day having brought packed lunches. In Luyengo and Bhunya,

34

interviewing also took place in the evenings, the only possible time
to meet some of the wage earners. The headquarters for the team was

the Luyengo campus.

Method of Analysis
The purpose of the investigation was to find the effect of
such factors as sex, age, number of children, and environment on
the dependent variables covered in the study. The statistical

analysis used provides:

a. Unbiased estimates of the effects of each level of
each factor.

b. Significance tests such that the

i. null hypothesis that effects of the various
levels of a factor are zero can be either
accepted or rejected at some satisfactory
level of probability.

ii. null hypothesis that the difference between
any two levels of a factor can be accepted
or rejected.

In this study the method of least squares was used to make
unbiased estimates of the factor effects. This procedure is
required because of the occurrence of unequal subclass numbers
which result in confounded estimates of the effects if simple cross-
classification is used to analyze the data. Least squares analysis
also provides mean squares which through the analysis of variance
give estimates of the variance components associated with each
source of variation which makes it possible to apply the F-test of

significance and Duncan's least significant range test.

35

Model
The theoretical model underlying the data, for the case of
two sources of variance affecting one dependent variable, is given

below as an example of the statistical method of analysis.

(1) Yijk = u + Pi + Aj + (PA)ij + eijk

kth

where Yijk is the value of the dependent variable of the respon-

h

dent in the it level of independent variable 1 and belonging to the

jth level of independent variable 2: u is the population mean; Pi

is the effect of the ith level of independent variable 1; Aj is the
effect of the jth level of the independent variable 2; and (PA)ij
is the effect of the interaction between the independent variables.

is a random error term associated with the response of
th

eijk
kth

level of independent variable 1 and

the respondent in the i

th level of the independent variable 2. 2.. is

13k
assumed to be normally distributed with a mean of zero and a stan-

belonging to the j

dard deviation of o. The variance is assumed to be homogenous.
The model also assumes that the effects are additive and

that

.= .+ .. .= ,, °=
2P1 ZAJ §(PA)1J for all J §(PA)1J for all 1 0

In this investigation the subclass numbers are unequal, that is, "ij
in each cell are not equal.

While the estimate of the effects (A1, A2, . . . , P1 . . .)
is straightforward with equal subclass numbers, this procedure gives

biased estimates if applied to cross-classified models with unequal

36

subclass sizes. Statistical methods using the principle of least
squares have been evolved by Kempthorne (1952), Cunningham and
Henderson (1968), and Harvey (1960). A version of Harvey's method
was available in Fortran code at the Ministry of Agriculture,
Mbabane. Hence, it was possible to use Harvey's routine to make
estimates of the factor effects and to apply significance tests. A
limitation of the program was that cell means were not provided, only
deviations from the grand mean. This shortcoming made it impossible
to interpret interaction effects.

In this study a .01 alpha level was chosen. Repeated analy-
ses, which is less than ideal, necessitated this level. Results of

these analysis.of variance tests are reported in Appendix 0.

Duncan Multiple-Range Test

In order to test which of the means differed significantly,
Duncan's multiple-range test was applied (Harter, l960).* Essen-
tially all possible hypotheses of the type u1 - “2 = 0 at a speci-
fied significance level are tested. In this study the .05 level

was chosen.
To be significantly different the range of any subset of

sample means must exceed a certain value. This value is called the

least significant range (Rp) for the (p) means, assuming that

 

*While Duncan's test is strictly applicable to situations
where the number of replications within each class are equal, in
most classes in the study the numbers were spread fairly evenly
between classes so that Duncan's test seemed appropriate.

37

(k) random samples are all of equal size (n). Rp is computed

through the following formula:

Jszln

Rp = rm

2 is the sample variance obtained from the error mean square

where s
in the analysis of variance table, and rm is least significant stu-
dentized range depending on desired level of significance and the
number of degrees of freedom of the error mean square. The rm is
obtained from Duncan's tables and n is the total number of observa-
tions divided by the number of classes.

After arranging the sample means in increasing order of
magnitude, comparisons can be made between the least significant
ranges (Rm) and the differences in the ordered means. An absolute
value greater than Rm indicates significant differences. In Tables
16-27 this is indicated by letters assigned to means such that

those sharing a common letter are not significantly different from

one another.

CHAPTER IV

DESCRIPTION OF RESEARCH AREAS AND SAMPLE

This chapter will cover selected background information of
Swaziland and a description of research areas, sample, and charac-

teristics of the sample.

Swaziland: Background Information

Swaziland is one of the smallest countries in Africa and is
situated in southeastern Africa between the Republic of South Africa
and Mozambique, without any direct access to the sea. The country
is divided into four clearly distinguishable climatic and geographic
zones, extending longitudinally from north to south. 0f highest
elevation is the ”Highveld" region, where two of the research areas
in this study (Velezizewni and Bhunya) are located. Characteristics
of the zone are steep, rocky hills, which in some areas are rich in
mineral resources. Subsistence farming predominates, with maize as
the most important crop, while afforestation projects have been
developed on a commercial scale. The "Middleveld," where the third
research area (Luyengo) is located, extends to the east of the
"Highveld." This area has rolling topography, with tall grassland
and good cultivatable soils. It has always been the most heavily
populated region of Swaziland, and is where the influence of Western

civilization has been the strongest, resulting in a higher degree

38

39

of industrialization and urbanization than in other parts of the
country.

In contrast to many African nations, Swaziland is fortunate
to be ethnically homogenous. The Dlamini clan, who form the nucleus
of the Swazi Nation, is a Bantu speaking group mainly Nguni in
language and customs (Schapera, 1966: 45-47). They settled in the
country at the end of the eighteenth century. The last thirty years
of the nineteenth century were marked by permanent European settle-
ments and their economic conquest of the country. Boers sought new
farming land and better grazing for their sheep, while Britons
mainly engaged in mining and the establishment of business. This
was a tragic period of Swazi history with an intensive economic and
cultural power struggle between Europeans and Swazis, which resulted
in the loss of Swazi political independence. The country was for
a short period administered from the Transvaal Republic, but the
British government gained control of the territory in 1902 after a
victory in the Anglo-Boer War. In 1968 Swaziland gained total
national independence and became a constitutional monarchy.

Swaziland is comparatively rich in natural resources, and
has experienced a more rapid rate of economic development than most
African countries. Nevertheless, economic development is a local-
ized phenomenon in Swaziland and it is relevant to refer, as Fair,
Murdoch and Jones (1969) do, to "Islands of Development." Tradi-
tional subsistence economy is still existing side-by-side with

modern money economy. Swaziland's economy is heavily dependent on

40

the economy of the Republic of South Africa. Migration of laborfrom
Swaziland is still a common phenomenon and the two nations form
together with Botswana and Lesotho a customs union.

The Swaziland government faced the usual challenges of
developing countries in transforming their country to a modern
nation. A rapid expansion of the formal educational system has taken
place in Swaziland since independence. In 1972, 50 percent of young
people aged 5-19 were attending school (Second National Development
Plan, 1973-1977: 151). Education is not free in Swaziland but sub-
sidized by the government and to a great extent by religious and
private organizations.

0n the average the Swazis are healthier than many other
peOple in developing countries, partly because many tropical dis-
eases are not endemic and partly because of the fast expansion of
curative and preventive services. Infant mortality is, however,
still high.

The main approach of family planning services has been to
integrate family planning into public health services; into the regu-
lar maternal health program. All governmental health centers offer
family planning services daily, while at the rural clinics services
are more irregular depending on the training of personnel and time
available. The Pill and the IUD are the most common devices, and at
rural clinics given free of charge. Before women are given contra-
ceptives they must present a signed form of approval by their hus-

bands or, in case of an unmarried woman, by the father. It is not

41

usually a policy to extend family planning services to young unmar-
ried women who have not yet had a child.

Agricultural and home economics extension programs are car-
ried out by the Ministry of Agriculture. Agricultural extension
started in the 19405 and has been of moderate success. To make pro-
grams more effective, extension efforts have been concentrated on
pilot areas and on key programs. Approximately five percent of rural
women participate regularly in home economics extension programs.
The home economics extension agents, "domestic science demonstra-
tors," work through women's associations, known as the Zenzele
groups, originally developed for this purpose in the early 19505.
The major components of the home economics extension program are
food and nutrition education, family life education, child develop-
ment, home gardening, budgeting and handicraft ("Programmes Related
to Family Living in Swaziland: A Handbook," Luyengo, April 1967:
53).

Research Areas
Three areas of diversified environmental characteristics
were selected for the implementation of the study. The three
selected areas were Velezizweni, Luyengo and Bhunya. They span from
one extreme to the other of development stages existing in Swaziland
at the time of the study. Velezizweni represents areas of a high
degree of traditional social and subsistence economic characteris-

tics, while Bhunya represents the other extreme of a high degree of

42

Westernized life style and money economy. Luyengo is assumed to

fall in between the two extremes.

Velezizweni

Velezizweni is located in the Highveld region in Manzini
District, approximately 10 km southeast of the town of Mankayane
(see map, Appendix C). The area is accessible all the year round
by gravel road from both Mankayane and Luyengo. Twenty-three house-
holds in a small valley, bounded by hills and the Mankayane-Luyengo
road, were, since 1974, already involved in research carried out by
the Faculty of Agriculture, UBS (Rose, 1975). A good rapport had
been established with the farm families, which provided an ideal
situation for the planned study, which would cover very personal
areas of family life. The households are scattered, following the
usual pattern in rural areas in Swaziland.

Velezizweni is located in a setting typical for that of the
Highveld region. The views over the hills and gorges are breath-
taking. The climate during the winter months is sunny and dry with
cold nights, while there are heavy rains during the summer months.
Even during the winter, however, there is seldom a shortage of
water for household consumption. Natural creeks and springs usually
provide enough water for livestock and irrigation projects.

The soil in Velezizweni is rocky and acid. The predomina-
ting crop is white maize, but various legumes, sorghum and cotton
are also grown. As is the case regarding most traditional Swazi

Nation land, the farmland in Velezizweni is fragmented. Most

43

farmers cultivate "lands“ close to the homestead, but also in other
places further away.

The majority of families in Velezizweni claim that they do
not normally purchase maize. Gifts of maize, however, seem impor-
tant between related families. A credit system in maize also seems
to exist in such a way that households not producing sufficient
maize to meet their needs over a particular season borrow from
another farm. They are expected to pay back the following season.

During the 1973/1974 season, 8 of the 23 families in the
valley sold maize to local traders immedately after the crop was
shelled (Rose, 1975).

Vegetables from the irrigated, communally run garden situ-
ated five km away, as well as handicrafts such as grass mats,
baskets, crocheted items and beadwork are marketed in Mankayane or
Manzini. These products are usually transported by the regular
passenger bus service.

Cattle breeding is an integral part of the economic system
of Velezizweni. Each family owns on the average about 18 head of
cattle. Besides giving the owner prestige, cattle are easily con-
verted into cash when a need arises. Many families also keep goats
and pigs, chiefly for household consumption. Poultry is kept in
almost every kraal.*

Velezizweni is governed by a Chief of the Dlamini clan, who

is appointed by the King following the traditional pattern. His

 

*Kraal means compound and is often used synonymously with
homestead.

44

kraal stands out from the rest of the kraals in the area, compris-
ing many individual huts. It is the natural meeting place fOr
meetings of different kinds. His subjects are expected to pay tax
to him and participate in work parties on his land or in his house-
hold. Several of the families living in Velezizweni are related to
the Chief. He has four wives and many children.

Most families in Velezizweni are, however, monogamous.
Except for the Chief's household, only three other husband/wife
couples were polygamous. 0n the other hand, six of the women inter-
viewed, not living with their husbands, claimed that they were
either senior or junior wives.

The size of families in Velezizweni varies from very small,
consisting of single women and their children, to very large with
many members additional to the immediate family. In many cases the
husband is not present and works as a wage earner in towns in Swazi-
land or abroad, coming home only to visit. In a few cases the
opposite situation exists, where the wife is working elsewhere,
often as a teacher or nurse, while the husband is staying in the
rural home.

A common family situation in Velezizweni is also that the
mother of the head of the household, usually widowed, lives together
with younger brothers and sisters of the head of the household, and
grandchildren, while the head of the household is living with his
immediate family in another place.

At the time of the study, farming activities were low in

Velezizweni. The farming season was over and little else but dry

45

grass could be seen around the farms. Some farmers and their wives
were involved in preparations for agriculture and home economics
shows, respectively--shows taking place during this time of year.
It is also the time for repairing and building new houses. Women
spend much time cutting thatching grass on the veld and men are
repairing the roofs. Beer parties take place on a rotating basis
all the year round, and are well attended.

There are two primary schools in Velezizweni--Velezizweni
Primary School and South African General Mission (SAGM) Primary
School, both supervised and run by the Swazi government. The SAGM
also arranges Sunday School teaching in three places in Velezizweni
every Sunday. The nearest junior high school is located at
Naqbaneni, about 20 km towards Luyengo, while a senior high school
is located in Mankayane town, 10 km away. Literary classes are
also conducted in the Mankayane area.

There are four churches in Velezizweni--Zionist, Methodist,
Apostolic, and Roman Catholic. The nearest medical facility is the
hospital in Mankayane. There are, however, two "African Doctors"
in the area itself, one of whom is a woman.

The Agricultural Extension Field Officer lives in the area.
Several farmers have frequent contacts with him, resulting in the
adoption of improved farming practices. The Domestic Science
Demonstrator assigned to Velezizweni lives in Mankayane, and carries
out extension activities in the area, but on a less regular basis.

A small grocery store is located in the area, but carries a

very limited assortment of goods.

46

Luyengo

The majority of the families interviewed in Luyengo live
next to the southwest border of the campus of the Agricultural Col-
lege at the University of Botswana and Swaziland. The college is
situated 37 km southeast of Mbabane and 28 km west of Manzini.
Before the establishment of the college in 1966, these families were
living on the campus grounds. When the King donated the land to
the establishment of the college they were moved to the present
settlement. For many families this was the second forced move in a
generation, and the decision caused much rebellion and resentment.

The main source of income for these families previously
was from breeding cattle. Since the 12 acres of land allotted to
each when moved could not support the number of cattle they owned,
it was stipulated that they had to limit the number of cattle to
five. It was suggested these cattle should be dairy cows, and milk
could thus be sold to the college community, providing families with
a new source of cash income. However, soon the college obtained
their own dairy herd, producing sufficient milk to supply the col-
lege community.

As a compensation for these sacrifices, housing of improved
standard was provided, water supply in terms of an irrigation chan-
nel was made available, and a cash lump sum was paid to each family.
Expert advice and equipment from the college were further made
available and utilized in developing the new settlement. Neverthe-
less, the relationship between the people and college continued to

be tense, and extension efforts from the college ceased. However,

47

over time the situation has normalized, and at the time of this study
people were found to be generally content with their situation.
Several men and women are now employed by the college or private
families in the college community.

No hostility was experienced by the enumerators and people
were cooperative in providing the information which was asked for.
This was doubtless due partly to the presence of the elder appointed
by the Chief to accompany the enumerators.

The settlement consists of two main groupings of houses:
one to the south of a creek which, at the western end of the settle-
ment joins the Usutu River, and the other is situated alongside the
Luyengo-Mankayane road between the point where the main Bhunya-
Manzini road and the Usutu River join. No interviews in the final
study were carried out in the former area since the instruments were
tested among some of those families.

