Front Cover
 Title Page
 Table of Contents
 Part I. Looking at rural women
 Part II. Approaches to learning:...
 About the authors
 Back Cover

Title: Studies in family planning
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00086911/00003
 Material Information
Title: Studies in family planning
Abbreviated Title: Stud. fam. plann.
Physical Description: v. : ill. (part fold.), maps (part col.) ; 28 cm.
Language: English
Creator: Population Council
Publisher: Population Council
Place of Publication: New York
Frequency: quarterly[<1997- >]
monthly[ former july 1963-]
monthly (with combined june/july, aug./sept. issues)[ former ]
bimonthly[ former ]
Subject: Birth control -- Periodicals   ( lcsh )
Family Planning Services -- Periodicals   ( mesh )
Gezinsplanning   ( gtt )
Régulation des naissances -- Périodiques   ( rvm )
Genre: periodical   ( marcgt )
Additional Physical Form: Also issued online.
Dates or Sequential Designation: v. 1- July 1963-
Numbering Peculiarities: Vol. 1- called also no. 1-<25>
General Note: Supplements accompany some issues.
 Record Information
Bibliographic ID: UF00086911
Volume ID: VID00003
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 01651215
lccn - 71001187
issn - 0039-3665

Table of Contents
    Front Cover
        Front Cover
    Title Page
        Title Page 1
        Title Page 2
    Table of Contents
        Table of Contents 1
        Table of Contents 2
        Page 309
        Page 310
        Page 311
        Page 312
    Part I. Looking at rural women
        Page 313
        Page 314
        Women, doctors, and family health care: Some lessons from rural Java
            Page 315
            Page 316
            Page 317
            Page 318
            Page 319
            Page 320
            Page 321
            Page 322
            Page 323
            Page 324
            Page 325
        Rural women's credit systems: A Nigerian example
            Page 326
            Page 327
            Page 328
            Page 329
            Page 330
            Page 331
        Marked preference for female sterilization in a semirural squatter settlement
            Page 332
            Page 333
            Page 334
            Page 335
            Page 336
        Women in the household economy: Managing multiple roles
            Page 337
            Page 338
            Page 339
            Page 340
            Page 341
            Page 342
            Page 343
        Women's reality: Critical issues for program design
            Page 344
            Page 345
            Page 346
            Page 347
            Page 348
            Page 349
            Page 350
            Page 351
            Page 352
        From research to policy: Rural women in India
            Page 353
            Page 354
            Page 355
            Page 356
            Page 357
            Page 358
    Part II. Approaches to learning: A sharing of experience
        Page 359
        Page 360
            Page 361
            Page 362
            Page 363
        Group I: Methodological issues
            Studying rural women in West Java
                Page 364
                Page 365
                Page 366
                Page 367
                Page 368
                Page 369
            Measuring rural women's work and class position
                Page 370
                Page 371
                Page 372
                Page 373
            Collecting data on women's employment in rural Java
                Page 374
                Page 375
                Page 376
                Page 377
                Page 378
        Group II: Time use and project planning
            Collection and analysis of data on rural women's time use
                Page 379
                Page 380
                Page 381
                Page 382
            Rural women's time use
                Page 383
                Page 384
        Group III: Instruments for learning
            Measuring rural women's economic roles and contributions in Kenya
                Page 385
                Page 386
                Page 387
                Page 388
                Page 389
            Learning about women through household surveys: An experimental module
                Page 390
                Page 391
                Page 392
            Assessment of body concepts and beliefs regarding reproductive physiology
                Page 393
                Page 394
                Page 395
                Page 396
                Page 397
            Project-oriented research on aspects of women's knowledge and experience
                Page 398
                Page 399
                Page 400
            Research priorities: Women in Africa
                Page 401
                Page 402
                Page 403
                Page 404
        Group IV: What rural women know
            Circumventing problems of accessibility to rural Muslim women
                Page 405
            How and what rural women know: Experiences in Bangladesh
                Page 406
                Page 407
            Women in rice cultivation: Some research tools
                Page 408
                Page 409
                Page 410
                Page 411
            Research on women by women: Interviewer selection and training in Indonesia
                Page 412
                Page 413
                Page 414
                Page 415
        Group V: The voice of rural women
            Profile of a female agricultural laborer
                Page 416
                Page 417
            Anita: A Mayan peasant woman copes
                Page 418
                Page 419
                Page 420
    About the authors
        Page 421
        Page 422
    Back Cover
        Page 423
Full Text


Volume 10 Number 11 / 12 November / Dccember 1979



guest editor

A Special Issue



guest editor

Studies in Family Planning

Editorial Committee Editorial Staff
George F. Brown, Chairman Valeda Slade, Managing Editor
Judith Bruce Robert Heidel, Project Editor
Ethel P. Churchill Renee Santhouse, Production/Design
Margaret McEvoy
Susan A. Robbins A Publication of
S. Bruce Schearer The Population Council
Irving Sivin One Dag Hammarskjold Plaza
Beverly Winikoff New York, N.Y. 10017

Learning about Rural Women

This special issue of Studies in Family Planning is one of a series of issues
dedicated to a single subject or program. Our intention is to provide an examina-
tion in depth of selected key topics beyond the range of a single article.
This special issue focuses on ways in which the roles and status of rural
women in different societies can be better understood. There is a vital need to
understand rural women, not as a set of statistics, but as individuals performing
crucial roles in society and therefore playing fundamentally important parts in
the development process.
A variety of ways to learn about rural women are described. Some ap-
proaches adopt standard sociological and demographic methodologies, while
others employ anthropological techniques. All attempt to begin from the point
of view of the individual woman, to understand how she sees herself in the soci-
ety around her and how she adapts to changes brought about by development
efforts and other social forces.
For those concerned with improving family planning programs, this spe-
cial issue is intended to provide a fresh viewpoint-a perspective on the re-
productive and child-rearing function as viewed by individual women, in the
context of their daily lives. A broadening of our understanding of rural women,
and particularly a deepening of our appreciation of the ways women perceive
their roles, can significantly help to improve program design and management
of family planning and other development efforts. Indeed, the material in this
special issue is intentionally wide-ranging, directed to other sectors of the devel-
opment field in addition to family planning. Most development actions have a
direct or indirect impact on rural women. We must understand rural women
better, in order to plan successful policies and programs and to avoid unin-
tended negative effects. The methods described in this special issue will, I hope,
aid in deepening our understanding.

Director, International Programs
Vice President, The Population Council

The Population Council, Inc. 1979


Sondra Zeidenstein

Part II
Approaches to Learning:
A Sharing
of Experience

Part I
Looking at
Rural Women

315 Women, Doctors, and Family Health Care:
Some Lessons from Rural Java
Valerie J. Hull
326 Rural Women's Credit Systems:
A Nigerian Example
Kamene Okonjo
332 Marked Preference for Female Sterilization
in a Semirural Squatter Settlement
Leela Gulati
337 Women in the Household Economy:
Managing Multiple Roles
Achola Pala Okeyo
344 Women's Reality: Critical Issues
for Program Design
Taherunnessa Abdullah
and Sondra Zeidenstein
353 From Research to Policy: Rural Women in India
Vina Mazumdar

361 Introduction
Judith Bruce

Group I: Methodological Issues
364 Studying Rural Women in West Java
Pudjiwati Sajogyo, Endang L. Hastuti,
Syarifah Surkati, Winati Wigna,
Krisnawati Suryanata,
and Benjamin White
370 Measuring Rural Women's Work
and Class Position
Carmen Diana Deere
and Magdalena LeOn de Leal
374 Collecting Data on Women's Employment
in Rural Java
Nancy Lee Peluso

Group II: Time Use and Project Planning
379 Collection and Analysis of Data on Rural
Women's Time Use
Brenda Gael McSweeney



383 Rural Women's Time Use
Vivian Havens Gillespie

Group III: Instruments for Learning
385 Measuring Rural Women's Economic Roles
and Contributions in Kenya
Audrey Chapman Smock

390 Learning about Women through Household
Surveys: An Experimental Module
Nadia H. Youssef and Coralie Turbitt

393 Assessment of Body Concepts and Beliefs
Regarding Reproductive Physiology
Michele Goldzieher Shedlin

398 Project-Oriented Research on Aspects
of Women's Knowledge and Experience
Taherunnessa Abdullah
and Sondra Zeidenstein

401 Research Priorities: Women in Africa
Achola Pala Okeyo

Group IV: What Rural Women Know
405 Circumventing Problems of Accessibility
to Rural Muslim Women
Marsha Safai
406 How and What Rural Women Know:
Experiences in Bangladesh
Gudrun Martius-von Harder
408 Women in Rice Cultivation:
Some Research Tools
Joan P. Mencher, K. Saradamoni,
and Janaki Panicker
412 Research on Women by Women: Interviewer
Selection and Training in Indonesia
Hanna Papanek

Group V: The Voice of Rural Women
416 Profile of a Female Agricultural Laborer
Leela Gulati
418 Anita: A Mayan Peasant Woman Copes
Mary Elmendorf

421 About the Authors


Sondra Zeidenstein

In this special issue of Studies in Faniily Planning, we
suggest why knowledge about rural women is essen-
tial for all aspects of policy, planning, and implemen-
tation directed toward goals of rural development
(see Part One), and how learning about rural women
can proceed rapidly (see Part Two).
Learning about rural women is essential because
of the nature of their responsibility. A basic feature
of traditional rural society is that the family is the
unit through which people seek to fulfill their needs
and improve their condition. An equally basic fea-
ture is that family responsibilities are usually di-
vided along sex lines among adult males and
females. Each sex has responsibility for carrying out
certain aspects of the work necessary for family sur-
vival, a division of responsibilities that has typically
been referred to as "complementary." Although this
description is familiar and seems to ring true, it does
not explain the situation fully enough to be useful in
interpreting many phenomena of rural life. Based on
evidence from many sources,' it seems more accurate
to say that, for members of each sex, responsibility to
the family usually includes production or provision
and control or management of the resources needed
to carry out their work.
What is the work of rural women? The specific
tasks vary because rural settings and, within them,
rural cultures differ markedly from one another. But
it is well documented that in most cases the work of
rural women includes childbearing and rearing,
household provisioning and management (cooking,
cleaning, washing clothes, household repair and
manufacture, fuel gathering, provision of water),

'AII the papers in this issue support this analysis (see Okeyo,
Okonlo, and G l particu n r lar), as do much ot the netv data emerging
from studies of rural \\omen in the context of rural development.

and aspects of agricultural production and process-
ing, livestock raising, artisan production, trade, and
income generation. These varied tasks are not per-
ceived by women as falling into separate categories
of familial and nonfamilial, domestic and economic,
nonproductive and productive, but as intrinsically
related. For women, childbearing and rearing are,
among other things, matters of maximizing family
opportunity for survival and security through deci-
sions involving numbers of children (Gulati), their
education (Okonjo), health care (Hull), and the kinds
of work they will do.
Some of the means by which women provide the
resource base for carrying out their family respon-
sibilities are the subsistence production of resources
for household consumption, participation in in-
come-generating work, and traditional, largely fe-
male-run systems of maternal and child health care.
As the economy and direction of opportunity have
become monetized (for various reasons in different
rural areas), women are less able to meet their re-
sponsibilities through subsistence production alone
(Okeyo). Reports of efforts to provide various serv-
ices to rural women in many countries indicate that
women everywhere are looking for opportunities to
increase income and that they are spending their in-
come on basic family needs: food, repair or improve-
ment, clothing, and children's education.
As is the case with men, women's access to and
control over resources affect the degree, mode, and
quality of performance in carrying out their respon-
sibilities. And as with men, shifts in the larger econ-
omy, planned or unplanned, affect women's access
to and control over resources and consequently their
behavior. Planners are aware that policy directed
toward achieving rural development goals depends
for its success on the response of rural people. But
planners have not recognized that women's need for



and use of resources is as basic an element of rural
dynamics as men's, or that the mode and degree of
access to and control over resources differs signifi-
cantly between men and women. By considering
only the behavior of males or assuming that it some-
how accounts for female behavior, planners have
only half the information base needed for determin-
ing policy. For example, poverty is a key policy is-
sue, yet rarely does one hear of policy addressed to
the fact that poverty and its consequences for all as-
pects of rural development result from women's un-
employment or women's alienation from the land as
well as men's.

Rural development policies, whether in agricul-
ture, land tenure, employment, population, health,
or any other area, that do not recognize and support
the resource needs of each sex (i.e., the genuine com-
plementarity of the rural family) will not have their
intended effect and will have unintended negative
side effects. There are many relevant examples of this
issue. Rural employment is an important and ob-
vious one. There has been little awareness of the fact
that women's earnings are an indispensable source
of income to poorer families, who make up the ma-
jority of most rural populations, or that women's
earnings are usually spent on their families' most
basic needs. Because the employment of rural wom-
en in such countries as India and Bangladesh, for ex-
ample, has not been accurately recorded in the past,
it has been difficult to measure the extent to which
women's unemployment has been increasing. The
loss of women's earnings has consequences for fam-
ily and national economies in the form of low sav-
ings, high dependency ratios, scarce family re-
sources (contributing not only to malnutrition,
poor health care, and low literacy, but also to increas-
ing alienation from the land), and a decline in the
value of women that shows up in the changing ratio
of male and female deaths. In many countries
women's employment is still not a policy issue.
However, as a result of important developments in
research about women, this issue is now being se-
riously considered as part of India's Sixth Five Year
Plan (see Mazumdar).
Food production is another major issue. Miscon-
ceptions about food production (who is engaged in it
or aspects of it; who owns the land; what technology
is being used; who makes decisions and on what ba-
sis) and the resulting misdirected programs to in-
crease food production (inappropriate extension
activities, development institutions, credit arrange-
ments) have been and continue to be well docu-
mented. The failure of policies intended to increase

food production is related to failure to understand
the nature and rationale of women's involvement in
food production. Much is being learned about the
nature of women's work, but an ignorance of the ra-
tionale of her involvement persists as new efforts are
under way to increase food production and influence
its distribution. At this critical point we must ask:
What will happen to the goals of rural development if
policy support is for large mechanized farms that
leave women out completely; if women are not given
adequate support to commercialize small farms or
the aspects of farming that are their responsibility; if
the only jobs for rural women are as laborers or on
marginal farms; or if investments in land, training,
and credit are directed only to cash crops, to the ex-
clusion of subsistence crops? There is no doubt that
the answers to these questions will be related not
only to food production, but also to migration, pov-
erty, and fertility.
Fertility is another major issue. Women's re-
source needs and their means of meeting them have
as much to do with desired family size as those of
men or "the family." Both a woman's point of view
about preferred family size and her willingness to act
on this preference are influenced by her access to val-
ued resources, her control over them throughout her
lifetime vis-a-vis males, and the economically based
social pressures to fulfill an accepted role. For exam-
ple, a woman's preference for a large family may be
related to competition for her husband's favor in a
polygamous marriage, to the need for children to do
her subsistence work while she earns money, or to
dependence on males for all contact with the outside
world. Desire for a small family may be related to her
perception, as a wage laborer, of the absolute finite-
ness of the resource base (Gulati). It is critically im-
portant for planners to understand women's perspec-
tive on their situation, not just the perspectives of
men or "the family," if they are to have a more com-
petent understanding of population dynamics and
the likely impact of various policies.
If we look at the institutions that implement pro-
grams involving women, we find the same lack of
understanding. Institutions can succeed only if what
they offer bears some relation to the needs and de-
sires of the groups they serve. The shortcomings of
institutions seeking to reach rural women are signifi-
cantly attributable to their failure to take into ac-
count, in the context of a given culture, women's
coequal role in the rural family and the way they seek
to carry it out. For example, family planning pro-
grams in many rural areas attract only a small num-
ber of eligible women and experience high dropout
rates among women who are willing to try contracep-


tives. This is true despite the fact that pilot ap-
proaches in those areas indicate that a much larger
number are interested in contracepting. Some prob-
lems are related to the difficulty of channeling re-
sources to rural areas through large, centralized
institutions. Many problems are related to the differ-
ence between the perspective of the program and the
perspective of the users. Yet there are few attempts to
understand what kinds of contraception and modes
of delivery women find usable in their circum-
stances, perhaps because there is little recognition
that women evaluate such resources in light of the
socioeconomic framework mentioned above. Studies
of women's attitudes toward menstrual bleeding or
abstinence make more sense when seen as aspects of
this larger framework than as determining factors
operating in isolation. An example of the criteria by
which highly motivated women evaluate available
services can be found in the article by Gulati.
Similarly, health clinics in many rural areas are
underused even though it is known that women
have a strong concern for the health of their children.
Again in analyzing this problem, very little attention
is paid to what rural women need from a health care
system in terms of distance, costs, access without de-
pendence on one's husband, and retention of control
over knowledge. An analysis of health care systems
in these terms can be found in the article by Hull.
Another kind of institution directly involving
rural women is the women's group or women's com-
ponent of rural development activities. Typically its
program has included information and services re-
lated to literacy, nutrition, child welfare, home
economics, labor-saving technology, and, more re-
cently, income-generating projects. In the past these
groups have had low rates of success in attracting the
women they were intended to reach or in changing
behavior, probably because what they had to offer
was of low priority for rural women. Articles by
Okonjo and Okeyo describe women's groups that
emerged organically out of the needs of women-for
example, the provision of resources for which wom-
en are responsible in a household economy-and
suggest that the priorities of these groups have a very
different focus from those of conventional programs
directed toward women.
One would guess that a new emphasis on in-
come-generating projects will succeed only with a
better understanding of the complexity of rural
women's behavior, including the pressures they are
under within the family, the direct and indirect
means they use to produce resources, and the con-
straints on their efforts from larger economic forces.
The articles by Mazumdar and by Abdullah and