The Luyengo area is located in the Malkerns Valley in the
Middleveld region at an altitude of 750 meters, with a climate that
could be described as ideal. An outstanding feature of this area is
the vast pineapple plantations expanding over the land. The valley
is surrounded by semi-mountains, which in the north are covered by
a man-made pine forest. The area is doubtless one of the most
attractive in Swaziland.

The area is part of one of the "corridors of development"
in Swaziland, stretching from Manzini to Bhunya. The most obvious
factor that has contributed to the high degree of development is

the irrgation system which was built and developed from the Usutu

48

River during the early 19505. At that time fruit farming, con-
sisting mainly of pineapple and citrus, was started by white set-
tler5--many of Brisih origin--on freehold land. A fruit cannery
was also established in the Malkerns village, five km northeast of
Luyengo. Production mainly consists of pineapple and citrus which
are processed for export. Approximately 3,000 workers are employed
in the factory and on the plantations.

An Agricultural Research Station is also located in the
area--it is associated with the College, employing about 120 people
in Luyengo. During agricultural seasons there is a shortage of
labor in the area. Labor is then transported to the Valley from
areas further away. The majority of agricultural and factory workers
are women.

The nearness of the University, commercial activities, and
tarred road with fast and frequent bus services to Manzini and
Mbabane place Luyengo somewhere in the middle of an urban-rural con-
tinuum. The area has many features in common with Velezizweni, with
the major part of the population still involved in farming activi-
ties, but more on a part-time basis, with one or more family members
earning wages in the neighborhood.

MoSt families cultivate maize, and vegetables such as cab-
bage, tomatoes and onions. The maize is usually consumed in the
household, while vegetables are also sold. Transporting vegetables
to markets is, however, difficult, and often products are Spoiled in
the field. Other sources of cash income are sales of handicraft

products and home-brewed beer.

49

The houses provided by the government are of similar design
--commonly, two bedrooms, living-room and kitchen. Walls are made
of cement blocks and roofs of corrugated iron. Some of the houses
have been extended, while, in addition, other types of houses such
as the traditional rural hut and typical square-shaped urban house
have been built. Water is fetched from the irrigation ditches or
the river. There is no electricity in the settlement.

Similar to Velezizweni, many families are split in Luyengo.
A large number of homes are occupied by single women, widowed or
with husbands living in other places. Monogamous family structure
predominates. Only four of the couples contacted lived in polyga-
mous unions.

There is one primary school, one junior high school, and one
secondary school in Luyengo--the latter only for boys. In spite
of the fact that the area is better provided with educational
facilities than most other semi-rural areas, there is keen competi-
tion for places in the schools.

There is one Zionist and one Anglican church in the near
neighborhood of Luyengo.

Towards the northern part of the settlement there is a large
store run by Indians, carrying groceries, fabrics, clothes, shoes,
and household equipment. A butchery and petrol station are also
run in conjunction with the store, as well as an outdoor vegetable
market. Warm local food and drinks are also sold at the vegetable

market. Opposite, on the road, a "Beer Hall" has recently been

50

established, and further along the main Bhunya-Manzini road there
are two smaller grocery stores.

The main bus stop is located where the Mankayne road joins
the Bhunya-Manzini road. People are seen there waiting for or
exchanging buses throughout the day. Buses also stop in front of
the Indian store and along the road towards the river.

A Government Health Clinic is situated a short distance
from the settlement. The charge is 30c per consultation. A maternal
and child health program is carried out, as well as family planning
services.

An Agricultural Extension Field Officer and a Domestic Sci-
ence Demonstrator live in the settlement, but since they both have
a wide area to cover, extension activities have not been intensive
in the settlement. Only recently, a "Women's Club" was started in

connection with the United Nations "Women's International Year."

Bhunya

The community of Bhunya is located in the middle of the
Usutu Forest--one of the largest man-made pine forests in the world.
The distance from Luyengo is 21 km in a northwest direction. The
climatic conditions of this Highveld region provided ideal condi-
tions for a gigantic afforestation project. In 1949 the planting
of pine trees started on the 100,000 acres which the forest now
covers. The rolling hills, previously not very productive, now

produce trees ideal for pulping in a 10-20 year cycle. In 1959, a

51

pulp mill was established at Bhunya, with a capacity of producing
100,000 tons of unbleached kraft pulp per year for export.

The creation of the community of Bhunya is directly linked
to this industry. Housing for the employees had to be provided.
Bhunya was planned for unskilled and semi-skilled workers, while
another community living at Mhlambanyati, a village 15 km away, was
established for employees in higher income brackets. The Usutu Pulp
Company, Ltd., employed in 1976 approximately 2,500 people, of whom
529 are living in Bhunya. With dependents, the population of Bhunya
comprises approximately 1,500 people. Bhunya is an all-black commu-
nity, while the proportion of black to white inhabitants in
Mhlambanyati is one to nine. About five percent of the employees
are women in Bhunya.

The streets in Bhunya are all connected to a big, open,
circular grassed area in the center of the village. A large general
store, butchery and post office are located in the northern part of
the open place. In this area are the bus station, the market, where
vegetables and hot prepared food are on sale, a restaurant, the
social worker's office, and the town clerk's office. Further to the
north and downhill are the stadium, sports fields, tennis courts,
the community hall and the club. There are seven churches in Bhunya
--all located in the western part of the village. A large primary
school and a junior secondary school are the main educational
facilities.

Houses are of one-family or semi-detached type, with

internal or external ablution facilities. The smallest unit

52

consists of two bedrooms and kitchen, while other units have a liv-
ing room, also. In a recently built extension of Bhunya--New
Bhunya--located in the eastern part of the community, all houses

are of one-family type, with three bedrooms, living-room, kitchen
and internal ablution facilities. Houses with four rooms and a
central kitchen, designed for four single people, have also recently
been built in the center of the village. Altogether there are 388
family and 140 single units. The rentals for units vary from E2-
E8,* which is a subsidized rate.

Vegetables are often grown outside the houses--mainly cab-
bage, onions and carrots. A nursery, privately owned, sells plants
to families. There is also a Poultry Association run as a coopera-
tive, to which the company has allocated land.

Most inhabitants in Bhunya consider their stay as temporary,
although the labor situation is close to being static. Almost
everyone has a home somewhere else to which they will return when
employment is terminated. Workers visit their homes over weekends
and holidays, or the family will come to Bhunya to visit. Sometimes
the wife and children return to the permanent home after an initial
stay in Bhunya, and sometimes a worker brings a part of his family--
for example, one wife and her children, or maybe only one child for
the purpose of keeping him company and attending school--while the
rest of the family stays in another place. Nevertheless, the
majority of inhabitants in Bhunya seem to live in what, with caution,

could be called a nuclear family system.

 

*In August 1976 one Emalangeni equaled 0.87 US dollars.

53

There is always money in cash around in Bhunya. The
majority of employees living there are classified as laborer,
unskilled and semi-skilled, and are paid hourly with an income per
month ranging from E53-E80. Artisans, chargehands, supervisors,
laboratory technicians, and chemical process operators are usually
paid weekly, earning on an average E150-E450 per month, while
chargehands, supervisors and administrative staff are paid monthly,
with salaries ranging from E1,000-E3,000 per year.

It is obvious that the relationship between productivity
and the health status of the worker is recognized by the company.
Free medical service is available at the company clinic for workers
not earning over approximately E80 per month, which includes their
dependents. At higher income levels, workers are eligible to join
a health insurance scheme at a cost of E4 per month, with a coverage
of 80 percent of health expenses--also for treatment in other places
than Bhunya. The health care system is nevertheless highly subsi-
dized by the company.

There is no family planning program, as such, implemented by
the Clinic, although family planning services are available. The
Pill is the main means of contraception. The Clinic charges the
wholesale price plus a five percent margin which comes to 60¢ per
cycle. The justification for not distributing the Pill free of
charge is that women not being able to afford the charge can be
referred to the Government Clinic just outside Bhunya. From a rough
estimate taken from purchase orders at the company clinic covering

the previous year, about 25 percent of women in Bhunya are on the

54

Pill. No approval from husbands, unlike at governmental clinics,
is asked for. A pre- and post-natal program is carried out by staff
at the clinic. Most pregnant women living permanently in Bhunya are
believed to be reached through this program.

A well-qualified social worker is employed by the company,
mainly to work with families living in villages in the forest.
She, however, also leads a women's group in Bhunya, having on their
program nutrition, food preservation, health, child care, handi-
crafts, and a savingscheme. Every afternoon she also welcomes a
playgroup of children 13-16 years of age to her office/demonstration
center. Women in Bhunya are also involved in the Red Cross Society
and in the Zondle School Feeding Scheme.

Sport and music are important leisure-time activities in
Bhunya. Tournaments of soccer between teams of workers from differ-
ent departments at the mill, or between Bhunya teams and teams
coming from outside, are common weekend activities. Competitions
between groups of singers are also organized. The club is run
entirely by the community. Films are shown there regularly, and
other activities such as darts, and table-tennis, take place there.
Much social activity is associated with the churches. Generally,

church attendance is high in Bhunya.

Sample
In order to obtain an adequate sample size for the analysis,

the criterion was set that at least 50 interviews from at least 30

different households should be carried out in each of the three

55

research areas. For the same reason, in at least 15 households in
each area both husband and wife should be interviewed. In
Velezizweni and Luyengo the total population within determined boun-
daries and some additional households were interviewed, while a
random sample was drawn from the total population in Bhunya.

In Velezizweni the 23 households in the valley and 7 addi-
tional on the opposite Side of the road were contacted. In Luyengo
the total population of 27 households alongside the Luyengo-
Mankayane road and 9 additional families north of the Bhunya-Manzini
road, close to the junction with the Luyengo-Mankayane road were
added. The latter area is named Mhlabubovu and has very similar
characteristics to the settlement area.

From the total population of Bhunya, excluding single people
without children (according to the records in the town clerk's
office), 40 households were selected at random With the help of
records from the town clerk's office, and a random table. Of these,
one family was on leave, two refused to be interviewed, and in six
no one was at home in spite of repeated calls, leaving the number
of contacted households at 31. The total number of people inter-
viewed was 54.

All in all, 169 men and women were interviewed: 52 in
Velezizweni, 63 in Luyengo and 54 in Bhunya. Table 2 presents a

summary of contacted males, females and couples in the three areas.

56

TABLE 2.--Respondents: Females, Males, and Couples by Area.

 

 

 

Area Females Males Subtotal Matched % of
# % # % # % Couples Subtotal
Velezizweni 31 18 21 12 52 30 15 28
Luyengo 39 23 24 14 63 37 15 28
Bhunya 26 17 28 16 54 31 22 42
Total 96 57 73 43 169 100 52 100

 

Characteristics of the Sample

Characteristics of the sample are presented in the tables

below.

Ages of respondents were distributed by areas as shown in

Table 3.

TABLE 3.--Age Distribution by Area.

 

 

 

 

Age Group
Area 20-30 31'45 46+ Subtotal
# % # % # %
Velezizweni 5 10 25 48 22 42 52
Luyengo 8 13 23 36 32 51 63
Bhunya 25 46 26 48 3 6 54
Total 38 22 74 44 57 34 169

 

57

There was a bias towards older ages in the Velezizweni and Luyengo
subsamples, while there was a bias towards younger ages in the
Bhunya subsample.

In analyses using only female and male respondents, ages
were classified into two groups; the age distributions are shown in

Tables 4 and 5. Generally, men in the sample were older than women.

TABLE 4.--Age Distribution of Females by Area.

 

 

 

 

Age Group
Area 20-35 36+ Subtotal
# % # % # %
Velezizweni 12 39 19 61 31 32
Luyengo 14 36 25 64 39 41
Bhunya 20 77 6 23 26 27
Total 46 48 50 52 96 100

 

TABLE 5.--Age Distribution of Males by Area.

 

 

 

Age Group
Area ‘ 20-35 36+ Subtotal
# % # % # %
Velezizweni 2 10 19 90 21 29
Luyengo 5 21 19 79 24 33
Bhunya 13 46 15 54 28 38

 

Total 20 27 53 73 73 100

 

58

The educational levels of respondents are shown in Table 6.
The educational attainments were considerably higher in Bhunya than
Velezizweni. More women than men had terminated with primary and
lower secondary school, while men outnumbered women in higher sec-
ondary school attendance.

The respondents' affiliation with religious denominations
is Shown in Table 7 (page 60).

Table 8 (page 61) shows the number of women who reported
they were legally married or not married. Although 21 women
reported that they were not married, most women in the study lived
in a marriage-like union.

The status of the legally married women is shown in Table 9
(page 61). From those figures, it is evident that a monogamous
family structure prevails in the sample. Thirty-six percent of
women in Velezizweni and 12 percent of women in Luyengo do, however,
live in a polygamous situation.

Women reported presence and absence of husband as shown in
Table 10 (page 62). Husband refers both to legal and nonlegal
partners. That 100 percent of the husbands in Bhunya were present
is, of course, related to the fact that the family would not live
in Bhunya if the husband was not present and employed by the
company.

The marital status of men, distributed by areas, is shown

in Table 11 (page 62).

59

 

 

 

 

 

 

 

 

 

.Feuopaam u mam ”opus u z umpesmm u a
m e m.o up m.m m.w Fe up NN N_ N m NN NF m— N
mop m N F mN m up on mN we FN mp m we FN NN emnssz
pouch
mp PF PN Fm mp as me mp «N e N N m o m ~euoun=m mo N
am N m P NP m m mN op m_ N P P m m N L.3532
eaczgm
N N u- m N o Fe m mm up PF m mm N_ m_ Pmuounzm mo N
mm P _ u- m _ e oN m _N op N m _N PF op L3:52
omcmxzu
-1 1- r- N -1 N mm m_ mN NF o_ N Na mp mN pepopnam eo N
Nm -1 u- r- F -1 P ON NF NF m m e NN N m_ L8:52
wcm3~wNm_m>
azm z m cam z a saw 2 a sum 2 cum 2 a
Pepe» » » mme<
ream eeucoumm concoumm
. cocoa: gaze; xeeewea xueempmg mcoz

 

.emc< can xmm an cmcweuc< cowpeuzum we Fm>m4 ummcmp:--.o mum<p

6O

.mc_euoou maowmvpme cm mwrumempwswm op mcwveooum amazoem mew; mco_umcweocmv mg»-

 

 

 

oOCOZ H &

.memgpo .mmmcupz m.;m>ocma u N

.umwcowN u a

.umwuqem .umpuocumz u u

.uwpogueu cmeom .cmuPpmc< n m

.ceemgu24 gmwnmzm .mcmemNmz .Fecowuez m~m3m .cowmmwz Peemcwu :mowew< spaom .uNpoumoa< u <

mop e m m mp NN we m m_ mp mN mm on Peach
em e N o m ON PF eN NF Pp o mm NF excscm
me 1- -- m_ N. NN N_ m N Nu my on m_ emeeNSS
Nm m e - 11 mm 0N - 11 NP 0 Ne FN wcozNVNmFm>
N N N N N N N N N N N N
Fepop .111111111 --111111 1111111111 1111-11111 1111-11111

-aem a w a u m < mete

 

emanoew maowmwpwm

 

.mme< an cowmwpmm-1.N m4m<~

61

TABLE 8.--Marita1 Status of Women by Area.