Zeidenstein analyze some of these issues in concrete

Concepts of rural development have thus far
failed to include the centrality of women's behavior
in predicting and explaining rural phenomena, and
planners have therefore not seen the need to under-
stand the frameworks governing that behavior in
order to formulate and carry out policy more effec-
tively. Thus, although research about rural women is
under way in many areas, there are strong indica-
tions that recognition of that research is not taking
place. One indication is that although the informa-
tion emerging from rural areas worldwide is being
disseminated, it is not being read by or at least not
incorporated in the thinking ot most rural planners.
With few exceptions, it has not entered the main-
stream of policy formulation. More likely it is toler-
ated as a fad or minimized in such handy but
misleading formulations as "beyond family plan-
ning" or even "integration of women in develop-
ment." Typically, research on rural women is
collected by a group isolated from the larger institu-
tion and relegated to a chapter of a report or an item
on an agenda without informing the thinking of the
rest of the report, meeting, or institution.
Another indication is that the study of rural
women does not generate the same kind of support
or carry the same status as, for example, migration,
household economics, nutrition, value of children,
and cropping patterns, even though an understand-
ing of rural women's role is a prerequisite to under-
standing each of these areas of study and most
others. The role of rural women is, appropriately, a
matrix issue, affecting what needs to be learned and
the interpretation of data on other issues. The result
of not understanding the role of rural women is that
knowledge accumulated on these issues has been
partial, as has development policy based on that
knowledge. Having measured only what men do,
our interventions seek to influence only men's be-
If there were recognition of women's coequal re-
sponsibility in the rural household and the conse-
quence of that role for rural dynamics, information
would be sought, incorporated in the knowledge
base for rural planning, and operationalized in rural
programs. Such conceptualization is implicit if not
explicit in all the articles and most of the research
notes in this issue, as it is in much of the new re-
search about rural women.
Learning about rural women is not mysterious
or exotic; it requires as sober an approach as any


other topic, although it plays a more central role than
most in explaining rural life. A great deal of informa-
tion is available, much of it collected and analyzed
since 1975 by and with the support of an interested
few, committed by an intuition of its central rele-
vance to rural development and then convinced em-
pirically by the data. Much more is now being
Unfortunately, the study of rural women labors
under several handicaps that have created a research
gap. One handicap that it shares with other areas of
rural studies is the relative lack of attention paid to
rural phenomena and the consequent lag in the re-
finement of social sciences, which evolved in urban,
industrial circumstances, to enable them to analyze
rural social and economic behavior. A matter of spe-
cial frustration is the inadequacy of the concepts,
terminologies, and methodologies-the tools of
research-of these social sciences for looking at and
describing rural women. For even as the social sci-
ences evolved to examine rural phenomena, they did
not appropriately conceptualize the role of women.2
The frustration is increased by the pressure from
some directions for hard data about rural women as a
basis for giving them serious attention even though
the instruments available for collecting hard data are
so flawed. The research gap has been aggravated by
the persistence, in the absence of more accurate data,
of urban, normative, male perceptions of rural
women, which conceal the relevance of rural devel-
opment issues to women's roles and hamper the col-
lection of necessary data through conventional
systems." And research has been hindered by the
protective invisibility of most rural women and the
scarcity of women researchers, for under these cir-
cumstances it is difficult to ascertain women's points
of view about their situation.
There is a need to close the research gap rapidly
so that rural dynamics can be better understood. To
do so requires the range of efforts and modes of
learning illustrated in Part II of this issue. Scholars
are working to correct comprehensive frameworks,
such as Marxism, and the biases of such social sci-

2The persistence of the terms productive and nonproductive to de-
scribe women's market-value and use-value work has been especially in-
'It is astonishing that the fact of women's participation in agricul-
ture-one of the most obvious phenomena of rural life-has had to be
proved in almost every country and then the nature of this participation
analyzed and its value measured.

ences as economics and anthropology. Researchers
are introducing in conventional frameworks of statis-
tics or sample surveys new analytic approaches and
new modules to collect more accurate information
about rural women. They are developing meth-
odologies to elicit new kinds of information and are
training interviewers in understanding new contexts
for asking questions. They are recognizing that the
lack of fundamental knowledge of the reality of
women's lives puts all attempts to study rural women
and men at a great disadvantage in regard to know-
ing what questions to ask, how to ask them, and how
to interpret answers. Therefore they are stressing the
continuing interaction of qualitative and quantita-
tive, micro and macro studies to reduce the chances
of misinterpretation of data or inappropriate conclu-
sions and policy recommendations. For example, one
can learn that rural women withhold water from
babies with diarrhea, conclude that folk medicine is
harmful and women ignorant, and recommend that
health systems be developed to replace women's tra-
ditional health care networks. Or one can look, as
Hull has in Indonesia, at the framework in which
women operate a traditional health care system and
recommend policy that will be much more acceptable
and practicable. Perhaps the greatest lack has been
the voices of rural women themselves expressing
their own point of view, which would expand enor-
mously our understanding of rural life. The in-depth
and open-ended interview is seen by many research-
ers as a very important tool in ending their long si-
Another source of information about rural
women basic to evolving an accurate framework for
understanding critical aspects of their behavior is the
institutions and projects that have already made con-
tact with them-the family planning clinic, the rural
health center, the functional literacy or agricultural
extension project. These institutions are in a unique
position to record the behavior of women in re-
sponse to new resources, always an indication of
what they consider to be priorities. Extension work-
ers could provide many of the answers researchers
are looking for if they are asked the right questions
and if their mode of knowing and explaining is val-
ued. In fact, without access to a record of what such
workers cannot help but know, but may not commu-
nicate in routine reports, one cannot understand
how programs and institutions need to be modified
to better serve the needs of rural women.

Part I
Looking at
Rural Women

Women, Doctors, and Family Health Care:

Some Lessons from Rural Java

Valerie J. Hull

The following article describes traditional health care sys-
tens of rural Indonesian women. It distinguishes the
central from th e peripheral aspects of syste('ns that women0
have developed to meet their own needs and conttrasts
them with modern health care systems. It ,.'..:.i what
can be done to keep the valuable features of each ap-

During the past two years the Population Institute of
Gadjah Mada University has been conducting a
longitudinal study of factors determining birth spac-
ing patterns in the subdistrict of Ngaglik, about 15
kilometers north of Yogyakarta in central Java.' Over
500 women have been followed through pregnancy,
childbirth, and the postpartum period, with inter-
views conducted twice monthly-once to obtain an-
thropometric data for mother and infant, and once to
obtain data on attitudes and behavior related to such
birth-interval determinants as lactation, amenhor-
rhea, and abstinence.
Through these frequent interview sessions, the
research uncovered a great deal about the main con-
cerns of women in Ngaglik. Our focus was not spe-
cifically on maternal and child health, yet intensive
interviews with pregnant women and new mothers
presented a unique opportunity to learn about these
matters. For example, a question on illnesses experi-

'The research was carried out in two villages (with a total population
of 15,799), which included areas of well-irrigated rice land as well as poorer
quality land suitable only for dry-cropping Most adults are engaged in ag-
riculture; however, there were also large numbers of traders and artisans, as
well as some civil servants who commute daily into the city of Yogyakarta.
Although the majority of the population is nominally Muslim, only a small
proportion strictly adheres to Islamic teachings.
The women selected for prospective interviews in the Ngaglik study
induded all those giving birth between January 1976 and March 1977. These
women were young, 50 percent of them between the peak childbearing ages
of 20 to 29. About a third have had no schooling, and just under 30 percent
have had only a few years of primary school. Average monthly household
income was estimated at approximately Rp. 10,000, or about US$25, and

enced during the month preceding each interview,
asked primarily to check the possible influence of ill-
ness on the anthropometric measures being col-
lected, yielded rich insights into women's per-
ceptions of illness and methods of treatment. A valu-
able supplement to our interview data was a system
of field notes kept by each interviewer. Conversa-
tions with village women encountered on the way to
and from interview sessions or while attending vil-
lage functions were noted on cards, which have been
labeled and subsequently filed according to topic.
The large proportion of these notes dealing with
health-related topics indicates the importance of
these matters in the women's lives. Open-ended in-
terviews were also conducted with the community's
traditional midwives, as well as with trained mid-
wives and local clinic personnel, in order to discover
more about their roles in the provision of health care.
In this article we present some of what we
learned about traditional beliefs and practices in ma-
ternal and child health care among women in
Ngaglik, and review some of the important changes
occurring as modern medical care is introduced. We
learned about positive and negative practices in both
traditional and modern medicine and considered
specific ways that this knowledge could be used to
improve health care programs. Perhaps more impor-

about half of all households owned no irrigated rice land, itself an important
indicator of economic and social status. The majority of women regularly
engage in economic activities in addition to household tasks, primarily
farming, farm laboring, and petty trading.
A community health center combined with a maternal and child
health center serves the approximately 40,000 people of the subdistrict of
N, iI' :I r i. I by paramedical personnel supported by weekly visits from
a doctor, the center is attended by an average of 20 to 25 patients a day
Women in our study lived anywhere from a few meters to seven kilometers
from this main health center, and also had access to one of several doctors
and trained midwives with private practices in the area. For a full descrip-
tion of the Ngaglik Study, including some preliminary findings as of Janu-
ary 1977, see Hull (1978).


tantly, we examined some of the underlying princi-
ples of the two approaches to health care. We found
that, contrary to the common image of peasant fatal-
ism in the face of disease, women played a resource-
ful, active role in family health care. In contrast to
this traditional self-reliance, modern health care
seemed to promote dependency, for medical profes-
sionals were providing services but not relaying

What Women Are Doing:
Traditional Health Care

Our examination of traditional health care turned up
many customs that are positive either in concept
(e.g., the value placed on child spacing) or in practice
(e.g., breastfeeding). We present a few of the beliefs
and practices dealing with childbirth, infant care,
and family planning that we felt were beneficial. We
follow this with a discussion of practices that could
be considered harmful.

Wisdom in Tradition

Childbirth. A woman giving birth in Ngaglik
will probably deliver in her own home, attended by
family members and a dukun bayi, an older woman
who performs the role of traditional midwife as a sec-
ondary occupation." During labor, the midwife
guides the woman's breathing, helps her control
contractions, and gives moral support. The woman is
usually advised to keep ambulatory during early la-
bor, and to get up and bathe as soon as she is able to
after the birth. Support from the dukun bayi con-
tinues for 35 days following birth, in the form of fre-
quent visits to give advice and to massage both
mother and infant. Overall, the atmosphere sur-
rounding childbirth and the supportive role of the
traditional midwife bring to mind much of what is
currently being advocated in the West by proponents
of natural childbirth and home delivery (Arms, 1975).
Some specific techniques of labor and delivery
may also serve as examples of traditional wisdom.
For example, contrary to modern medical practice,
the umbilical cord is not cut until the placenta, the
spiritual "younger sibling" of the infant, has been
expelled. Although usually based on the belief that
the disconnected placenta can migrate up into the

'Over 90 percent of our respondents delivered at home, and all but a
few with the assistance of a dukun bayi (Amin Yitno and Tri Handayani,

woman's body, this practice in fact ensures that the
baby will receive blood from the placenta for as long
as the umbilical cord pulsates, helping to prevent
anemia in the newborn infant.

Breastfeeding. Breastfeeding is still virtually
universal in Ngaglik, and infants are breastfed well
into their second year, often longer. The breastfeed-
ing mother receives support and advice from the du-
kun bayi and from experienced female relatives and
friends, a factor shown to be of critical importance in
the establishment and maintenance of successful lac-
tation (Raphael, 1976). The attitude of the women
themselves is positive: the importance of mother's
milk for infant health is widely acknowledged, and
milk is even thought to have curative powers for cer-
tain infections. Although women were not com-
pletely free of such problems as nipple soreness and
variable milk supply, such difficulties were usually
surmounted by the women themselves, again with
advice from elders and peers. Significantly, fewer
than 10 percent of the sample women reported hav-
ing experienced inadequate milk supply, though this
is one of the most common fears and frequent com-
plaints among urban, educated women in Indonesia,
as it is in the West. As with childbirth, many of the
breastfeeding practices of rural Javanese women are
now being taught as "wisdom" in developed coun-

Infant Health Care. A widely read Indonesian
journal, which devoted a special issue to health, in-
cludes the following statement: "It is not the usual
practice for the mother herself to either prevent or
treat her child's illnesses; she will probably bring the
child to a traditional healer" (Ryanto and Marsis,
1974, p. 58; translation mine). In contrast to this and
similar notions of passivity or reliance on traditional
healers, mothers in Ngaglik reported a wide range of
home remedies used to prevent and treat illnesses in
their children, and indeed all family members. These
consist of such cures as jamu, or herbal medicines,
which are either consumed as teas or used as com-
presses or salves; various oils; massage; and the ap-
plication of mother's milk, saliva, or urine. Some
cures are fairly simple, others involve the combina-
tion of as many as a dozen or more ingredients. Jamu
are by far the most popular, and there is a specific
mixture for almost every common ailment. Treat-
ment is sometimes "indirect"-for example, the
breastfeeding mother drinks the jamu specific to her
child's ailment, in the expectation that it will be
transmitted through her milk. Women either know
of these remedies themselves or seek advice from


family members. Only rarely did Ngaglik women
consult traditional healers, usually after their own
efforts had not produced the desired results, or for
some unusual symptom.
The wisdom in infant health care lies not in the
actual remedies, but in the principle of self-help
health care. Mothers are in fact following basic medi-
cal procedure: identifying symptoms, selecting the
appropriate cure that has been shown to be effective
previously, and applying the remedy. In the case of
Ngaglik women, at least, the image of women fatalis-
tic in the face of illness, or dependent on mystical tra
ditional healers, is a myth.

Family Planning. A strong traditional value is
placed on child spacing-three years being consid-
ered the ideal interval-surely an example of the
wisdom of tradition. Wide birth intervals have been
achieved traditionally through the effects of both lac-
tational amenhorrhea and postpartum abstinence.
Some women believe that intercourse can contami-
nate breastmilk; however, it is important to note that
child spacing is not merely a result of this taboo.
Women feel that pregnancies should be spaced for
economic reasons, children's health, and to facilitate
child care. Long periods of postpartum abstinence
have been supported by the high value Javanese cul-
ture and traditional religion place on all forms of ab-
stinence and self-control (C. Geertz, 1960; Singarim-
bun and Manning, 1974).

Some Harmful Practices

The preceding examples represent only a few of the
beliefs and practices we felt showed beneficial as-
pects. At the same time, descriptions of the positive
efforts of women in health care can sometimes tend
to overemphasize, even romanticize, the wisdom of
traditional ways, in complete disregard of the harm-
ful or extremely dangerous practices that are also part
of tradition. In our conversations with women we
learned about several practices related to childbear-
ing and rearing that would be considered harmful,
and collected information on the reasoning behind

Childbirth. Our earlier description of childbirth
focused on positive aspects of the delivery situation
and the advice of the midwife; however, certain
techniques used are potentially very dangerous.
While the belief that the placenta can go up into the
woman's body results in a delayed cutting of the um-
bilical cord-a positive effect-it can also prompt the
midwife to insert her hand to extract a placenta that

is unusually long in being expelled, a dangerous
practice that can lead to serious infection. In addi-
tion, the umbilical cord is traditionally cut with a
bamboo blade, and while some midwives claim to
use a freshly cut blade for each birth, others use old
blades that have been stored without regard to clean-
liness. This practice introduces the risk of neonatal
tetanus from a cord cut under septic conditions.

Breastfeeding. Certain breastfeeding practices
can also have adverse consequences. Colostrum, the
early milk that contains important antibodies and
nutrients, is believed to be "dirty" and to cause ill-
ness in the infant. It is thus discarded, often in the
same place as the child's placenta, which is cere-
monially buried shortly after birth. Over half the
women studied reported that they expelled and dis-
carded this important early infant food.