 

 

 

 

 

 

 

 

 

Married Not Married Sub-
Area total
# % # %
Velezizweni 22 70 9 30 31
Luyengo 33 85 6 15 39
Bhunya 20 77 6 23 26
Total 75 78 21 22 96
TABLE 9.--Marital Status of Married Women by Area.
Senior Junior Only
Area Wife Wife Wife Other Sub-
-——-————- total
# % # % # % # %
Velezizweni 3 l4 5 23 13 59 l 4 22
Luyengo 3 9 1 3 29 88 -- -- 33
Bhunya -- -- -- -- 20 100 -- -- 20
Total 6 8 6 8 62 83 l l 75

 

62

TABLE lO.--Husband Present/Not Present by Area.

 

 

 

 

 

Present Not Present
Sub-
Area total
# % # %
Velezizweni 25 81 6 19 31
Luyengo 30 77 9 23 39
Bhunya 26 100 -- -- 26
Total 81 84 15 16 96

 

TABLE ll.--Marital Status of Men by Area.

 

 

 

 

Married Not Married
Sub-
Area total
# % # %
Velezizweni 17 81 4 19 21
Luyengo 18 75 6 25 24
Bhunya 27 96 l 4 28

 

Total 62 85 .11 15 73

 

63

Men reported number of wives by the three areas as shown in

Table 12.

TABLE 12.--Number of Wives by Area.

Wife refers both to legal and nonlegal partners.

 

Number of Wives

 

 

 

Sub-
Area 1 4 total
# % % # % # %
Velezizweni 13 76 3 18 1 6 -- -- l7
Luyengo 12 66 4 22 1 6 1 6 18
Bhunya 25 93 2 7 -- -- -- -- 27
Total 50 81 9 l4 2 3 l 2 62

 

Comparing the figures of reported status of married women in

Table 9 and the figures of marital status of men in Table 12, some

disparities are noticeable in the Bhunya subsample.

explained by the fact that women have referred to "only wife in

Bhunya."

Presented in Table 13 is the actual number of children

reported by men and women in the three areas.

TABLE l3.--Number of Children by Area.

It might be

 

Number of Children

 

 

 

Sub-

Area 0-2 6+ total
# % # % # %

Velezizweni 3 6 13 25 36 69 52

Luyengo 9 14 27 43 27 43 63

Bhunya 16 30 31 57 7 13 54

Total 28 17 71 42 70 41 169

 

64

A greater number of children is predominant in Velezizweni
where, however, keeping the age bias of the sample in mind, a
greater number of families can be assumed to be completed.

The mean number of people in households in the three areas

is shown in Table 14.

TABLE l4.--Mean Number of People in Households by Area.

 

 

Area Mean Number in Households
Velezizweni 7.93
Luyengo 8.16
Bhunya 6.13
Grand mean 7.41

 

In spite of a greater number of children reported from Velezizweni,

the household size was greater in Luyengo.

CHAPTER V

FINDINGS

In this chapter, the results of the testing of the hypothe-
ses using Harvey's least squares analysis and Duncan's multiple
range test are reported. A summary of F-ratios for tested relation-
ships is presented in Tables 15 and 16, followed by sections for
each set of hypotheses referring to respective source of variance:
sex, age, desired number of children, and environment. In each of
these sections, a summary table of mean deviations is included.

The chapter ends with a general summary.

In Tables 15 and 16 the F-ratios are summarized for tested

relationships and interaction between sources of variance included

in the particular analysis.

Sex As a Source of Variance
The total sample was used to test the hypotheses:

1: There is no effect of sex on (a) desired number of chil-
dren, (b) ideal number of children, (c) attitude about
fertility control, and (d) knowledge of reproduction.

The null hypotheses 1:c and 1:d were rejected. The magni-
tude of the F-statistic (see Tables 15 and 16) suggested that there
was an effect of sex on attitudes about fertility control and the

knowledge about reproduction. Women scored a higher population

attitude score than men. The figures were 9.57 and 8.21,

65

66

._m>m_ Fo. um mocmqupcum moumowucme

 

 

 

empm.o «momN.m~ NNNN.N mN msoqu
womm.o eemN.m mon.N mN weoum cowumNNFPua
m_¢o.o «momm.m ¢~e_.o mN meoum muwpueea seem
Nom_.o mFmN.N m¢¢F.o mm meoum :omppeuaz
Npmm.— Pmo_.o mwmp.o om «Loom memo u~wgu
NFNN.N eNNpo.m mmom.m om weoum mcmpm»:
eowm¢.¢ mmmm.N omoo.o emmNm.op mop mgoum corpusuoeama
emme.o «¢ONm.m mONw.o «Npmm.o_ mop «Loom unapppam comampsqom
momm.o mmN_.m mpmp.o ompm.o mm_ :mequgu No amass: meum
mmNo.P «Nwme.e NweN.o mFmF.N amp cmevagu No amass: umepmwa
comuueewch “cascoew>cu mm< xmm

z mpnmwcm> pcmucmama

wocw Tpm> $0 mUe—aom

 

.mucewem> mo mmueaom

me Newscocm>cm use .mm< .xmm cpwz mmmxpe=< mmemsam umem4 seem mowumN-m No Newssam-1.mp NAm<p

67

.Fesep Po. ea

muceuwmwcmwm mmueuwccme

 

 

mmo~.o eoeN~.e memm.o mm aeseeeea ee_ea~=aeaemmmweu
QOmNeo ¢©N©op MNFmeo Nm ”LOU"... mmmc—auvaLDOme @PQJOQ
_Nme.o mMNe.P mmpm.P «Nepm.p_ me_ aeeem Nme_eee eeeeeeeeeaem
Nmom.o amome.mP ommm.p asmwe.mp amp eeeem eeeeeuea eeeee_=aea
meme.o eeop.m eooo.o Nmmm._ me, eaeee_ee to Lease: Faeee
w~m_._ memo.e FNFP._ eoom.m mop eeeepeee Le eaee=e eeewaeo
cowuumemucH pcwscoe_>=m :weumwgu xmm
No oz 2 mpaewee> pcmucmamo

 

wucewem> No woezom

 

.mucmwem> No mmueaom we ucmscoew>cm ecu

.cmguprgu mo emaszz umewmmo .xmm saw; mmmxpmc< mmeesam ammo; Eoew mowymm-m we Ngm553m1-.mp N4m<h

68

respectively. Also, the reproduction score was higher for women
than men. The means obtained were 2.98 and 2.50, respectively.
TABLE 17.--Summary of Mean Deviations Based on Sex, for Desired

Number of Children, Ideal Number of Children, Population
Attitude Score, and Reproduction Score.

 

 

 

. Sex

Dependent Variable IRENE

Females Males
Desired number of children 6.32 -O.47 0.47
Ideal number of children 6.80 -0.68 0.68
Population attitude score* 8.89 0.68a -o.68b
(max. 14)
Reproduction score* (max. 7) 2.74 0.24a -o.24b

 

*Refers to significance on .01 level.
a’bThe levels bearing "a" are significantly different from
levels bearing "b."
Age As a Source of Variance
Hypotheses 2, 3, 4 and 5 tested the effect of age on all of
the dependent variables in the study. Findings will be reported in
relation to the hypotheses and based on the different sample sizes
used in the analyses.
The total sample was used to test the hypotheses:
0 2: There is no effect of age on (a) desired number of chil-
dren, (b) ideal number of children, (c) attitude about
fertility control, and (d) knowledge of reproduction.

The null hypotheses were not rejected. The effect of age,

based on three classes, was non-significant.

69

TABLE 18.--Summary of Mean Deviations Based on Age, for Desired
Number of Children, Ideal Number of Children, Population
Attitude Score, and Reproduction Information Score.

 

 

 

Dependent Variable Mggad Age Groups
20-35 31-45 46+
Desired number of children 6.32 0.38 0.15 0.24
Ideal number of children 6.80 -0.16 -O.4l 0.57
Population attitude score (max. 14) 8.89 0.13 0.23 -0.36
Reproduction score (max. 7) 2.74 -0.02 0.03 0.00

 

A subsample of women in the study was used to test the

hypotheses:
Ho 3: There is no effect of age of females on (a) knowledge

of hygiene, (b) knowledge of child care, and (c) knowl-

edge of nutrition.

The null hypotheses were not rejected. Age, based on two
classes, did not have a significant effect on female level of infor-

mation.

TABLE 19.--Summary of Mean Deviations Based on Age of Females on
Hygiene Score, Child Care Score, and Nutrition Score.

 

 

 

Age
Dependent Variable Sgggd
20-35 36+
Hygiene score (max. 8) 4.16 0.30 -0.30
Child care score (max. 8) 5.31 0.05 -0.05

Nutrition score (max. 8) 3.37 -0.05 0.05

 

70

A subsample of males in the study was used to test the
hypothesis:
H0 4: There is no effect of age on males' farm practices.

The null hypothesis was not rejected. The age, based on two
classes, did not have any effect on farm practices.

TABLE 20.--Summary of Mean Deviations Based on Age of Males on Farm
Practice Score.

 

 

 

Age
Dependent Variable Sgggd
20-35 36+
Fann practice score (max. 10) 3.78 -0.15 0.15

 

For the effect of age on family use of community resources
and family income, as reported by males, the following hypotheses

were stated:

H0 5: There is no effect of age on (a) family use of community
resources and (b) income.

The null hypothesis was not rejected.

TABLE 21.--Summary of Mean Deviations Based on Age of Males on
Family Utilization Score and Income.

 

 

 

Age
Dependent Variable agggd
20-35 36+
Utilization score 48.15 2.88 -2.88

Income 4.26 0.49 -0.49

 

71

Number of Children As a Source of Variance

 

Hypotheses 6 and 7 tested the effect of the actual number

of children on fertility goals, standards, and attitudes about fer-

tility control and resourcefulness.

The total sample was used to test the hypotheses.

O 6: There is no effect of actual number of children on

(a) desired number of children, (b) ideal number of
children, (c) attitudes about fertility control, and

(d) knowledge of reproduction.

The null hypotheses were not rejected.

The actual number of

children, based on three classes, did not have any effect on the

desired number of children, ideal number of children, attitudes

about fertility control and knowledge of reproduction.

TABLE 22.--Summary of Mean Deviations Based on Actual Number of
Children for Desired Number of Children, Ideal Number of
Children, Population Attitude Score and Reproduction

 

 

 

Score.

Number of Children
Dependent Variable agggd

0-2 3-5 > 6
Desired number of children 6.32 -0.64 -0.05 0.69
Ideal number of children 6.80 -0.03 0.02 0.00
Population attitude score 8.89 -0.48 -0.05 0.53
Reproduction score 2.74 0.10 -0.18 0.08

 

A subsample of matched couples was used to test

$653

the hypothe-

H 7: There is no effect of actual number of children of

0
about fertility control.

couples on (a) resourcefulness and (b) attitudes

72

The null hypothesis was not rejected. There was no effect
of actual number of children on couples' resourcefulness or atti-
tudes about fertility control.

TABLE 23.--Summary of Mean Deviations Based on Actual Number of

Children of Couples for Resourcefulness Score and Popula-
tion Attitude Score.

 

Number of Children

 

 

Grand
Dependent Variable

Mean < 5 > 5
Resourcefulness score (max. 14) 23.70 -4.15 4.15
Population attitude score (max. 14) 7.71 -2.78 2.78

 

Environment As a Source of Variance
Hypotheses 8, 9, 10, 11 and 12 tested the effect of environ-
ment on all the dependent variables in the study.

The total sample was used to test the hypotheses:

0 8: There is no effect of environment on (a) desired number

of ch11dren, (b) ideal number of children, (c) attitudes

about fertility control, and (d) knowledge of reproduc-

tion.

Hypotheses 8:b and 8:d were not rejected.

The effect of environment on the desired number of children
and attitudes about fertility control was, however, significant. In
the rural area (environment) the mean number of desired chldren was
7.25, in the semi-rural area, 6.83, and in the urban area, 4.85.

The means for the rural and urban area were significantly different
from the semi-rural area. The population attitude score was highest

for the urban area, followed by the rural and semi-rural areas. The

means were 10.04, 8.55, and 8.08, respectively (Table 24).

73

TABLE 24.--Summary of Mean Deviations Based on Environment for
Desired Number of Children, Ideal Number of Children,
Population Attitude Score, and Reproduction Score.

 

 

 

Environment
. Grand

Dependent Variable Mean Rural Semi-Rural Urban
(Velezizweni) (Luyengo) (Bhunya)

Desired number of a a b

children* 6.32 0.96 0.51 -1.47
Ideal number of a ab b

children 6.80 2.45 -0.51 -1.93
Population attitude a a b

score* (max. 14) 8.89 -O.34 -0.81 1.15

Reproductive informa- 2 74 0 23 -0 07 _0 16

tion score (max. 7)

 

*Refers to significance at .01 level.
a, b’ abRefer to Duncan's test. The levels sharing the

same letter are not significantly different.
The female subsample was used to test the hypotheses:
9: There is no effect of environment on (a) knowledge of
hygiene, (b) knowledge of child care, and (c) knowl-
edge of nutrition.
Hypotheses 9:b and 9:c were not rejected.
However, the effect of environment on knowledge of hygiene
was significant. The highest mean score, 4.64, was reported from
the rural area, followed by the semi-rural area, 4.46, while the

urban area had the lowest, 3.37. The mean score for the rural and

semi-rural areas differed significantly from the urban area.

74

TABLE 25.--Summary of Mean Deviations Based on Environment for
Hygiene Score, Child Care Score, and Nutrition Score.

 

 

 

Environment
Grand
Dependent Variable Mean Rural Semi-Rural Urban
(Velezizweni) (Luyengo) (Bhunya)
Hygiene score* a a _ b
(max. 8) 4.16 0.48 0.30 0.79
Child care score
(max. 8) 5.30 -0.07 0.06 0.01
Nutrition score
(max. 8) 3.37 -0048 0310 0038

 

*Refers to significance at .01 level.
a’bRefer to Duncan's test. The levels sharing the same letter

are not significantly different.

The male subsample was used to test the hypothesis:
10: There is no effect of environment on farm practices.

The null hypothesis was rejected.

Farm practices differed significantly between the three
environments. The highest mean score, 5.35, was obtained in the
rural area, followed by the semi-rural area, 3.57, and the urban
area, 2.41. The semi-rural and urban area differed significantly
from the rural area. (See Table 26.)

The male respondents were asked about the use of community
resources by the family, and about the yearly family income to
test the hypotheses:

H0 11: There is no effect of environment on (a) use of commu-
nity resources and (b) income.

The null hypothesis 11:a was not rejected.

75

TABLE 26.--Sumnary of Mean Deviations Based on Environment for
Farm Practice Score.

 

 

 

Environment
. Grand
Dependent Var1able Mean Rural Semi-Rural Urban
(Velezizweni) (Luyengo) (Bhunya)
Farm practice score* 3.78 1.57a _0.20b -1.36b

(max. 10)

 

*Refers to significance at .01 level.
a’bRefer to Duncan's test. The levels sharing the same letter

are not significantly different.

Income, however, differed significantly between the three
social and economic environments. The highest mean level of income
was reported from the urban area, 6.16, followed by the semi-rural
area, 3.45, and the rural area, 3.16. The mean levels from the
rural and semi-rural area differed significantly from the mean
level from the urban area.