Supplementary Food. Another custom related to
breastfeeding is the practice of early supplementa-
tion with solid foods, often begun within the first
weeks of life. A child's appetite is a constant concern
for many mothers, and early supplemental feeding is
supposed to aid the baby in becoming accustomed to
solid food, which in turn will help to avoid later
feeding problems. Most infants, however, do not re-
quire nourishment other than breastmilk until the
age of 4 to 6 months, and giving this unnecessary
food to very young infants exposes them to an earlier
risk of infection through contaminated food. The in-
cidence of diarrhea, a serious health problem in In-
donesia, was found to be higher among Ngaglik
infants receiving supplementary food in the first few
months of life than among totally breastfed babies.

Pregnancy. Pregnant women perceive little
need for routine antenatal checks, as long as no un-
usual symptoms arise. Dukun bayi are not called
upon until labor actually begins, except to officiate at
any ritual ceremonies that might be conducted dur-
ing pregnancy, and this ritual care occurred only
among some primiparae. Ethnographic accounts
tend to portray gestation as a protected, fearful pe-
riod, with dietary and behavioral restrictions, such
as staying behind closed doors and windows after
sundown; yet most Ngaglik women were more mat-
ter-of-fact, carrying on activities as usual until late in
pregnancy. The symptoms of "morning sickness,"
unusual behavior, and food cravings were more
likely to be found among younger, educated women.
For most women experiencing a normal pregnancy,
the lack of antenatal checks may not entail undue
risks; yet it cannot be denied that maternal mortality,


fetal mortality, and the incidence of low-birth-
weight babies could be reduced by the identification
during pregnancy of conditions that would cause
these problems.
It is important that harmful traditional practices
not be viewed merely as irrational behavior, but
rather as having a basis within the woman's percep-
tions of relative risk and of causality. A clear under-
standing of these underlying factors is necessary to
any meaningful attempt to try to change this behav-

In Between: Some Harmless Practices

Obviously not all practices can be classified as either
intrinsically helpful or harmful. Many may have no
actual physical effect, although they are helpful psy-
chologically to those who believe in them. Par-
ticularly during childbirth, such customs as the
recitation of mantras by the dukun bayi or the posi-
tioning of the mother to face in a certain direction
during labor add to the woman's psychological pre-
paredness for delivery. Saving the dried umbilical
stub as one of the child's "protectors"-an indicator
of the perceived high risks to the infant and child-is
another harmless practice, as are such home reme-
dies as bathing eye infections with mother's milk or
scoring the skin with the edge of a coin (kerok), a very
common treatment for symptoms that the Javanese
describe as masuk angin (entrance of wind into the
body). Based on our study, we would also include as
harmless some of the dietary taboos during preg-
nancy, for they generally involved foods that would
not be eaten normally or those with little nutritional
value, such as cucumber. This is not to imply that di-
ets were adequate, for poverty prevented women
from eating many protein-rich foods.' In the case of
Ngaglik women, an assumption that insufficient nu-
trition during pregnancy is caused by superstitious
taboos would ignore the more basic underlying pat-
terns and their causes.4

"Our survey showed that, over all observation months for pregnant
women, over half (55 percent) of the women from the lower income group
reported that they ate no meat at all the preceding month, and 60 percent
did not eat eggs. Corresponding figures for pregnant women in the upper
income group were 20 and 24 percent.
4We should point out that in collecting information on dietary taboos
we asked specifically whether there was anything the woman normally ate
but (a) avoided or (b) limited her intake of during the interview month be-
cause she was pregnant. Had we simply asked "What foods are taboo during
pregnancy?" we would undoubtedly have obtained many more examples of
restrictions-for example, pineapple is a well-known taboo item. However,
women would rarely eat pineapple normally (it is an expensive fruit), so
this taboo does not result in changed behavior due to pregnancy, the variable
we were investigating.

Traditions of Unknown Effect

Finally, a number of health care practices are of un-
known efficacy-probably the most important being
the many kinds of jamu, or herbal medicines, that
enjoy enormous popularity among all classes of Jav-
anese. The large majority of women take jamu during
pregnancy (usually around 85 percent in any given
interview month) and during lactation (95 percent),
to improve breastmilk quality and quantity. Some
take mixtures purported to "bring on late periods,"
implicitly abortifacients. As mentioned above, jamu
are also an important home remedy for many infant
illnesses. We need to know more about the possible
value or harm of these mixtures. Among the few
studies already done, some effective pharmacological
agents have been identified. The patterns of taking
jamu also deserve attention, since many are con-
sumed regularly as preventive medicine, an impor-
tant concept in health care that might provide a basis
for generalization to other activities, such as vitamin
A campaigns.
It would also be useful to gain a more complete
understanding of other commonly used remedies,
including techniques of massage in general and a
specific type of massage performed by traditional
midwives to retroflex the uterus and prevent concep-

Diversity in Tradition

Most accounts of traditional practices tend to gener-
alize about beliefs and customs, giving the impres-
sion that these are found uniformly across the whole
society. In fact, peasant societies are rarely as homo-
geneous in either behavior or attitudes as some eth-
nographers would have us believe. There are always
contrasts across generations and by sex and social
class. Importantly, there is also individual variation in
custom and belief: not everyone follows societal
norms in the same way or to the same degree.
This flexibility in custom is both a cause and an
effect of social change. Despite the conservative con-
notation of the word "tradition," societal customs do
not remain static. In Java, as elsewhere, existing tra-
ditions incorporate a history of change. Elements of
an indigenous Javanese culture were overlaid by
Hindu, then Muslim influences. Years of Dutch colo-
nialism also had an impact in many areas of social
and economic life, and today elements of westerniza-
tion are evident in even the most isolated village.
Present traditions are changing in both form and
content. New products and ideas entering the village
are often combined with time-honored ways of
doing things. Some dukun bayi today, for example,


may use a razor blade to replace the welat, or bamboo
knife, that was used to cut the umbilical cord.
A very significant change in tradition has been
an apparent decline in the practice of prolonged
postpartum abstinence, a phenomenon associated
with wider changes in the marital relationship and
attitudes toward sex. Because the traditional wisdom
of child spacing through abstinence is not always re-
placed by reliance on modern forms of birth control,
birth intervals tend to be shorter among young edu-
cated women (Hull, 1975; Hull et al., 1976).

Self-Reliance in Traditional Health Care

Writing on maternal and child health care in devel-
oping countries, Williams and Jelliffe (1972, pp.
84-85) made the following observation on the lives
of women in preliterate societies: "Traditional prep-
aration for motherhood is often more appropriate in
preliterate villages than in a sophisticated urban con-
text . the preliterate mother normally takes mar-
riage, pregnancy, lactation, hard work, child care
(and often child mortality) in her stride." Without
idealizing the situation of village women-for in-
deed their choices are limited, and part of their pres-
ent self-reliance has been forced upon them by their
conditions of life-many observers have admired the
self-reliance of the Javanese woman, her autonomy
within the family, and her economic independence
(H. Geertz, 1961; Mintz, 1961; Smith, 1961). Tradi-
tional health care among women in Ngaglik is an im-
portant aspect of this self-reliance. Specialized
traditional healers do exist, but they are generally
consulted only after one or more of the wide array of
home remedies have been tried. Health, including
childbirth, infant care, nutrition, and contraception,
is a matter to be met and dealt with at home, with the
advice and support of family members. Although
specific practices are not always effective and some
customs are varied and changing, the one important
principle underlying women's traditional approach
to health care is their attempt to deal with their en-
vironment within the very real limitations imposed
by poverty and lack of medical knowledge. Can mod-
em health care build on the kind of self-reliance we
found among Ngaglik women?

What Doctors Are Doing:
Modern Health Care

Recent national development plans place strong em-
phasis on rural outreach in health, with the first-year

plan (PELITA I) establishing over 2,500 rural health
centers, and another 900 planned under PELITA II
(Subyakto, 1974, p. 51). There are inoculation pro-
grams in the village to immunize young children
against smallpox and tuberculosis. The government
is also actively training dukun bayi in modern deliv-
ery techniques and in family planning motivation.
The national family planning program is working
toward a goal of "institutionalizing the small family
norm and contraceptive practice" in every Indone-
sian village. The program relies on 7,000 family plan-
ning fieldworkers who make house-to-house visits
and supervise thousands of acceptors' clubs in Java
and Bali, and works through local officials and accep-
tors' clubs in the outer islands.
The problems of providing health services are
formidable in a country of 140 million people but
with only 6,000 doctors and 14,000 paramedics, most
of whom are concentrated in the cities. Despite the
goal of rural outreach, the approach has been largely
passive and based mainly on the subdistrict health
center, or PUSKESMAS, which serves an average of
40,000 people. These health centers have been vastly
underutilized (Sulianti, 1974), a fact frequently at-
tributed to social traditions:
It is clear that . .societal traditions interfere with
modern medicine. The preference for consulting du-
kuns results in patients either not coming to a health
center, or coming too late . Before la patient is]
brought to a hospital there must also be lengthy con-
sultation among members of the extended family,
which further postpones going to the health center
[Hashem, 1974, p. 69; translation mine].