In Emalengeni the mean yearly income reported fell in be-
tween 200-440 in the rural area and 10-49 in the semi-rural and
rural area. A bias of under-reporting the income seems likely.
(See Table 27.)

The sample of matched couples was used to test the
hypotheses:

H0 12: There is no effect of environment on couples' . .
égztngsourcefulness and (b) att1tudes about fert111ty

The null hypothesis 12:a was not rejected.

76

TABLE 27.--Summary of Mean Deviations Based on Environment for
Utilization Score and Income.

 

 

 

Environment
. Grand
Dependent Var1ab1e Mean Rural Semi-Rural Urban
(Velezizweni) (Luyengo) (Bhunya)
Utilization score 48.15 5.66 -0.42 -5.23
Income* 4.26 4.09a -o.80a 1.90b

 

*Refers to significance at .01 level.
a’bRefer to Duncan's test. The levels sharing the same letter

are not significantly different.

The effect of environment on couples' attitudes about con-
trol fertility was, on the other hand, significant. The mean scores
for the rural and semi-rural areas differed significantly from the
urban area. The means were 6.40, 6.81, and 9.91, respectively.
TABLE 28.--Summary of Mean Deviations Based on Environment on

Couples' Resourcefulness Score and Population Attitude
Score.

 

 

 

Environment
. Grand
Dependent Var1able Mean Rural Semi-Rural Urban
(Velezizweni) (Luyengo) (Bhunya)
Resourcefulness score
(max. 4]) 23.70 0.93 1.40 1.21
Population attitude 7 7] _] 30a -0 89a 2 20b

score* (max. 14)

 

*Refers to significance at .01 level.

a’bRefer to Duncan's test. The levels sharing the same let-
ter are not significantly different.

77

Summary

The effect of the environment was significant in a number of
relationships tested in the study. The greater F-ratios were
obtained for population attitude score and income, followed by farm
practice score and hygiene score (in Tables 15 and 16).

Sex was found to have a significant effect on attitudes
about control of fertility and knowledge of reproduction, while
non-significant F-ratios were obtained for the effect of age and

number of children on all the dependent variables in the study.

CHAPTER VI

DISCUSSION, CONCLUSIONS, AND IMPLICATIONS

In this chapter, the study will be summarized, limitations
of the study presented, findings discussed in relation to related
research and literature, and implications for home economics pro-

grams, development policy and research stated.

Overview of the Study

The study was undertaken for the purpose of examining rela-
tionships of demographic variables such as sex, age, number of chil-
dren and environment on factors affecting family size decisions in
a developing country. The study focused on such factors as fer-
tility goals, standards, and attitudes about control of fertility,
levels of information, use of community resources and income.

A home-management framework with an ecological emphasis was
employed in the study. The reported desired family size was
regarded as a fertility goal, and the ideal number of children as a
fertility standard. Measuring attitudes about fertility control,
being of importance in directing fertility decisions, was of special
interest. For this purpose a Guttman type scale was developed. In
order to determine such a resource as levels of information related
to fertility decisions, scores were computed from answers on fixed-

alternative questions. Use of community resources was determined

78

79

by a ratio of utilized resources during the previous year, over
existing resources in the community.

Three areas in Swaziland were selected, representing dif-
ferent social and economic environments. Data collection took place
in July-August, 1976. Altogether, 73 men and 96 women were inter-
viewed. The data were collected by students from the College of
Agriculture, University of Botswana and Swaziland.

Least squares analysis was used in order to test relation-

ships.

Limitations of the Study

The limitations of survey research in studying such a com-
plex phenomenon as aspects of family size decisions are recognized.
Data obtained through simple closed-end questions, as used in this
study, fulfill the criterion of objectivity, but leave much to be
desired when it comes to interpretation of, for example, a "yes" or
a "no," "agree" or "disagree." It is hazardous to derive meaning,
implications and predictions from any research findings within the
social sciences, however ambitious the researcher might have been
with validity and reliability testing. It is indeed questionable if
two-way variable relationships ever can be indicators of complex
sociocultural behavior patterns.

It is, further, generally assumed that the meaning of
empirical relationships discovered through statistical manipulation
derive entirely from the data themselves and is not imparted by the

researcher. Findings from a study of interpretation of data from a

80

KAP study in Bangladesh showed the opposite (Studies in Family
Planning, 1976: 322-33). It was found that the analyst's degree
of familiarity with the culture varied directly with his interpre-
tation and conclusions. The Western analysts were found to be con-
cerned more with context free "rates" and objectives and less with
people, while analysts more familiar with the Bangladesh culture
related their analyses to the understanding of problems.

In addition, the cross-cultural setting of this study must
be considered in interpreting research findings. The cross-cultural
setting was, in fact, three dimensional; the research leader using
a second language as a medium of communication throughout the
research process as well as being a foreigner in the cultural set-
ting itself. The latter was probably the most difficult to overcome.
In spite of the study of the Swazi culture and language, the fact
remains that the problems were seen through Western eyes, Western
research methods were applied and Western-biased conclusions and

implications were drawn.

Discussion and Conclusions
The findings will be discussed under the following headings:
studying fertility behavior, and the influence of environment, sex,

age, and number of children, respectively.

Studying Fertility Behavior
Efforts, in trying to understand human fertility behavior,
or rather human fertility control behavior, and how the onset of

fertility decline takes place, have so far given disappointing

81

results. Few predictions have been accurate, attitudes have not
conformed with behavior, identified key-variables have started to
lose influence, to mention some examples. The sudden rise in birth
rates in North America and Europe after World War II, as well as the
subsequent marked decline in the 19605, came as a surprise. The
"baby boom" generation delayed the birth of their children and gen-
erally deCided on less children than their parents. Recently, the
slight drop in birth rates in developing countrieshas come about
without a matching rise in living standards, which often has been
considered as a necessary condition (The Economist, Jan. 8, 1977:
72-73).

Some of these inconsistences in fertility research might
originate in the fact that there are some unique aspects of study-
ing fertility behavior. Firstly, the subject in itself was, until
recently, completely taboo. Secondly, it is difficult to find valid
indications of attitudes and actual behavior, let alone change, in
these areas. Actual number of children or intervals between chil-
dren are not necessarily good indications either of attitudes or
behavior. Thirdly, if motivation and readiness for controlling fer-
tility exist, the means must be socially accepted and conveniently
available. The means, as history has shown, do not necessarily have
to be modern contraceptive devices, but must be perceived and
accepted by the society. Fourthly, it has more often than not been
forgotten that there are always two people involved--only one, the

woman, has generally been studied.

82

Various approaches have lately been used in trying to
explain the mechanism, on the micro level, leading to decline in
fertility and change in family size norms. The significance of a
change from consumption to production aspects of children has been
advocated (Becker, 1960). Similarly, there is the necessity of a
switch of wealth flow to occur; traditionally flowing from children
to parents, but changing direction in the course of social and eco-
nomic deve10pment (Caldwell, 1977). De Tray (1972) found female
opportunity time and female level of education crucial, and Ryder
(1973) and Clay and Zuiches (1977) stressed the importance of
studying fertility behavior in relation to family size norms of the
family of orientation and reference groups.

A family decision-making framework has not yet been employed
to any great extent in the search for explanations and reliable pre-
dictions of fertility behavior. As stated earlier, family decision-
making is a complex mental process, which does not easily lend
itself to investigation. In this study, just a few factors that
influence this process were examined. Of these factors, it was
found that the environment had a significant effect on several
aspects of fertility decisions. These findings are not surprising,
nor unique, but the implications are of extreme importance in plan-

ning development activities.

The Influence of the Environment
Social and economic aspects of family environments have been

found to have significant influence on fertility behavior all over

83

the world (Kiser and Wilson, 1968; Caldwell, 1967; Mauldin, 1953;
Berelson, 1966; Allen, 1973). Generally, findings have indicated
that more traditional fertility values and more negative attitudes
towards control of fertility exist in rural, compared with urban,
settings. The findings in this study support this generalization.
Both the number of desired and ideal number of children were highest
for Velezizweni and lowest for Bhunya, while Luyengo fell in between.
Moreover, the respondents in Bhunya were the most positive towards
control of fertility. However, the respondents in Luyengo were the
least positive with Velezizweni falling in between. Possibly, the
hardship of raising children was felt less in the Luyengo area,
where life is sustained on a mixture of the two extreme economic
systems. The actual number of children was largest in Velezizweni,
closely followed by Luyengo, with a considerable drop for Bhunya.
Interdependence between the environment and the family
decision-making process is also well documented in home management
literature (Deacon and Firebaugh, 1975: 95-101; Paolucci, Hall,
and Axinn, 1977: 29-49). The culture, embracing social and eco-
nomic aspects of the environment, has an indisputable influence on
decision-making. Paolucci et a1. (1977) state that "Culture is an
adaptive mechanism, providing ready-made solutions to the problems
encountered by the group" (p. 47). There is a limitation of choice
in a traditional culture, while decision-making in the course of
social and economic development tends to become more conscious,

goal-oriented, and rational.

84

It can, of course, be argued that it is not the influence of
the environment that has created differences, but rather the per-
sonality traits of people living in different environments. In
studies concerned with the urbanization process, it has, for example,
been found that older, less educated, more traditional, less flexible
people tend to stay in rural areas while young, better educated, more
progressive and achievement-oriented people are more prone to migrate
to urban areas. In Swaziland, this is to some extent also true,
particularly related to age and education. However, in the case of
Bhunya where labor for the industries mostly has been recruited
locally and trained on the job, the personality factor can be
assumed to be minimized.

In order to draw conclusions from the findings in this study,
it would be desirable to know where,<u1the continuum from tradi-
tional Swazi to a complex Westernized society, the three research
areas are located. Obviously, Velezizweni has more components of
Swazi traditional culture than Bhunya.

Generally, objective information about the extent of the
assimilation of Western culture and how families adapt to complex
and accelerating change presently taking place in Swaziland is
almost an unexplored field. Comparatively more is known about the
traditional Swazi life through anthropological and ethnological
studies carried out in 1930-40 (Englebrecht, 1930; Marwick, 1940;
Kuper, 1947, 1963, 1965).

In 1940, Marwick wrote:

85

Most impressive is the fact that Swazi culture remains a
working whole, in spite of the influence of European civi-
lization. Many elements of European culture have been.
assimilated, e.g., the use of the plow and of blankets,
but the whole synthesis of essentially Swazi culture is
still preserved. Sanctions and institutions which existed
prior to European influence still operate even where
attempts have been made to uproot certain elements of cul-
ture. Examples are frequently of Christian converts who,
as soon as some trouble befalls them, hasten to consult
the witchdoctors.

This led me to the realization that Swazi culture is
still very real, and that it is closely woven into the
fabric of the people's lives and maintained there by their
own institutions (p. xiii).

Since then much has happened, but to some extent Marwick's
observations still hold. There has been, particularly since Inde-
pendence, an outspoken desire to cherish Swazi traditions, beliefs
and customs. The fact that Swazi culture has been more resistant
to outside influence, qumalo (1976) attributes to circumstances
such as Swaziland being a unified and integrated nation before the
imposition of British rule, in contrast to most other modern Afri-
can countries. A common language was spoken, a common culture
existed, and the country owed allegiance to a single king.

Nevertheless, erosions of Swazi culture are noticeable,
particularly among people living in towns or those otherwise
affected by modern economy such as employees of the large rural
industrial companies, of which Bhunya is one example. This is par-
ticularly true for the second generation growing up in this environ-
ment, which is radically different from traditional patterns of
Swazi life. Maybe the introduction of Western education first
brought by the missionaries and later controlled and expanded by the

British has in itself been the greatest agent of change. Other

86

factors which have accelerated change in Swaziland, as summarized by
qumalo (1971), are the imposition of taxes combined with the effect
of the division of land which forced the rural adult Swazi male into
various types of paid employment. Even if the homes were left, only
temporarily, for employment in mines, farms, or in industries
developing in urban areas abroad or in Swaziland, the workers brought
home impressions from a totally different world which doubtless has
had its impact on cultural change.

Being unable to determine exactly where in the transforma-
tion process the three communities are, it is still safe to draw the
conclusion that the environments of the three selected research
areas differ considerably. Most obvious is the difference in physi-
cal environment, particularly between Velezizweni and Bhunya. There
is a sharp contrast between the spacious hills of Velezizweni, with
each family commonly residing in several huts, and the restricted
yards of Bhunya that surround the small housing units. This is
bound to have a significant influence on family size standards and
attitudes about control of fertility. Nevertheless, the dual living
pattern of families in Bhunya should be kept in mind. If too much
crowding is experienced, it is most likely that a part of the
family will be sent off to the rural home.

Social and economic aspects of the three environments are
equally contrasting. A summary of some of the community resources
available in the three areas is presented in Table 29. For a
verbal description, see Chapter IV, pages 42-54, The majority of

families in Velezizweni live at a subsistence level, while Bhunya

87

TABLE 29.--Summary of Available Community Resources by Area.

 

Resource Velezizweni Luyengo Bhunya

 

Water supply River Irnigation Bays
urrow

Grocery store X X
Primary school X X
Home economics extension X
Nursery school

Lower secondary

Clinic

Family planning services

Outdoor market

Post office

><><><><><><><
XXXXXXXXXX

Beer hall

Agriculture extension X

X
Boys only

Electricity X

Higher secondary

Sport field x

Cinema X

 

was created for the very purpose of housing wage earners at the pulp
mill. Although the Luyengo settlement was once planned and built by
outside authorities, the community can now be assumed to be devel-
oping at its own pace and direction, resulting in a blend of tra-

ditional as well as modern patterns of living.

88

The exposure of new ideas and a new life style is inevitable
in Bhunya, regarding such community resources as frequent communi-
cations with Mbabane and Manzini, film shows, sports and music fes-
tivals, availability of daily newspapers, better educational
facilities for both children and adults, and not the least of spe-
cial interest in this study, family planning services with slightly
different features than just the "clinic approach." The daily con-
tact with expatriates, as well as Swazis with training or work
experience commonly from the Republic of South Africa, is also
likely to have great impact.

Although the well-planned community of Bhunya offers plenty
in terms of community resources, the degree of utilization was found
to be lower than for Velezizweni and Luyengo. The findings might be
attributed to the well-known supply-and-demand relationship. It
might also result from a feeling of non-belonging in this community
that was created by representatives from a different culture with
very different premises of living.

Also unexpected are the low levels of information in Bhunya.
In spite of the most active family planning approach, the level of
reproductive information was the lowest. It is only in the level
of nutrition information that Bhunya scored higher than the other
communities. In line with a theory of information overload, infor-
mation might be blocked out in this more complex environment.

Regarding the low level of reproductive information in
Bhunya, one interpretation is that the traditional system of sex

education has ceased to function in Bhunya. No other institution,

89

certainly not the church, which rather has discouraged spread of
such information, nor the school, have taken over this function in
the urban society. In Velezizweni, on the other hand, having the
highest level of reproductive information, parents and grandparents
still seem to carry out their traditional roles as sex educators
with some success.