Underutilization and Mixed Blessings

Many women in Ngaglik were in fact reluctant to at-
tend the rural health center, although not necessarily
because of a "preference for consulting dukuns."
Modern health services, not surprisingly, are built
largely on Western models. Patients coming to a
clinic sit on hard benches (whereas most villagers are
accustomed to sitting on the floor), waiting their turn
to be called by health personnel whose middle-class
origins are evident in their dress, bearing, and
speech patterns. A mother bringing her child to the
health center is unlikely to be given an explanation of
her child's ailment or its treatment, yet may be told
that she was wrong to try to treat the child herself.
Clinics are usually open only in the mornings, the
busiest time for women. Services are free only to
those who present a letter from their village adminis-
trator attesting to their inability to pay. Obtaining
such a "poverty certificate" is far too humiliating for


most people. Even transportation costs to the clinic
are often beyond the means of village families. Per-
haps mainly for these reasons, the clinic approach to
health services is not effectively reaching Javanese
Unfortunately, some of the existing outreach
efforts in health care have met only limited success.
In Ngaglik, for example, most of the dukun bayi do
not want to attend the government training course,
some of their comments being, "I'm illiterate, I'd be
embarrassed in front of everybody"; or "The clinic
where the course is held is too far away" (Amin
Yitno and Tri Handayani, 1978). Some of those who
did receive training were visited by the instructor af-
ter completion of the course, and their equipment
kits taken away from them. The instructor claimed
that the kits were not being cared for properly. This
action resulted in ill feeling and ended the dukuns'
willingness to learn modern practices and to cooper-
ate with the trained midwife.
Despite the low overall rates of clinic attendance
and the limited success of such efforts as the midwife
courses, modern medicine has had an observable im-
pact. The inoculation campaign in Ngaglik has
achieved fairly wide coverage, most mothers readily
taking advantage of the free injections offered in
their village. Trained medical practitioners assisted
in nearly 10 percent of the deliveries experienced by
women in our study. And people are beginning to
add simple patent medicines and injections to the
various remedies attempted.
Modern medicine has undoubtedly benefited
some villagers: pneumonia victims have been saved
by penicillin; women with prolonged labor have
been helped by trained midwives or doctors; some
tuberculosis cases are under treatment. The blessings
of modern medicine, however, are mixed. The intro-
duction of benefits must be weighed against ques-
tionable and even harmful effects, some of which we
now discuss.

Bottlefeeding. "Are you going to give us pow-
dered milk?" was a frequent query when we began
our monthly weighing sessions of mothers and in-
fants. The World Food Program's powdered milk
supplies have been distributed through the maternal
and child health clinics in Indonesia, a fact widely
known in the villages. The milk is given primarily as
an incentive to get patients to come to the clinic, and
women are generally not given full information
about either the dangers of bottlefeeding or the ben-
efits and techniques of breastfeeding.
In the city, modern medical practitioners often
give incorrect advice about and little support for

breastfeeding, so that women who want to begin or
continue breastfeeding sometimes fail. This would
have potentially tragic consequences in rural areas
such as Ngaglik, where the constraints of poverty
preclude safe and high-quality infant feeding sub-
stitutes. Still, there are early signs that bottlefeeding
is becoming known, and infant formula is clearly
seen as a high-status food. In the early months of in-
terviewing, when all of the infants were under a year
old, anywhere from 10 to 15 percent had their breast-
feeding supplemented with bottles of infant formula
or skim milk.

Hospital Care. "I kept hearing my baby cry in
the nursery, but they wouldn't bring her to me" was
the complaint of a young woman who gave birth to
her first child at the local clinic. Although rooming-in
is practiced in some maternity clinics, separation of
mother and infant is the more frequent arrangement.
This practice can interfere with initiation of lacta-
tion, an unfortunate result that has been reported in
Indonesian cities (Poernomo, 1978).
"When my baby went into the hospital with di-
arrhea, they told me I couldn't breastfeed him." This
restriction, which contradicts sound medical prac-
tice, was imposed on one of our research assistants.
Mothers are also forbidden to stay at a sick child's
bedside. In Indonesia, as elsewhere, hospital regula-
tions are frequently based more on staff convenience
than patient welfare.

Prescription Drugs. "I'm taking little yellow
pills and large brown pills." Women in Ngaglik who
were taking modern medicines seldom knew the
name of the medication or its purpose. Pharmacies
do not write the names of drugs on prescription or-
ders, and doctors do not explain to patients what is
being prescribed. One woman was given a potent
fertility drug and told it was "to make her body
healthier." There is inappropriate use of antibiotics:
in the city, especially, it is not unusual for a child
with a mild cold to be given tetracycline or other
antibiotics. People sometimes purchase only partial
dosages of prescriptions when they cannot afford the
full order, and patients are given prescriptions for
expensive medication when cheaper substitutes are
available. These near-universal problems associated
with prescription drug use are obvious in the cities,
but are also beginning to appear in rural areas.

The Magic Injection. "I took my child to the
clinic, but they only gave him medicine, no suntik
[injection]. Next time I'll take him to Dr. S., he al-
ways gives a suntik." Injections are very popular, in


part stemming from the yaws campaign of the 1950s
in which penicillin injections produced dramatic and
widespread results. This popularity often leads med-
ical practitioners to give injections (usually of vita-
min B complex) where not medically indicated. "I
paid 300 rupiahs for my injection at the clinic, but
my neighbor paid only 200 rupiahs. Why was I
charged more?" Many people do not understand that
the contents of injections vary, for the cure to them is
in the suntik itself.

Expropriation of Health

Modern forms of treatment can alleviate many of the
illnesses suffered by Ngaglik women and their chil-
dren. As noted by Illich (1975) and others, this new
knowledge is becoming the exclusive province of
modern medical practitioners. Women now know
that, even if they themselves cannot afford the serv-
ices, doctors and clinic workers can work minor mir-
acles with their suntik and other forms of tech-
nologically advanced care. These modern treatments
are the new magic and are in fact stimulating irra-
tional belief to an extent seldom found in traditional
medicine. Unlike herbs and most other home reme-
dies, which are used according to sound principles of
applying appropriate known cures for specific symp-
toms, a suntik may be sought for everything from
eye disease to malnutrition. The doctor, in many
ways, is the new mystical healer: someone with spe-
cial powers whose orders are followed on faith. Even
the traditional healer is more likely to offer an expla-
nation for the cause of an illness than is the modern
The almost blind faith that characterizes the pa-
tient-doctor relationship in many developed coun-
tries is now becoming evident among educated,
urban Indonesians. Urban women increasingly rely
on the services of modern medical practitioners, not
only for illness but also for advice on such matters as
infant feeding and other aspects of child care. Unlike
most women in Ngaglik, the daughters of the urban
elite have not been trained since childhood for their
motherhood role. Once married, they often live away
from parents and members of the extended family
who were formerly available to advise as the need
arose. The medical professional-usually a male-is
called upon to fulfill the roles once held by these fam-
ily members. This is usually seen as a sign of prog-
ress, a valid assumption only to the extent to which
modern medical care is more technically effective
than traditional care. In other cases, however, re-
liance on modern medical advice may result in de-
creased chances of success (as is the case with

discouragement of breastfeeding) or even in new
forms of danger (such as the overuse of drugs).