Comparing community resources between Velezizweni and
Luyengo, one finds a rather equal distribution. In the push for
rural development, Velezizweni, being centrally located and fairly
close to Mankhyane, one of the major towns in the country and cen-
ter for an integrated rural development project, cannot be consid-
ered a forgotten, backward rural area. For example, agricultural
extension Services exist on a regular basis in Velezizweni but not
in Luyengo, in spite of farming activities, even on a cash crop
basis. In the case of home economics extension activities, the
situation is the same, although a women's group has recently been
started in the Luyengo settlement. Levels of information are,
nevertheless, lower for Velezizweni, with the exception of repro-
ductive information, hygiene information and farm practices.
Regarding the influence of modernization and the degree of social
and economic development, however, there is doubtless a difference
between the two communities. Being located close to one of the
main roads in one of the economic "core regions" in the country, the
population of Luyengo is more frequently exposed to new influences

and has incorporated more of them into their own lives.

90

The near social family environment is equally different,
particularly comparing Velezizweni and Luyengo on the one hand, and
Bhunya on the other (see Tables 8 and 11). The degree of nucleation
of the family structure is obviously higher in Bhunya. This is to a
certain degree a forced process, related to housing regulations by
the company and can, nevertheless, be assumed to have an impact on

the relationships here examined.

The Influence of Sex

It was felt necessary to interview both men and women.
First of all, information from men about family size decisions is
rare, and also for the reason that differences in goals, standards,
and attitudes were anticipated. However, only for two variables
reflected significant differences between men and women were
recorded: population attitude score and level of reproductive infor-
mation. Women scored higher in both cases. Men desired more chil-
dren and considered a greater number of children ideal for a Swazi
family, although the finding was not significant.

Women's motivation for controlling fertility is in line
with findings from several other studies (Mauldin, 1953: Allen,
1953). Increasingly, Swazi women are experiencing the hardship of
raising children, often being left alone with this responsibility.
Even if the majority of women in the study were living together
with a man at the time of the study, many unions are of a temporary

nature.

91

Among men, polygamy is still the ideal family structure,
although legalized polygamous unions are decreasing (Ministry of
Local Administration, 1970). The influence from the Church as well
as economic pressures have contributed to this development. In
urban areas, particularly where family support is dependent on
wages, which hardly allow for supporting one wife and her children
as well as an increased cost of living, the polygamous family cannot
survive. However, still guided by traditional values, it is not
unusual that a man when getting tired of his wife will pick another
female partner. Due to the impossibility of supporting them both,
the first wife is usually left on her own. This is not imcompatible
with Swazi law and custom, but there is no legislation implemented
to protect women in this sense. Instead of being a respected senior
wife, she will have to look for a job outside the home in order to
support herself and her children.

Among educated women with possibilities of supporting them-
selves and their children, there seems to be a novel attitude
against payment of dowry and traditional marriage bonds. Without
payment of dowry, the children belong to the mother and her family,
and the risk is not taken of having to support not only the husband,
but also additional wives. There is doubtless a growing resentment
among women about their increasingly unfair and insecure situation,
accentuated by economic and social changes.

African men have often been accused of being lazy, and
doubtless they have always had a privileged position. The fact that

men lost their important roles as warriors and hunters a long time

92

ago provides one explanation. An unequal distribution of duties
and responsibilities is often witnessed between Swazi men and women.
The majority of women have gained little from the development that
has taken place. Their situation has rather become more demanding,
particularly for those having to combine work outside the home, on
the one hand, and child rearing and other household duties on the
other.

A significant pooled interaction between sex, age, and
social and economic environments was recorded in relation to level
of reproductive information. Unfortunately, the analysis does not
provide information to draw conclusions of between which variables
there was an interaction and the direction of these interactions.
Because the effect of sex in itself was significant this would sug-
gest that in spite of interaction within the levels of age and

environment, clear differences based on sex were apparent.

The Influence of Age
No significant effect of age was found on any variable in
the study. In a fast-changing society, one would expect that younger
people would more easily adopt new norms, and also, being better
educated, have higher levels of information. This, however, was
not noted.
The Influence of Actual
Number of Children
Using the actual number of children as a source of variance,

no significant effect on desired or ideal number of children, nor on

93

population attitude score and levels of reproductive information,
was found. These findings hold for both men and women and matched
couples. Thus, the conclusion can be drawn that decision-making
about family size is not yet rational nor goal-oriented. However,
even if relationships are weak, having fewer children is related to
a smaller family norm and more positive attitudes about control of
fertility.

Although the traditional Swazi culture could be described
as a maximum fertility culture, there exist concepts of spacing
children. As in many other polygamous African cultures, it was a
custom not to have another child until the first-born was weaned,
usually resulting in a two-year interval between children. It is
believed that if the mother became pregnant again before the child
was weaned, her milk would become poisonous. This norm might also
originate from the times of frequent warfare when villages suddenly
had to be vacated. Women Should then have only one child that
needed their full attention (Marwick, 1966: 33).

Traditionally, women also stopped bearing children when
their first child reached the age of puberty. Moreover, the limita-
tion of one son for the woman designated to become the future Queen
Mother, to reign the country together with the son, is another tra-
ditional concept.

The major conclusions from findings in this study are that
traditional fertility values tend to be prevalent in less advanced
stages of social and economic development. Women were more motiva-

ted than men to control fertility, although men and women differed

94

slightly in the number of children they desired or considered ideal.
The level of reproductive information decreases in the transforma-
tion to a modern society. Age was not a determining factor on
attitudes about fertility control, nor fertility goals, standards or
levels of information. Finally, since the actual number of children
did not influence fertility goals, standards and attitudes about
fertility control, it was concluded that fertility behavior in the

three investigated communities was on the whole not goal-oriented.

Implications for Home Economics Programs

Based on the reported association between environments and
attitudes toward fertility, the overall implication for home econ-
omics program development, based on findings in this study, is the
need for program activities to be more directed towards changing the
environment than delivering information to individuals. This would
be facilitated through close coordination with other development
activities as well as more active involvement in community affairs.

Another means of initiating change in the environment, which
home economics programs can support, is the creation of job oppor-
tunities, particularly increased participation of women in paid
employment or self-employment. Promotion of increased agricultural
production being the backbone of economic development, involvement
in related activities should be of special concern.

Both men and women should be involved in home economics pro-
gram activities. Many of these activities are not, as often

believed, related only to women's roles. To prepare, particularly

95

youth, for a modification of sex roles in a more pluralistic society
is of great importance.

Including population education components in home economics
programs would have implications for fertility decisions. The low
level of reproductive information as well as unfavorable attitudes
towards control of fertility among men found in this study point to
the necessity to direct these programs also to men.

However, subject content of home economics programs must
focus on family goals competing with the goal of having a large
family. Family health, particularly child health, is such a goal.
Making mothers and fathers aware of the needs of children to grow to
healthy, productive individuals may be an effective means of deliver-
ing family planning messages as well as changing family size norms.

Educating children is another family goal of great priority
in Swaziland, which home economics programs should capitalize on in
order to create a desire to limit the number of children in favor of
a stronger emphasis on human resource development. In stressing
the importance of education, home economists must also concern them-
selves with the content of the school curriculum and methods of
teaching, contributing to the relevance of home economics teaching
as well as other curricula taught.

Home improvement is another home economics subject area
which might be attended to with implication for environmental
changes. Improving the home or building a new home are family goals

often expressed by Swazi families.

96

All home economics subject areas lend themselves to the
training of decision-making skills, which should be given high
priority. As findings in this study indicate, there is a need for
home economists to serve as a link, and information processor,
between the community and the family, when complexity increases in
the environment. This can be facilitated through formal and non-
formal educational channels. An important function should be to
provide information and feedback of already existing environmental
support systems in order to maximize utilization of these services

and ensure that needs of families are met.

Implications for Development Policy

The strong environmental influences on factors affecting
fertility decisions found in this study support the argument that
improved living conditions accelerate decline in fertility and
development toward a smaller family size norm (genes, Nov.-Dec.,
1973). The position taken by earlier advocates of the theory of
the demographic transition that urbanization and industrialization
are pre-conditions for fertility decline, runs, however, against
sound development policy. First, the social problems and human
misery, particularly in connection with too rapid urbanization, are
too well witnessed in the Third World today and need not be elabor-
ated on any further. Second, there are examples of nations where
fertility has declined without massive urbanization and industri-
alization. The People's Republic of China provides an outstanding

example.

97

For these reasons, and based on findings from this study,
forceful rural development policies, changing the family environment
and raising levels of living in high-fertility areas are imperative
for the evolution of a smaller family norm. The findings also
point to the fact that semi-rural and semi-urban societies, where
high fertility values often linger, should not be neglected. Often
belonging to "no one's land" in terms of development programs, as is
the case in Swaziland, these areas are left behind.

With 90 percent of its population living in rural areas,
rural development programs do occupy a central position in Swazi-
land's development strategy (Second National Development Plan, 1973-
77: 15-26). A family-oriented program with special emphasis on
population education, "Planning for Better Family Living," has also
been approved and preparatory activities started (Government of
Swaziland, Ministry of Health, 1976). Nevertheless, Swaziland
still lacks a population policy per se, which would state objectives
of altering or changing existing population trends or character-
istics.

It is also questionable if the magnitude of these efforts
is large enough to have an impact on lowering fertility, to the
extent that would be desirable. The allocation of resources is,
doubtless, still more favorable for urban areas.

Establishment of industrial communities in rural areas is
another means of changing social and economic environments without
massive urbanization. The creation of Bhunya is one example.

Findings in this study indicated that change in fertility behavior

98

has taken place in this new environment. This has happened in less
than 25 years, which must be considered a very short time in rela-
tion to changing attitudes and behavior. If, in addition, genuine
emphasis on creating sources of cash income for women was instituted,
the magnitude of the change might have been greater still.

Bhunya can, in fact, be looked upon as an integrated devel-
opment program on a large scale. However, created with private
capital and having private profit interests as the ultimate goal,
there are some less desirable aspects of this establishment. Some
of these are the low wage level created by a static labor situation
and ample supply of labor, the lack of participation of workers in
the management process and, generally, the company's paternalistic
way in handling matters.

Regarding relationships between goals, standards, and
attitudes, the findings in this study indicated a trend in Bhunya
towards a more rational and goal-oriented fertility behavior. These
findings fall in line with previous assumptions of increased ration-
ality in the course of social and economic development. Neverthe-
less, it can be argued that fertility decisions in societies of
every type and stage of development, in fact, are rational. It all
depends on what kind of rationality the discussion is based. If it
is economic rationality, which often is the case, it then depends
on how familial and societal goals are transferred into money terms.

To identify these values must be the task of individual
families. The development worker, whether a home economist, agri-

culturist, health or community development worker, on the other

99

hand, should assist families with bringing decision-making processes
to the level of awareness, particularly stressing the relationship
between family goals and resources.

In an increasingly complex environment, decision-making
skills will be of the greatest importance. Any development program
needs to incorporate micro decision-making. It can be facilitated
through decentralization of political decision-making, through
establishment of cooperative societies for different purposes,
through consumer education, and farm and home management informa-
tion. This means an active participation in the development process
at all levels.

To summarize, development activities leading to an overall
decline in fertility must first of all concentrate on reshaping the
environment in rural areas; making rural living more profitable,
more attractive and comfortable. Much greater proportions of
resources need to be allocated to these areas to create such a
change. New job opportunities must be created to raise income
levels, through agricultural development programs and through the
establishment of rural industries. In general, development efforts
must be integrated, incorporating improved formal and nonformal
education, health and family planning services, less guided by
Western tradition but building on cultural and economic realities.
Of special importance is the training of decision-making skills,
at all levels, bringing the family decision-making process to the
level of awareness, resulting in rational and conscious decisions

about family size.

100

Implications for Further Research

Beginning 15-20 years ago, when rapid population growth
became a concern, valuable time was lost in trying to understand
global population trends, while research on a family level was
neglected. This study attempted to explore factors affecting fer-
tility decisions, involving both men and women. Nevertheless, the
data gathered can only begin to integrate information about the
effects of selected demographic variables and environmental influ-
ences.

Only a part of the data that were collected have been
reported in this dissertation. For further managerial research,
secondary analyses could be performed. Of special interest would be
to carry out factorial and item analyses on data recorded on atti-
tude statements in relation to Specific variables. Concerning data
collected from the matched couples, it might be more fruitful to
aply correlation analyses of husband's and wive's answers rather
than the analysis used in the study.

The population attitude scale, proven to be of acceptable
reliability and validity, could be used in studying effect of popula-
tion education, such as campaigns and/or audio-visual materials.

In order to study interrelationships within the family system
as well as between the family and the environment, an in-depth study
with stronger social anthropological emphasis, over a period of
time, would need to be undertaken. Preferably a few families could

be selected from each of the three research areas used in this

101

study, and studied over a period of time. This would contribute to
a family ecological perspective. An interesting dimension would be

added if families of different stages in the family life cycle would

be selected.

APPENDICES

102

APPENDIX A

COMPLETE QUESTIONNAIRES

103

APPENDIX A
COMPLETE QUESTIONNAIRES
Pages marked with "h" were information gathered only from
husbands, while pages marked with “w" were information gathered

only from wives. In the case of pages marked both "h'l and "w,"

questions were addressed both to husband and wife.

104

105

Name of enumerator:

 

Date of interview:

 

QUESTIONNAIRE A

Questions to be asked of all wives

Could you please start with giving some information about yourself.

Area:

 

Name:

 

Name of head of household:

 

Relationship to head of household:

Husband present:

Age:
Occupation:

No. of children:

No. of children living with

you other than your own:

Education (highest level completed):

Religion:

U‘I-hWN-J
I O O

—-l
C

Senior wife
Junior wife
Only wife
Other, specify
Widow

Yes
No

 

 

 

 

O'tU'l-th-l

None

Literacy

Primary

Lower Secondary
Higher Secondary
Other, specify

 

106

Name of enumerator:

 

Date of interview:

 

QUESTIONNAIRE B

Questions to be asked of husbands/head of households

Could you please start with giving some information about yourself.

Area:

 

Name:

 

(Age:
Occupation:

No. of children:

Education (highest level completed):

Religion:
Religion:

How long have you lived in this area?
From where did you move?

15 the house your family is living in
owned or rented by you?

 

 

 

GmkwN-d
o 0

None

Literacy

Primary

Lower Secondary
Higher Secondary
Other, specify

 

 

 

Owned

Rented rent/
month

Other

107

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.ngzzou

macaw .ucep =.N a P No come

1P~mzm ouwm mpmsmw ..m.m .mem553:

.oc so 1uao econ NH com emcee mcwaewumam

mm» .cowp .puwep N .upocmmao; on»

NmemszmmFm New: saw; 1eosno mewp 1mwo mumpm memo» .mpes No mamasms emspo

mcw>w_ 3o: mcw>pp 3o: 1PP=N new .ucmpp~mzm umpmpasou com ace one; on upcm

ago News 30: men News 3o: 1ucmupm 302 :N econ NH :H z -:owpepme mpmum
neon cmwmwwszMHNNHo m;m\w; mH moepacuewm mm< xmm awcmcowgmpmm meez

 

zpwsew Lao» cw memnEme on» mcwmmzumpn saw: pemum m.pmp .zoz

mzmzuu ogozmmao:
gp

 

108

2 h V
VELEZIZWENI

Utilization of Environmental Resources

Environmental Resources Checklist for each family

 

Have Have Household Would
never heard member use but
heard but has used, cost is
Resources of it never attended, expensive
(here) attended or
consulted
How often

 

Hospital, Mankayane

Sangoma

Inyanga

Primary School

Velezizweni Primary,
S.A.G.M.

Junior Secondary School,
Nkabanene

Senior Secondary School,
Mankayane Central

Adult Literacy Class,
near Mankayane

DS. Demonstrator

Agriculture Extension

Sunday School, S.A.G.M.

Farmer's Association

Farmer's Coop. Society

Communal Garden

Churches (Zionist,
Methodist, Apostolic
Church, Roman Catholic)

Bus Service

Grocery Store

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

109

2 h L
LUYENGO

Utilization of Environmental Resources

 

Resources

Have

never
heard
of it
(here)

Have
heard
but
never
attended

Household
member
has used,
attended,
or
consulted

How often

Would

use but
cost is
expensive
or other
reason

 

Clinic, where?

 

Inyanga, where?

 

Sangoma, where?

 

Nursery School
where?

 

Primary School
where?

 

High School
where?

 

Adult Literacy class,
where?

 

Creche, where?

 

Sunday School
where?

 

Churches (Zionist,
Methodist, Episcopal,
Roman Catholic, other)

Shops:
Grocery, where?

 

Butchery, where?

 

Bus Service
Taxi Service
Beer Hall

Sports Field

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

110

2 h B
BHUNYA

Utilization of Environmental Resources
The following questions are about the help you get from the community.

 

Have Have Household Would
never heard member use but
heard but has used, cost is
Resources of it never attended, expensive
(here) attended or consulted or other
How often reason

 

Clinic, where?
Inyanga, where?
Sangoma, where?
Schools:
Creche, where?
Sunday School
Primary School
Secondary School
Zondle feeding scheme
School committee
Adult Literacy Class,
where?

Domestic Science Class,
where?

Red Cross Club,
where?

Churches (Zionist,
Methodist, Anglican,
Nazarene, Other)

Shops:

Grocery, where?
Butchery
Vegetable market
Dry cleaning
Restaurant
Beer Hall
Post Office
Bank, where?

Town Clerk Office

Club House
Are you a member?

Town Hall
What activity?

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

111

Utilization of Environmental Resources, 2 h B, Bhunya, continued.

 

Have Have Household Would
never heard member use but
Resources heard but has used, cost is
of it never attended, expensive
(here) attended or or other
consulted reason
How often

 

Stadium
What activity?

 

Sports field
What activity?

 

Tennis courts

 

Nursery

 

Poultry scheme

 

Facilities in Mhlembanyathi,
What kinds?

 

Bus Service

 

Taxi

 

 

112

Farm Practices

 

Now, could you please give us some information about your

farm.
1. Do you grow more than 4 crops other than maize? 1. yes
2. no
2. Did you produce more than 18 bags of maize in 1974/75? 1. yes
2. no
3. Did you produce a surplus over your requirements in 1. yes
1974/75? 2. no
4. Do you own the three basic farm implements: 1. yes
mouldboard plough, planter, and cultivator? 2. no
5. Do you own oxen? 1. yes
2. no
6. Do you use hybrid maize seed? 1. yes
2. no
7. 00 you use a grain tank? 1. yes
2. no
8. Is your storage capacity more than 18 bags? 1. yes
2. no
9. Do you chemically treat maize? 1. yes
2. no
10. 00 you sell maize? 1. yes
2. no

We are also interested in your yearly cash family income.
This is just for our use.

Cash Income

What was approximately the cash family income in Emalangeni,
last year?

1. 0.00- -- __ 5. 200.00- 440.00 _
2 1.00- 9.00 _ 7. 450.00- 599.00 __
3 10.00-49.00 __ a. 700.00- 999.00 __
4. 50.00-99.00 __ 9. 1,000.00-1,999.00 _
5 100.00-199.00 10. 2,000.00-over

113

p..-
3N

Family Size

1.

9.

Now, I would like to discuss factors which concern the size
of your family.

If you could have exactly the number of
children you want, what would that number
be?

Would you like to have more children?

If yes, why?
If no, why?

Would you like to have less children than
you have?

If yes, why?
If no, why?

What do you think would be the ideal number
of children for Swazi families, if they
could control the number as they wish?

 

 

 

 

___No. of children

1.
2.

yes
no

yes
no

___No. of children

 

 

 

 

 

 

 

 

 

 

 

 

 

Have you ever discussed the number of 1. yes

children you want with anyone? 2. no

If yes, with whom?

(Wives only)

What are the names of your children?

1. 7.

2. 8.

3. 9.

4, 10.

5. ll.

6. 12.

When was your last child born?

15 the child still alive? 1. yes
2. no

Did any of your children die? 1. yes
2. no

If yes, how many?

Suppose you have a very close friend in the same circumstances
as yourself, and she asked you for advice on the convenient
number for her. What is the number of children you would advise

her to have, if she could?

___No. of children

Marriage

Now, I would like to have your opinion about marriage.

10.

11.

Do you think a man should have more than

one wife?
If yes, why?

mm
3'2

114

 

If no, why?

 

Are you married?

If yes, how were you married?

Which marriage ceremony would you like for

your children?

Do you think a couple can consider them-
selves married if "Lobola" has not been

paid?
(Wives only)

Did your husband's parents pay Lobola

for you?

Do you think it is right for a young
couple to have sexual relationship before
they are married?

If yes, why?

(B)

 

If no, why?

 

Do you think it is better for a girl to
become pregnant before she is married?

If yes, why?

 

If no, why?

 

How old do you think on the average, that
a girl and boy should be getting married?

00 you think it is something strange when
people have been married for a couple of
years without having any children?

If yes, why?

 

If no, why?

 

1. yes '
2. n0
1. yes
2. no
1. civil ceremony
2. traditional
3. Christian
1. civil ceremony
2. traditional
3. Christian
1. yes
2. no
1. yes
2. no
1. yes
2. no
1. yes
2. no
a boy
a girl
1. yes
2. no

What do you think about a couple who have a baby every year?

What do you think about grandmothers bringing up children?

115

0100
3'2

Family Goals

What do you think about your family and the future?

00 you believe that actions you and other family
members take today will influence the future of
your family?

00 you sometimes discuss the future with other
family members?

00 you think families nowadays are trying to plan
more for the future than in olden days?

00 you think that what you are going to do in life
is predetermined?

Have you at any time discussed with other family
members anything Specific you wanted to achieve when
it comes to the following?

a. Children's education
If yes, with whom did you discuss?

 

What were the goals agreed upon?

 

What did you do to try to achieve them?

 

b. Moving to another place?
If yes, with whom did you discuss?

 

What were the goals agreed upon?

 

What did you do to try to achieve them?;

 

c. Introducing new agricultural methods
If yes, with whom did you discuss?

 

What were the goals agreed upon?

 

d. Extending or building a new house
If yes, with whom did you discuss?

 

What were the goals agreedTupon?‘

 

What did you do to try to achieve them?

 

N—l

e—n N—l N—l N—'
O O O O C

d
.

—-l
.

yes
no

yes
no

yes
no

yes
no

yes
no

yes
no

yes
no

yes
no

116

4 w
7 h

Family Goals, continued.

 

e. The number of children in your family
If yes, with whom did you discuss?

 

What were the goals agreed upon?

 

What did you do to try to achieve them?

 

f. Marriage of your children
If yes, with whom did you discuss?

 

What were the goals agreed’upon?

 

What did you do to try to achieve them?

 

9. Family problems
If yes, with whom did you discuss?

 

 

What were the goals agreed upon?

 

What did you dfiito try to achieve them?

 

Is there anything in life that you would very much like
to achieve and you believe that you probably could?

If yes, what is that?

 

What means would you use to reach that goal?

 

What is most important for you in life?

 

What would you do first if you suddenly got a lot of
money? Select three and rank in order:
send all children to school
buy cattle
improve and extend your home
buy a refrigerator
buy a stove
buy new clothes
buy a new car
buy a tractor
buy imputs for the farm
start a business
offer to ancestors
don't know

—I

d
.

u—l

. yes
. no

yes

.110

. yes
. no

yes

.110

 

117

N01
3'2

Population Attitude Scale

Now we will read some statements that people have made, and
we want to know if you agree or disagee.

Agree Disagree

l. Most mothers want fewer children because
men fail to support them nowadays.

2. There is no limit to the number of children
a person should have.

3. It would be desirable to be able to decide
how many children one gets.

4. Most mothers want fewer children, because
then each can get the best in life.

5. A woman should have many children in case
some die.

6. One cannot decide how many children one
would like to have.

7. More children should be born in Swaziland.

8. It is better to decide on the number of
children one wants, because by doing so
one will be able to manage them properly.

9. In order to decide how many children one
wants, husband and wife must cooperate.

10. Nowadays there are means to decide how
many children one wants.

ll. Parents should have as many children as
they like.

l2. There is no way to prevent pregnancy.

l3. In the next generation no woman should
have more than five children.

l4. Most women want to have a lot of children.

118

5 w
7 h

Population Attitude Scale, continued. Agree Disagree

Additional Population Statements

15. It is better for an unmarried woman to have
an abortion than bear an unwanted child.

16. All young people should be educated in means
to avoid pregnancies.

17. I believe that a man cannot feel fulfilled
if he has never fathered a child.

18. No woman should be denied the right to use
means to avoid pregnancies when she does
not want children.

19. Availability of effective means to avoid
pregnancies will lead to an increase in
promiscuity.

20. If the population increases rapidly in a
poor country, there will be less possi-
bility for economic development.

21. To limit the number of children born should
not be a concern, because there will be
enough resources for all children born in
this world.

22. I believe that having children is not
necessarily the best way for couples to
enrich their marriage.

 

001
3S

Family Expenditures

Have you ever thought about the expenses you have in your
family--your expenses daily, weekly, and yearly? We are interested
in knowing if your income is covering your expenses, and if you
still are able to save.

 

001
:72

Family Expenditures, continued.

1. Approximately how much do you spend on the following items?
How often do you have this expense?

 

Expenditure Value/time unit Value/year

 

Groceries

Alcoholic drinks

Tobacco

Textiles, clothing

Transportation

Milling

Taxes

Rent

Cash savings

Seeds

Hired labor

School fees, textbooks,
uniforms, health

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Other

2. Do you sometimes buy on credit? 1. yes
If yes, what? 2' no

3. Do you sometimes borrow money to manage expenses? 1. yes
If yes, what kind of expenses? 2’ no

4. Do you sometimes get help from relatives with 1. yes
expenses? 2. no
If yes, what expenses?

5. Have you received any substantial gifts during 1. yes
the last year? 2. no
If yes, what or how much?

6. Have you saved any cash money during the last 1. yes

year? 2. no
If yes, where do you keep the money?
___jn the house ___jn the bank ___other place

120

7 w

Hygiene

We are concerned about the health of your family, so could
you please give us some information about the following?

 

1. Do you boil your drinking water? 1. Never
2. Sometimes
* 3. Almost always
2. Do you think one should boil the * 1. Yes
drinking water? 2. No
If yes, why?
3. Do you boil milk before giving it to * 1. Yes
the children to drink? 2. No
* 3. Using commercial
milk

4. How do you store food?

 

In what

. ?
conta1ner? Where

Food

 

Sugar
Bread
Mealiemeal
Beans
Potatoes
Vegetables
Onions
Porridge
Eggs
Water

 

 

 

 

 

 

 

 

 

 

 

5. How do you wash cooking and serving dishes?
1. Cold water in a basin
* 2. Heated water in a basin
3. Under running water from a tap

6. Are you using any soap when washing * 1. Yes
dishes and utensils? 2. No
7. Do you sometimes try to protect your * 1. Yes
food from flies? 2. No

If yes, why?
How?

8. What do you think is the best treatment 1. Injections
to decrease the frequency of diarrhea? * 2. Cleanliness
3. Local medicine

 

 

 

121

Hygiene, continued

. Every day

. Twice a week
Once a week
Twice a month
. Once a month
Other, specify

Every day

. Twice a week
Once a week
Twice a month
. Once a month

. Other, specify

. Every day
Twice a week
Once a week

. Twice a month
Once a month
Other, specify

. Every day
Twice a week

. Once a week

. Twice a month
Other, specify

Every month

Four times a year
Twice a year

. Once a year
Other, specify

Several times a day
Once a day

. Once a month

Other, Specify

Several times a day
. Once a day

Once a week

Once a month

Other, specify

9. How often do your children wash
their bodies?

10. How often do you take a bath?

11. How often do your children put on
clean clothes?

12. How often do you yourself put on
clean clothes?

13. How often do you wash your blankets?

14. How often do you clean the floor in
the kitchen?

15. How often do you clean the floor
where you sleep?

U'i-hMN-i :wa-J 014:de m-wa—a mmwa—I 0101-5de mmth-a
O O O O O O I I O D

Employment/Special Skills

The next questions deal with employment and the help you get
in household work.

1. Do you receive cash money for any 1. Yes
work you are doing? 2. No

122

7 w

Employment/Special Skills, continued.

If yes, what type of work?
If selling own product, what type of

product?

1. Selling own product
2. Employed

 

How much did you earn last year?

Family Service

 

Do you get any help with the following from family members
who do not have meals together with you and people living outside

the household?
If so, how often?

 

 

Help with From family members From people living
who don't have meals outside the house-
together with you hold

D W M Y D W M Y

 

Household work
Agricultural work
Herding cattle

Finding employment
Caring for children
House, building/maint.
Arranging social events

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Reproduction

CDO
3'£

That children are born is important for families--1et's dis-

cuss that for a while.

1. How long does it take for a baby to develop 1. Five months
inside the mother before being born? (22 weeks)
* 2. Nine months
(40 weeks)
3. One year
(52 weeks)
2. If the mother is in a general state of poor * 1. Yes
health, does that affect the baby before it 2. No

is born?

123

(DO
3'2

Reproduction, continued.

3. How soon after giving birth can a 1. Immediately
woman become pregnant again? * 2. When the uterus has
stopped bleeding
3. When she has stopped
nursing the baby

4. If the mother becomes pregnant while 1. The baby will be

she is still nursing the last born poisoned
baby, what will happen? 2. The fetus will be
aborted
* 3. The breast milk will
disappear

5. Have you heard of any way to avoid * 1. Yes
having children too often? 2. No

If yes, what?

Have you used it yourself? 1. Yes
2. No
If no, would you like to use something 1. Yes
to avoid having children too often? 2. No
6. If only girls are born to a mother, 1. The mother
what is that depending upon? 2. The father
* 3. Both mother and father
4. Other, specify
7. When is it likely that a woman gets 1. During her periods
pregnant? 2. Just after a period
3. Just before a period
* 4. In between two periods
5. Don't know

8. What do you think is a good time
interval between pregnancies for a
woman?

 

9. Who do you think would be most suit-
able to tell people about how to
space children?

 

124

10 w
Child Care

1. What do you think is healthier for a
baby, to be breast fed or bottle fed?
Why?

2. How long did you breast feed your
last child?

3. At what age do you think a baby
needs additional solid food to
breast milk or cow's milk?

 

 

4. What is the first food of the fol-
lowing that should be introduced
to a baby?

5. If a child has diarrhea, what
would you give it to drink?

6. If the baby is bottle fed, how do
you think the bottle should be
cleaned?

How often?

7. Do you think it is important to
take a child to the clinic?
Why?

If yes, at what age should it be
brought to the clinic?

8. Do you think play other than helping
' parents with work is important for a
child's development?

Why?

 

 

* 1.