What Women and Doctors
Can Learn from Each Other

Building on Self-Reliance

The custom of self-treatment, the idea that health is a
family matter, and the consulting of dukuns are all
seen as barriers to implementing modern health
services. Because traditional cures are often ineffec-
tive or harmful, the implication usually drawn is that
we should make health care the sole responsibility of
those with the necessary knowledge-the medical
We have found, however, that there is much that
is positive in the traditional approach to health care.
Such practices as home delivery and breastfeeding
and the principle of self-help health care are now
being advocated even in countries with a high de-
gree of modern technology and availability of health
professionals. In Java, where modern health services
are severely limited, the growing reliance on pro-
fessionals seems distinctly inappropriate. Modern
health programs can achieve a wider and more last-
ing impact by providing knowledge in addition to
services: teaching people how to prevent illness, to
make home care more effective, and to recognize
those problems for which professional help must be
sought. This kind of approach led to the training pro-
gram for traditional midwives, but it can be taken
much further-to everyone in the community:
"Some doctors talk about self-care as if it were dan-
gerous. . But in truth, most common health problems
could be handled earlier and better by people in their own
homes" (Werner, 1977, w2; emphasis in the original).
A shift in priority from providing health services
to teaching is not a revolutionary idea (the word
"doctor" comes from the Latin docere, "to teach"), yet
it is sometimes overlooked in discussions about
building health care infrastructure, importing mod-
em drugs, and offering curative services based on
sophisticated technology.
Health education was in fact the basis of an
active hygiene and public health program in Indo-
nesia in the 1920s and 1930s, and the philosophy be-
hind that program has much to recommend it today:
Doing things to people is often easy, but it is expen-
sive and of temporary benefit. Showing people how to


do things for themselves may take a little time, but it is
relatively inexpensive and its results are lasting. More-
over the people are strengthened by the latter process
and weakened by the former [Turner in Hydrick, 1937,
p. 6].
The resurrection of this approach may come in the
form of current plans for the expansion of a village
health worker (kader) program, which will help pro-
vide health training at the village level. Several
schemes already function in various parts of Indo-
nesia, and the government is considering making the
training of volunteer village health workers a part of
the Third Development Plan. Current programs vary
(e.g., some kader are salaried, some are volunteers),
but they all represent a first step toward grass-roots-
level health workers on a large scale, and as such the
first stage in bringing health education to the people.
One major scheme, for example, proposes to train
family planning fieldworkers in basic health and nu-
trition so that they can train village mothers. Writing
of this proposed program, Rohde and Northrup
(1978, p. 154) have described what will be, in effect,
the recruitment of millions of Indonesian mothers as
health workers:
Rather than viewing the mother as a target of health
propaganda, as an unthinking recipient of directives
from medical professionals, this approach makes her a
fully integrated basic health worker, whose task it is to
carry out certain health interventions and to initiate
where necessary the referral process that will carry her
and her child to the appropriate level of competence
within the system. This is real integration where it
The ultimate impact of such efforts will depend
largely on the approach used in communicating
knowledge. The approach of the dukun bayi, as op-
posed to that of many trained midwives, is of a fel-
low villager, not unlike her patient in social status,
who takes time and shows sympathy for the woman
she is helping. With the expansion of the kader pro-
gram, Indonesian health services have an opportu-
nity to establish direct communication with women,
the coordinators of family health care. If village
health workers can maintain a status as sympathetic
fellow villagers, they will be a key element in the
spread of knowledgeable self-help health care.

Building on Current Practices:
Ideas for Local Programs

Two years' experience with women in Ngaglik not
only convinced us that self-help health care should
receive priority, but also yielded specific ideas for

the design and implementation of local MCH pro-
grams. The following points illustrate how a few
ideas drawn from our discussions of traditional and
modern health care might be translated into policy

Antenatal Care. In addition to the usual barriers
to clinic attendance-cost, time, distance, status dif-
ferentials-women in Ngaglik perceived little need
for regular visits to a clinic during pregnancy. A
health program might achieve better results and
more efficient use of its resources if it determined its
priorities in line with the community's perceived
needs. Thus, rather than establishing a complete sys-
tem of clinic-based antenatal care, a more modest an-
tenatal program could be planned, which could
eventually be developed into a more regular program
as women begin to demand it.
The system might involve only a single visit for
every woman who thinks she is pregnant. The visit
would be used to confirm the pregnancy and to
screen for high-risk women (primiparae and those
with small stature, severe anemia, or a history of pre-
vious birth difficulties), who would then receive pe-
riodic monitoring by village-level health workers.
Inadequate nutrition is a major problem for pregnant
rural women. Thus on the same visit a nutritional as-
sessment could be made, and women given simple
messages for basic nutrition using cheap, locally
available foods. ("Eat an extra handful of rice a day,"
which has been estimated to provide an additional
25,000 calories during pregnancy, is one example.)
Tetanus injections to prevent neonatal tetanus in the
child could be administered on this same visit.
Women would also be instructed about such danger
signs as swelling or vaginal bleeding, for which they
would report back to the clinic.

Childbirth. Modern medical practitioners can
learn a great deal from the approach of the dukun
bayi. She is a trusted village resident who gives the
needed attention and moral support during delivery
and at frequent home visits following birth. Clinic
and hospital services might well improve attendance
if their routines were flexible enough to allow for
much of what is good in home delivery, including at-
tendance by relatives and rooming-in of infant with
At the same time, modern medicine has much to
teach traditional midwives in more hygienic and effi-
cient delivery techniques. The current dukun bayi
training program might be more effective if some of
the following strategies were tried:


Locating the course in a modest (nonelite) vil-
lage home rather than at the clinic, with lessons on
hygiene and aspects of delivery technique being
geared to the village home environment. Sympathet-
ic trainers, perhaps a dukun bayi who had suc-
cessfully completed a previous course, could instruct
under the guidance of clinic personnel.
Incorporating beneficial traditional practices,
such as delaying the cutting of the umbilical cord.
Traditional midwives may be more receptive to a
training course that includes or at least acknowledges
these familiar practices.
Retaining a neutral attitude toward such
harmless practices as the recitation of mantras.
Providing sensitive and clear explanations of
why some current practices are harmful, and of what
safe practices may be substituted.
Establishing continued two-way communica-
tion between instructor and dukun bayi; there are
undoubtedly many ideas to be learned from her.

Breastfeeding. The importance of breastfeeding
is recognized by the national health department and
is being encouraged with a campaign that uses the
approach: "Continue the tradition of breastfeeding
your baby." However, individual medical practi-
tioners must be re-educated so that they are able to
give women the necessary support and knowledge to
successfully establish and maintain lactation. Wom-
en also have the right to complete and accurate infor-
mation on alternative feeding methods, information
free of the influence of commercial concerns.
Based on the pattern we observed in Ngaglik,
the negative practice of discarding colostrum might
be amenable to change through a well-designed in-
formation campaign. This custom is not found
among all women, and even for those who did dis-
card colostrum nearly 20 percent could not state a
reason, but were only following what they heard was
the "right" practice. A large minority of women do
feed colostrum, and they could stand as graphic ex-
amples that this practice does not result in harm.
There is a need for research into the ubiquitous
herbal mixtures taken by lactating mothers. Their
main effect may be psychological; they may indeed
include a galactogogue; or their primary benefit may
lie in the extra daily fluid intake for the lactating
woman. It is apparent that, whatever the mech-
anism, women report the desired results. A practice
this widespread and reportedly effective cannot be
ignored by modern health care programs.

Infant Health Care. It is in the field of infant
health, an area of constant concern to Ngaglik
women, that training the mother in basic preventive
and curative health care may have the greatest im-
pact. Women can learn to recognize early signs of ill-
ness, give basic curative treatments in the home,
provide adequate nutrition, monitor child develop-
ment, and improve the home environment to pre-
vent illness.
Child weighing programs and health cards (re-
cording children's monthly weights and periodic in-
oculations) are simple but valuable ways for the
mother to follow her child's development. Ngaglik
women responded enthusiastically to the regular
child weighing during our research. It gave them an
objective and readily understandable measure of
their children's progress: continued weight gain was
a good sign, faltering weight a cause for concern.
Comments such as "My child gained 800 grams this
month!" or "My child has had a bad cold-I'm sure
her weight gain won't be as good this time as it was
last month" were often heard at weighing sessions.
The monthly gathering of women at these sessions
would provide a valuable opportunity to provide
health care training, as well as direct intervention by
village health workers.

Family Planning. Family planning fieldworkers
in Ngaglik have already shown the kind of flexibility
required of those who work in this type of program
in rural areas. By talking with women, they realized
that a number of couples eligible for accepting con-
traception were in fact abstaining from sexual rela-
tions for extended periods, and thus could not be
motivated to accept modern contraceptives. How-
ever, their superiors-who did not have this same
knowledge of women's behavior-continued to ask
for higher proportions of new acceptors. Fieldwork-
ers solved the problem by providing couples with
condoms, "just in case" they resumed sexual rela-
tions, and registering them as condom acceptors.
With the apparent decline in the practice of ab-
stinence among young educated women, field-
workers will have increasing responsibilities for
providing accurate information and services to these
women. Their role should also be to convey what
they learn from women to the program administra-
tors, so that policy can be responsive to needs.
Fieldworkers have also expressed to us their de-
sire to provide services other than family planning.
They are frequently asked for advice or medicine for
sick children, and some have begun supplying sim-
ple medicines on their own. The 7,000 family plan-
ning fieldworkers currently making house-to-house


visits to all women in the reproductive ages are a
valuable potential resource for the provision of basic
health training.