2.

awn:

Nd #wN-fl OSU'Ith-d th—r' (”45de U1
0 C O O O O C

m...
0

Breast fed
Bottle fed

. Smiles at the mother

or other family mem-
bers (2 months)

Sits up with support
(3-4 months)

Creeps and crawls
(7-9 months)

Walks (10-12 months)
Other, specify

. Mealiemeal

Eggs

Fruit juice
Greens

Other, specify

Boiled water

. Milk
. Sour milk
. Water and ashes

By your hand only
With cold water
With water and soap
With hot water
Boiled in water

. Other, specify
. Never

Some times
Every day
After meals

. Yes

No

Yes
No

opinions do you have about the following?

1.

125

11 w

Nutrition

It is important that your family gets the right food. What

Who in the family do you think
has the greatest need for

l.
2.
nutrition? Rank in order: * 3.
4.

What is the single most nutritive
food item?

Which one of the following foods
is the best body-building material?

Which one of the following foods is
good for the blood?

Which one of the following foods
provides the most energy?

Which one of the following foods is
good in preventing infectious
diseases?

Which one of the following loses its
food value first, if over-cooked?

What would you add to porridge and
beans to make the meal well-balanced?

___The wife

___The husband

___The wife when pregnant

___Children older than
five years of age

Spinach
E99
Bread
Mealies

hWN-d
o o o o

. White bread
Jugobeans
Spinach
Mealie-rice

430de
O Q I

Milk

Liver
Sugar
Bread

wa-J
o o o o

Butter
Jugobeans
. Cabbage
Milk

DNNd
o 0

Bread
Oranges
Oil
Milk

thd
O O O

. Eggs

Cabbage

Jugobeans
Mealiemeal porridge

#QDN-H
o o 0

White bread
Spinach

. Butter

4. Meat

*
“Nd
o

126

11 w
Nutrition, continued
9. Food consumption record

What food did you serve your family yesterday?
(Include beverages and snacks)

 

 

 

 

 

 

 

 

 

 

 

8
Last night?
During the day?
In the morning?
Did you purchase any food yesterday? 1. Yes
2. No

If yes, what did you buy?

 

How much did you pay?

 

 

 

APPENDIX B

GUIDANCE QUESTIONS FOR
UNSTRUCTURED INTERVIEWS

127

 

 

 

 

01 h u N
o o o o

11.

12.

13.

14.

APPENDIX B

GUIDANCE QUESTIONS FOR UNSTRUCTURED INTERVIEWS

How would you like your children to earn their living?
Sons? Daughters? Why?

Where would you wish your children to live? Why?
How many children would you wish for them to have? Why?
What kind of house would you like for them to live in? Why?

Are there any special things you would like for them to get?
Why?

Why is it good to have children?

How many children do you think a woman should have? A man?
33:: it make any difference if they are boys or girls?

Do you think children should go to school? Why?

Would it be good to decide how many children one wants? Why?

Do you think it is possible to decide how many children one

wants? Can a woman decide? Can a man decide?
Who should decide? Why?

Do you think most women would like more children?

0r less children? Why?

Do you think most men would like more children?

Or less children? Why?

Do you think it would be good for Swaziland if more children
were born? 0r less? Or is it good like it is?

Why?

If one of your children should be seriously sick, where would
you go for help?

128

 

 

15.
16.
17.
18.

129

What is determining how many children people have?
What is the biggest expense of having children?
If you suddenly got a lot of money, what would you do?

What are the biggest problems you have?

 

 

APPENDIX C

MAP OF SWAZILAND

130

 

 

131

   

 

       
  
     
  
 
  

H A f a I C A
o “ I
5 ..
, -’ \\ \ N
g a . /Ie Newman"
0 .I _.~ rf¥
I! m ’0‘)? >\\

c /'

\ .IL"“

\, t >

0‘ 41/
Q // ’1
Q, , a."
Q- ‘K ' c ;
In Dotbwon - {-L o
lunar, lam-mu I “I
/. ‘ f . ’$~::;f1l‘ ’

/ k \

/‘ \ , ,

/‘ ‘) \/

 

I -
/ ~/
/ ‘./o I: that M\
to Emlo .\ I r _
Owe. ) I

 
  
   

. \ ' -
w Iolel X ‘\ , ‘. manna \
W- « A
Y / \ ... .- \\
. \ ' Em...— ,
/ I ;" 1 \ 1
II 5\\ K .5 917’),- \ I
f \ A . obiub‘v’ r

I MNombcnyul. ' an ’ .'~ J (”'73
I \ ““u'. / . (mush
I /’ ' \

 

    

Mia». ‘11--
I \§’\"q
!

l -

| o P ...-"3".

I ‘ vuwzwem

{Ask .ANZINI

. , / ..."-
. ~\. ‘/ (1.395143. 1:_
. \ ”lATIlUlU
/ ‘6". \\/ .1 Moaimn\~
\~ |
,(

 
 
 
    
 

1. ma Inc-.4 ( vl \

x a
v. r... 0.. .1 \p \3’, 4‘ \% \\/ VI
\ I

   
  

 

 

   

Cohen}:
'6 ‘\\‘\ r EN )' """ ....9
; 1’ — , ) I
o o \ .EKX [WEI/C «‘1 x i
uonm l \ \{\ I ' I
MM‘Jonqd — I C I K I
. . (”um -== ' , A
Haw, load —— —— I
s O .‘ — . I A unus-

U 7 H A f . I C karma 10.0.4an
.o ... ...-1..

u. '1
l .l....l .I - I
.lAIl mm

Figure 2.--Map of Swaziland

APPENDIX D

RESULTS OF ANALYSIS OF
VARIANCE TESTS

132

 

 

133

.Pm>m_ yo. no mucouwwwcmwm mmum6wucma

 

 

oooo.o oooo.o oooo.o oooo.m~o.m mm_ pmuo»

oooo.o oooo.o mpmm.ep oomm.vw_.m map mommmpunzm cpzuwz
_-¢.o ammo.P comm.e~ Nmmm.~m— m mcowuomgmpcw umpooa
oooo.o oooo.o oooo.o momm.~mw.o o Amucmumcou __mv cowpuscmm
oooo.o oooo.o oooo.o eo.mmm.o mp cmpumgcouca mommmpunsm
mpep.o mpmp.m ompp.—m ompp.Pm F xmm m
moms.o Nm¢~.o cmmm.m mm~o.u N mm< N
«omFo.o emoe.¢ mmm~.mo mom¢.-~ N mmg< F
wummuMwwmmwm mzpm>uu mgmzcm cow: mmgmzcm mo 23m .u.o mocmwgm> we mugaom mmmpu

 

.cmgupwsu mo gmasaz umgwmmo so» 9mm» mucmpgm> mo mwmapmc< mzu mo mpF=mmmuu.om m4m<h

 

 

 

vi
hr
In
.C
3
1'7
$5

4.3.2 1... ......T, u ... .1... mm aflhhfi

 

134

 

.Fa>ap Po. an

eeceewwcmwm meueewecmw

 

 

oooo.o oooo.o oooo.o oooo.muo.m mop pouch
oooo.o oooo.o mmoo.v~ mmom.m¢o.N new mmmmmpunzm :wcuwz
Nmmm.o wNmp.F Foam.mp vav.cmp NP mcowuuegwucw umpeoa
oooo.o oooo.o oooo.o oemN.nmw.o o Amgcmamcou prev :owuozumm
oooo.o oooo.o oooo.o emmm.NNo.m mp umuuwwsouca mmmmepunzm
oomm.o rm-.— emvo.m~ nmmN.~m N cmwopwcu we Lmnsaz m
mpmo.o voom.m wemN.m¢ memN.mv P xmm N
«mm—o.o Nmmo.¢ omem.mm Pomo.m_p N mms< —
meceewwwemwm eepe>um egeeem cue: mmgeeem we saw .m.o eecewge> we eegeem mmmpu

we Pe>e4

 

.emgepwzo we geesez eegwmec Lew anew eueewge> we mwmxpe=< an» we mapememun.pm m4m<w

 

 

 

135

 

 

 

oooo.o oooo.o oooo.o oooo.m¢N.m_ mow FeuOP
oooo.o oooo.o ONmn.~n comm.oem.o_ pmp mmmmepunam cwnuwz
mwmn.o mome.o menm.me moum.om¢ m mcewuumwmucw umpoee
oooo.o oooo.o oooo.o Pmmm.mnm.n o Amvcepmcou Ppmv cowuuaumm
oooo.o oooo.o oooo.o p¢.¢o¢.w mp umpowswouca mmmmmpunam
ommm.o ompm.o PNeo.om FNvo.mm F xwm m
mmNm.o mpmp.o m~m~.mp nmom.NN N mm< N
mveo.o comp.m Nmmo.mNN mopp.mme N emw< _
eunmewwwmmwm ee_e>-w eweeem new: meweeem we 23m .w.o eecewwe> we eeweem mmepu

 

.eewewwee we Leesez weeee eew emew eeeewee> we eweaweee me» we eeweeem--.~m wee<w

it!»

  

 

 

.25.: .4. _.. , :3 1,: ....1. e. ......» .

 

   

I ,1 1 l‘l‘ulI'D‘.
.. .1.
. . 15111.

,.r, I
.

  

136

 

 

 

oooo.o oooo.o oooo.o oooo.on.mF mow Pepe»
oooo.o oooo.o ommp.NN mmme.nmw.o_ mew memmepeeem cwcuwz
mmom.o momm.o mmmw.me empm.mmm NP mcewueeweacw em—eee
oooo.o oooo.o oooo.o mwmm.¢wo.w o Amucepmcee —Fev cewpeeeem
oooo.o oooo.o oooo.o Nwom.mom.m mp eeueewweece memmepeeem
emmm.o coco.o emeo.o moao.o N :ewewwce we Lease: m
wmmN.o Nmmm._ Nmmm.oow mmwm.oop w xem N
Nneo.o coop.m omNN.eNN ommm.mee N emw< _
eewmewwwmmwm m=Fe>1w eweeem :eez meweeem we 53m .w.o eecewwe> we eeweem mmepo

 

.eewewwee we weeEez weeee wew emew eeeewwe> we ewmawee< age we eewemem--.mm meeew

 

1

 

137

 

.Fe>eF Fe. Fe eeceewwwcmwm meueewecF«

 

 

oooo.o oooo.o oooo.o oooo.mmw.eF me moF FeueF
oooo.o oooo.o memN.m OONm.mom emF memmeFeeem :Fsuwz
eoom.o «mme.o ommm.N Nmmm.mN m mcereeweucw eeFeee
oooo.o oooo.o oooo.o weFN.mmN.mF m qucepmcee FFev :eFuueeem
oooo.o oooo.o oooo.o mm.mFm.mF mF eeueewweece memmeFueem
*eFoo.o Nme.oF mmFm.mm mmFm.mo F xem m
FNee.o mONm.o NeoF.m mmNm.oF N em< N
«eFoo.o eon.o meeN.me mmwe.mw N eew< F
eewmuwmemFm eeFe>nw eweeem ceez mmweeem we Eem .w.a eecewwe> we eeweem mmeFu

 

.eweum eeepwup< eeFueFeeee wow “mew eecewwe> we mwmaFec< ecu we muFememnu.em m4m<w

138

 

 

.Fe>eF Fo. we eeceuwwwcmwm meueewech

oooo.o oooo.o oooo.o oooo.mm~.eF moF Feuew
oooo.o oooo.o FNNN.o mmmo.Fem FmF memmeFeeem cwguwz
mmmm.o Nmom.o mNoo.m memm.no NF mcewpeewmucw eeFeee
oooo.o oooo.o oooo.o mme.mNN.mF m quceumceu FFeV :eFuueeem
oooo.o oooo.o oooo.o umem.nem.mF NF eepuewweece memmeFeeem
«FNN.o omNm.F mFme.m mem.mF N cowerge we weasez m
wFooo.o NNme.mF mome.om mome.mm F xem N
«oooo.o mome.mF Fmou.mm moo¢.~oF N eew< F
eemmewwwmmwm eeFe>1w eweeem see: meweeem we saw .w.o eecewwe> we eeweem mmeFu

 

.eweem eeepwuu< :eFFeFeeee wew umeF eecewwe> we mFmFFec< ecu we muFemem--.mm mem<w

1|

 

 

.23.... .w .1... .. 2:31.: 2.2254,: .. 3.; q . . ..

 

..0 ..NT

u“
I“ 11'1"...) 11.14.

T
. all}.-.

 

139

 

 

 

.Fe>eF F0. we euceewwwcmwm meueewecFa

0000.0 0000.0 0000.0 0000.NFN.mwm Fm Fave»
0000.0 0000.0 wme.0ww moom.wN¢.mm 0e mmmmmFuazm cwguwz
mNF0.0 000N.0 0F¢0.00F mmwm.FNm N mcowuuewwucw vwFooe
0000.0 0000.0 0000.0 NOF0.00¢.Nvm e Ampcepmceo FFeV cewuusumm
0000.0 0000.0 0000.0 Nm00.mmN.Nem 0 umuumwwouc: mmmmeFunsm

. . . . eewewwee

mee e meNm e wwew eeN erN eeN F we weesee eeez N
*mm00.0 00NN.0 mmNm.mmm.e Nmm0.w00.m N emw< F
eummewwwmwwm eeFe>1w eweeem gee: meweeem we saw .w.0 eecewwe> we eeweem mmeFu

 

.eweem eeeuwup< cewpeFeeee see: wew pmew eecewwe> we mwmeec< use we mpFememuu.0m mem<w

140

‘r

 

 

 

.Fe>eF F0. we eeeeewwwcmwm mopeewecF«

0000.0 0000.0 0000.0 0000.0ee.F 00F Feeew
0000.0 0000.0 000N.0 0000.FNF emF memmeFeeem cwsuwz
*0000.0 omme.e 0000.0 Fomm.Fm 0 mcewueewepcw eeFeee
0000.0 0000.0 0000.0 000N.FON.F 0 quceumceu FFeV ceFueeeem
0000.0 0000.0 0000.0 mm.mNm.F mF eepeewweeee memmeFenem
*0F00.0 mmNm.0F meF.0 meF.m F xem m
FN00.0 0000.0 eFmo.0 NNOF.0 N e0< N
«FN0.0 mmmm.N mFNF.N mNeN.e N eew< F
eunmewwwmmFm eeFe>1w eweaem :eez meweeem we Eem .w.0 eecewwe> we eeweem mmeFu

 

.eweem :eFeeeeeweem wow “mew eecewwe> we mwmaFec< ecu we muFememlu.Nm 000<F

t... ..

1. n... .. we“. .... r if

....vII'\ ...0'. 5.... clifl|lvl1e11lf11btf .

 

 

 

141

.Fe>eF F0. Fe eeceewwwcmwm meeeeFecFa

 

 

0000.0 0000.0 0000.0 0000.000.F 00F FeHOF
0000.0 0000.0 0000.0 0000.0NF FOF mwmmeFunzm :Fsng
00F0.0 FN00.0 0000.0 0000.0 NF mcoFuuewmucw umFoom
0000.0 0000.0 0000.0 000N.0F0.F 0 quceumcou FFev :oFuuauwm
0000.0 0000.0 0000.0 0000.0N0.F 0F nouumwwouca mummeFunzm
000F.0 00F0.F 00N0.F 0000.0 N :mquwsu we wensaz 0
«0F00.0 NOF0.FF 0000.0 0000.0 F xmm N
0FOF.0 0000.F 0000.F N000.N N eww< F
eemwewwwmmwm ezFe>1w eweeem new: meweeem we 5:0 .w.0 eeeewwe> we eeweem mmeFu

 

.eweum cewuezeeweem wow “mew eecewwe> we mwmaFe=< we» we maFemem-u.00 000<F

 

 

, 5.7.1,... 1... E.“ ._ 0-0.1.... 4......” m .. ex}; 1.

 

142

 

.Fe>eF F0. we euceowwwcmwm 0.33.55.w

 

 

0000.0 0000.0 0000.0 0000.000.F 00 Fequ

0000.0 0000.0 0N00.N FFON.FFN 00 memmeFoeem cwguwz
NF00.0 NF00.N 000w.0 0000.0F N mcewueeweucw eeFeee
0000.0 0000.0 0000.0 00N0.00N.F 0 Fauceumcee FFev cewuoeeem
0000.0 0000.0 0000.0 000N.NON.F 0 eouuewwouee memmeFoeem
0000.0 0000.0 NOF0.N NOF0.N F m0< N
«0000.0 0NF0.0 0000.FF 0NFO.mN N eew< F
mommewwwmmwm eeFe>1w ewezem new: meweecm we 5:0 .w.0 mocewwe> we eeweom mmeFu

 

.eweom ocew0xz wow “wow eocewwe> we mFmaFec< ecu we mpFememul.00 000<F

143

 

 

0000.0 0000.0 0000.0 0000.000.N 00 Fewew
0000.0 0000.0 0000.F N000.00F 00 memmeFenem :Fcuwz
000F.0 NF00.F 0000.0 0000.0 N mcowueewoucw eeFeew
0000.0 0000.0 0000.0 0000.00N.N 0 quceemcee FFev ceFuueeem
0000.0 0000.0 0000.0 0000.0FN.N 0 eepoewwoecz memmeFunem
8005 $25 0.85 285 F ee< N
FN00.0 FmoF.0 000F.0 00Nm.0 N eew< F
eewmuwwwmmwm eeFe>uw eweeem :eoz meweeam we sew .w.0 moceww>> we eeweom mmeFu

 

.eweom eweu 3.20 wew umew mocewwe> we 0.5.32.0 e5 we $3801.00 0005.

 

144

 

 

 

0000.0 0000.0 0000.0 0000.00N.F 00 FeueF
0000.0 0000.0 F000.N 0FNN.0NF 00 mmmmeFunam cwguwz
NON0.0 ~00F.0 0N00.0 FO0N.0 N mcowuuewmucw 0mF000
eeee.e eeee.e eeee.e meee.eee.w e Aeeeeeeeee Fwev eeweeeeem
0000.0 0000.0 0000.0 00Nm.000.F 0 nmuumwweuca mmmmeFunzm
000N.0 000F.0 000N.0 000N.0 F m0< N
0000.0 0F00.N 0FF0.0 0NNO.FF N emw< F
eowmoflwwmmFm eeFe>nw eweeem :eez meweeem we 5:0 .w.0 oucewwe> we oeweom mmeFu

 

.ewoom cowuwwuez wow “wow eucewwe> we mwmaFec< esp we mpFemomuu.F0 0400F

145

.Fe>eF F0. we moceonF=0Fm meweewech

 

 

0000.0 0000.0 0000.0 0000.0N0.F NN Feuew
0000.0 0000.0 NNON.0 NN00.000 N0 memmeFoeem ewguwz
0000.0 0F00.0 N00m.0 0N00.0 N mcewpueweucw ueFeoe
0000.0 0000.0 0000.0 000N.0FF.F 0 qucepmcou FFeV :owuoeeem
0000.0 0000.0 0000.0 mNm0.0FF.F 0 eeuoewweoc: menmeFoaem
FO0N.0 0FOF.0 000F.F 000F.F F e00 N
*0000.0 0000.0 NNOF.00 00FN.N0 N eow< F
mommuwwwumFm eeFe>nw eweeem eeez mmweecm we 2:0 .w.0 eecere> we eoweom mmeFu

 

emsoom worn—.00.:— Emm Low. Hmmh mucwwLm> mo mwm>._.mc< m...“ ....o mupsmmmll.N.—N mama:

 

146

 

 

0000.0 0000.0 0000.0 0000.000.00F NN FeueF
0000.0 0000.0 NNOF.00F N000.00N.FF N0 memmeFoeem cwnuwz
00F0.0 0000.0 0000.00F NN00.00N N meowpoeweucw ueFeee
0000.0 0000.0 0000.0 N000.000.00F 0 quceumceo FFeV cewueeeem
0000.0 0000.0 0000.0 0000.0N0.00F 0 eeueewwoece memmeFenam
NOFF.0 00N0.N 00F0.0N0 00F0.0N0 F e00 N
00N0.0 000N.0 0N00.Fm0 0000.NON.F N eew< F
eemmewwwmmwm eeFe>nw eweecm ceez meweeem we 5:0 .w.0 oocewwe> we moweom mmeFu

 

.eweem eeweeewwwee wew emew eeeewwew we mwewweee eee we eewemee--.ee eee<w

 

147

 

 

.Fo>oF F0. we euceonF=0wm moweowecF«

0000.0 0000.0 0000.0 0000.0FN.F FN FeweF
0000.0 0000.0 0000.0 0NON.00N 00 093283 cwcwwz
00NN.0 00Fm.0 NOF0.F 00N0.N N meowweewewew eeFooe
0000.0 0000.0 0000.0 00Nm.FNO.F 0 mecewmcoo FFeV cewweeeem
0000.0 0000.0 0000.0 0NON.0N0.F 0 eewuowwooce mommeFunem
meeF.e NONN.N emeF.NF emeF.NF F e00 N
*0000.0 000N.0F FON0.F0 FO0N.0NF N eew< F
oommowwwmmFm eeFe>1w oweecm :eez meweeem we 5:0 .w.0 eoeewwe> we eeweom mmeFu

 

.mEooeH wow wmeF eocewwe> we mwszec< mcw we mermomul.00 000<F

 

148

 

 

0000.0 0000.0 0000.0 0000.0N0.000.0 F0 FeweF
0000.0 0000.0 NF00.FON.F NO0N.N00.00 00 mmmmeFuasm :stwz
NONN.0 0NON.0 0000.000 NFNN.F00 N mcowwuewwwcw neroa
0000.0 0000.0 0000.0 FN00.000.000.N 0 mecewmceu FFeV :onuaumx
0000.0 0000.0 0000.0 000N.N00.000.N 0 vmwuwwwouca mmmmeFuaam
. . . . :ewerce
emwm e mer 0 meme 000 e000 000 F we weesee eeez N
NuoN.0 00N0.F FONN.000.N 0000.N00.0 N emw< F
eonmowwwmmFm eeFe>uw eweecm :eez meweecm we 5:0 .w.0 mocewwe> we eeweem mmeFu

 

.msoum mmmcpawmuszowmm Low Hmmh wucmem> ....o 0.092.005 0:». $0 mupzmmm11.m¢ m.._m<._.

 

BIBLIOGRAPHY

149

 

BIBLIOGRAPHY

"Agricultural Sample Census, Swazi-Nation Land, 1971-72." Mbabane:
Central Statistical Office, 1973. [‘

Allen, C. J. "Dimensions of Swazi Households in Rural and Urban
Areas.“ Ph.D. dissertation, University of Massachusetts,
1973.

 

"Analyst Biases in KAP Surveys: A Cross-Cultural Comparison."
Studies in Family Planning 7 (November 1976): 322-33.

Becker, G. An Economic Analysis of Fertility--Demographic and
Economic Change in Developing Countries. Columbia Univer-
sity Press for National Bureau of Economic Research, 1960.

Berelson, 8., ed. Family Planningpand Population Programs: A
Review of World Develppments. Chicago: University of
Chicago Press, 1966.

Borgstrom, G. World Food Resources. New York: Intext Educational
Publishers, 1973.

Caldwell, J. C. Population Growth and Family Change in Africa:
The New Urban Elite in Ghana. Canberra: Australian
National University Press, 1967.

. "Toward a Restatement of Demographic Transition Theory."
Population and Development Review 2 (Nos. 3 and 4): 321-66.

Ceres, November-December, 1973. Rome: Food and Agricultural Orga-
nization of the United Nations.

Coal, A. J. "The Demographic Transition Reconsidered.“ In:
International Union for the Scientific Study of Population
(IUSSP). International Population Conference, Liege, 1973.
Liege, IUSSP, 1973: 53-72. [Population Reports, Series 7,
No. 12 (November 1976): 7-213:]’

Clay, D. C., and Zuiches, J. J. "Reference Groups and Family Size
Norms." Department of Sociology. Michigan State University.
Paper prepared for submission to the annual meeting of the
Population Association of America, Session "Fertility Trends
in Developed Nations," St. Louis, April 21-23, 1977.

150

 

151

Cunningham, E. P., and Henderson, C. R. "An Iterative Procedure
for Estimating Fixed Effects and Variance Components in
Mixed Model situations." Biometrics 24 (1968): 13-25.

Davis, K., and Blake, J. Social Structure and Fertility: An Ana-
lytic Framework. The Bobbs-Merrill Reprint Series in the
Social Sciences, s-37l, a reprint from Economic Development
and Cultural Change, Vol. 4, April 1956.

 

 

Deacon, R., and Firebaugh, F. M. Home Management: Context and Con-
cepts. Boston: Houghton-Mifflin, Inc., 1975.

De Tray, P. "The Substitution Between Quantity and Quality of
Children in the Household." Ph.D. dissertation, Department
of Economics, University of Chicago, 1972.

Englebrecht, J. A. Swazi Customs Related to Marriage. Cape Town:
Nationale Press, 1930.

Fair, T. J. 0., et al. Development in Swaziland: A Regional Anal -
sis, Johannesburg:‘TWitwatersrand university Press, |969.

"First Draft of a Report of a Home Economics Survey in Swaziland."
Mbabane: Ministry of Agriculture, 1970. (Mimeographed.)

Freidman, R. "Application of the Behavioral Science to Family
Planning." Studies in Familprlanning,23 (October 1967): 6.

Gani, A. Report on the First Round of Demographic Survey. Swazi-
land, 1975.

Gross, I. H.; Crandall, E. W.; and Knoll, M. M. Management for
Modern Families. New York: Appleton-Century-Crofts,
Educational Division, Meredith Corporation, 1973.

 

Guttman, L. "The Basis for Scalogram Analysis." In Measurement and
Prediction, ed. Samuel A. Stouffer. Princeton, N.J.:
Princeton University Press, 1950.

 

Harter, H. L. "Critical Values from Duncan's Multiple Range Test."
Biometrics, 16, No. 4 (1960).

Harvey, W. R. "Least Squares Analysis of Data With Unequal Sub-
class Numbers.“ U.S. Department of Agriculture, ARS 20-80,
1960.

Holleman, J. F., ed. Experiment in Swaziland: Report of the Swazi-
land Sample Survey, 1960. Cape TBWn: OXford University
Press, 1964.

Hook, N., and Paolucci, B. "Family as Ecosystem." Journal of Home
Economics 62 (May 1970): 315-318.

 

 

152
Jones, H. H. Report on the 1966 Swaziland Population Census.
Mbabane: Swaziland’Government, 1968.
Kempthorne, O. The Design of Experiments. New York: Wiley, 1952.
Kiser, C. V., and Welpton, P. K. "Resume of the Indianapolis Study

Social and Psychological Factors Affecting Fertility."
Population Studies 7 (1973): 95-110.

 

in Fertility in the United States." In Vital and Health
Statistics Monographs, American Public Health Association.
Cambridge: Harvard University Press, 1968.

Kiser, V.; Wilson, H. 6.; and Campbell, A. H. "Trends and Variation E

 

Koza, C. M. T. "Proposed Guidelines and Model for a Home Economics
Family Planning Program in Swaziland.” M.S. thesis,
University of Wisconsin at Madison, 1973.

 

Kuper, H. An African Aristocracy: Rank Among the Swazi. London:
Oxford University Press, International AfricaniInstitute,
1965.

. The Swazi: A South African Kingdom. Case Studies in
Cultural Anthropology. New‘York: Holt, Rinehart and Win-
ston, 1963.

 

. "Swaziland (Native Races)." The Uniformity of Colour.
Witwatersrand Press, 1947.

Leibenstein, H. Economic Backwardness and Economic Growth. New
York: John Wiley & Sons, Inc., 1957.

Liston, M. I. "Managerial Functioning." In Behavioral Aspects of
Management, Proceedings of a conference fOrTWestern Regional
Teachers of Home Management-Home Economics, Salt Lake City,
Utah, 1971. PP. 57-78. '

 

Malthus, R. I. "An Essay on the Principle of Population." London,
1803.

Marwick, B. A. The Swazi: An Ethnographic Account of the Natives
of the Swaziland Protectorate. Londfin: ’FFank Cass & Co..
Ltd., 1966. *(First edition, 1940.)

Nickell, P.; Rice Smith, A.; and Tucker, P. Management in Family
Living. New York: John Wiley & Sons, Inc., 1976.

Notestein, F. W. "Population: The Long View." In Food for the
World, ed. T. W. Schultz. Chicago: University ofiChicago
Press, 1945, pp. 39, 41.

 

153

qumalo, S. S. Our Way of Life. Swaziland: Swaziland Printing and
Publishing Co., Ltd., 1976.

 

Paolucci, 8.; Hall, A.; and Axinn, N. W. Family Decision Makin :
An Ecosystem Approach. New York: John Wiley & Sons, 7.

 

 

"Population Figures: Provisional Census Figures as of Midnight
25-26th August, 1976." Mbabane: Central Statistical
Office. (Mimeographed.)

"Programmes Related to Family Living in Swaziland: A Handbook."
Luyengo, Ministry of Agriculture, 1976.

"Project Request for United Nations Fund for Population Activities."
Mbabane: Ministry of Health, 1976.

"Report of the Committee on Juvenile Delinquency, Sexual Immorality
and the Maintenance of Illegitimate Children in the Urban
Areas of Swaziland." Mbabane: Ministry of Local Admin-
istration, 1970.

Report to the Government of Swaziland: Development of a Home
Economics DipToma Course at theiSWaziland Agricultural Col-
lege and University Centre. Rome: Ebod'and Agricultural
Organization of thé’United Nations, 1971.

 

"The Reproduction Function." The Economist (January 8, 1977): 72-73.

 

Rose, T. J. "An Investigation Into the Economics of Grain Storage
Among Subsistence Farmers in Swaziland." Report One,
Faculty of Agriculture, University of Botswana, Lesotho and
Swaziland (October 1975). (Mimeographed.)

Schapera, I., ed. The Bantu Speaking Tribes of South Africa. Lon-
don: Routledge and Kegan Pau1,iLtd., 1966.

Second National Development Plan 1973-77. Government of Swaziland.
Steidl, R. "An Ecological Approach to the Study of Family Behavior."

In The Family: Focus on Management. Proceedings of a
National Home Management Conference, June 1969.

 

 

"World Fertility, 1976: An Analysis of Data Sources and Trends."
Population Reports (Series J, No. 13 [January 1977]:
J-237-J-238.

 

"World Population Data Sheet." Population Reference Bureau, Inc.

nIcwran STATE UNIV. LIBRnRIEs
IIHIWIWIIW”WW”IWIIUIIHIIVWIWIINWI
31293100395122