Women in Ngaglik have always been active partici-
pants in family health care, using whatever limited
resources are available to them. Most modern health
care programs, perceiving traditional patterns as ob-
stacles to development, work to remove health care
from the responsibility of the woman and her family
and to place it in the hands of professionals. The em-
phasis is thus on providing health care services.
An alternative approach is to provide women
with the knowledge to more effectively meet their own
and their family's health care needs, with profes-
sional services providing appropriate support. Our
two-year research project among Ngaglik women
showed us many ways in which modern programs
can benefit from an understanding of traditional
health care: by building on the basic approach of
self-reliance, by incorporating specific traditional
practices that are beneficial, and by sympathetic
treatment of those traditional practices that are inef-
fective or potentially harmful. We became aware of
the complexities of studying traditional health care,
yet convinced that careful and sensitive study of
what women are currently doing is essential to the
design and implementation of effective modern pro-


Amin Yitno and Tri Handayani. 1978. "Hasil
penelitian dukun bayi" (Results from a study of tra-
ditional midwives). Seminar No. 83. Yogyakarta:
Population Institute, Gadjah Mada University.
Arms, Suzanne. 1975. Immaculate Deception: A
New Look at Women and Childbirth in America. Boston:
Houghton Mifflin.
Geertz, Clifford, 1960. The Religion of Java. Glen-
coe, Ill.: The Free Press.
Geertz, Hildred. 1961. The Javanese Family: A
Study of Kinship and Socialization. Glencoe, Ill.: The
Free Press.
Hashem, O. 1974. "Peranan dokter dalam pen-
ingkatan masyarakat: Pengalaman seorang dokter di
daerah" (The role of the doctor in improving commu-

nity health: Experiences of a country doctor). Prisma
3, no. 5: 64-74.
Hull, Valerie J. 1975. "Fertility, socioeconomic
status, and the position of women in a Javanese vil-
lage." Unpublished Ph.D. thesis, Department of De-
mography, Australian National University, Can-
--- 1978. "A study of birth interval dynam-
ics in rural Java." In Nutrition and Human Reproduc-
tion. Ed. W. Henry Mosley. New York: Plenum.
-- Kodiran, and IrawatiSingarimbun. 1976.
"Family formation in the university community."
Report Series No. 9. Yogyakarta: Population In-
stitute, Gadjah Mada University.
Hydrick, J. L. 1937. Intensive Rural Hygiene Work
and Public Education of the Public Health Service of
Netherlands India. Batavia-Centrum, Java. (Includes
quotation from a letter by Dr. C. E. Turner, chairman
of the Health Section of the World Federation of Edu-
cation Associations.)
Illich, Ivan. 1975. Medical Nemesis: The Expropria-
tion of Health. London: Calder & Boyars.
Mintz, Jeanne. 1961. Indonesia: A Profile. Prince-
ton, N.J.: Van Nostrand.
Poernomo, S. S. 1978. "Dicari: Rumah bersalin
pendukung ASI" (Wanted: A maternity hospital
which supports breastfeeding). Femina, no. 137: 161.
Raphael, Dana. 1976. The Tender Gift: Breastfeed-
ing. New York: Schocken Books.
Rohde, Jon R., and Robert S. Northrup. 1978.
"Mother as the basic health worker: Training her and
her trainers." In The Changing Roles and Education of
Health Care Personnel Worldwide in View of the Increase
of Basic Health Services. Ed. Ronald W. McNeur. Phil-
adelphia: Society for Health and Human Values.
Ryanto, H., and I. V. Marsis. 1974. "Masalah
kesehatan di Kampung-kampung kota besar" (The
problem of health in city kampung [slums]). Prisma
3, no. 5: 56-63.
Singarimbun, Masri, and Chris Manning. 1974.
Fertility and Family Planning in Mojolama. Monograph
Series No. 1. Yogyakarta: Population Institute, Gad-
jah Mada University.
Smith, Datus C. 1961. The Land and People of Indo-
nesia. Philadelphia: J. B. Lippincott.
Subyakto. 1974. "Pusat kesehatan masyarakat:
Konsep dan pembinaannya" (The rural health cen-
ter: Concept and practice). Prisma 3, no. 5: 51-55.
Sulianti Saroso. 1974. "Politik kesehatan nasio-
nal" (The politics of national health). Prisma 3, no. 5:
Werner, David. 1977. Where There Is No Doctor: A
Village Health Care Handbook. Palo Alto, Calif.: The
Hesperian Foundation.


Williams, Cicely D., and Derrick B. Jelliffe. 1972.
Mother and Child Health: Delivering the Services.
London: Oxford University Press.

ACKNOWLEDGMENTS The data presented in this
paper represent the work of a team of researchers
associated with the Population Institute of Gadjah

Mada University who have worked with this project,
including Amin Yitno, Terence Hull, Elizabeth
Meyer, Jon Rohde, Suprawardhani Suryono, Tri
Handayani, and others. I acknowledge their contri-
butions, while accepting responsibility for any errors
or misinterpretations. I would also like to thank
Terence Hull, Marie Claire Malingreau, Nancy Pel-
uso, and Jon Rohde for their valuable comments on
an earlier draft of this paper.

Rural Women's Credit Systems:

A Nigerian Example

Kamene Okonjo

The following article looks in detail at a rotating credit
system used by rural women of Nigeria. This system,
which women use to ensure the availability of cash to
meet their family's financial needs, has analogues
throughout the world. This close study of behavior-in
this instance, about how women form groups to save and
why-reveals rural women's roles in the family and the
economic priorities created by these roles. This kind of in-
formation, rarely recorded or attended to, is an indispens-
able base for learning about how women are affected by
and affect development policy. Further, understanding
the mode of operation of this credit system and its value to
the women who organized it provides guidance for the de-
sign of projects that work and are meaningful to women.

It has become increasingly recognized in the last ten
years or so that growth in the developing countries is
not so much a problem of lack of resources for invest-
ment as it is a problem of mobilizing these resources
-both human and material-and apportioning them
rationally. Attention has thus been focused on the
rural areas of developing countries, for the wide-
spread poverty of their populations constitutes the
critical problem of development. Since it has become
evident that centralized decision making, with its
limited knowledge of local conditions, can play only
a minor role in stimulating development in small
rural communities, the conduct of micro-studies be-
comes very necessary. Through such studies, con-
straints on small communities pursuing self-reliance
will become more widely known.
In an effort to reappraise the role of African
women in traditional life and to evaluate their role in
development, this study examines the contribution
clubs devised to assist members in small-scale capital
formation in two rural communities of Nigeria.' The

uses to which such pooled monetary contributions
are put, as well as their social significance, are also
discussed. The particular financial institutions se-
lected for study are generally known as Esusu in most
parts of Nigeria and as Otu-Utu or Otu-Ofu in the
two rural communities studied. Although associa-
tions organized solely for men or as mixed associa-
tions of men and women (with men in the majority)
exist in both communities, the present study concen-
trates on women-dominated clubs. Such associations
exist widely, not only in Nigeria and West Africa,
but also in such countries as China, India, Indonesia,
Malaysia, South Africa, and Vietnam (Ardener, 1964,
p. 51).

The Communities

The towns of Obamkpa and Ogwashi-uku are Igbo
towns situated on the western side of the Niger
River, in the Bendel State of Nigeria. In 1972 the two
towns had a combined population of around 23,000
inhabitants, with over 20,000 of these living in Og-
washi-uku, while just over 2,000 lived in the rural
community of Obamkpa, situated in an inland valley
lacking all modern amenities. Ogwashi-uku had a
few scattered standpipes for water in some of its nine
villages and no electricity. As the district headquar-
ters of Asaba Division, it contained the district office

'Material on the contribution clubs was gathered during my field-
work in the towns of Ogwashi-uku and Obamkpa in 1972. More recent
fieldwork, conducted in 1976, has revealed many changes that have been
reflected here. One such change is the conversion of the national currency
from pounds, shillings, and pence to naira and kobo. Meeting days are
also now mostly on Sundays, while weekdays and Saturdays are reserved
for economic pursuits.


and a few other government offices, an old post of-
fice, but no banks. The nearest banks were 15 to 20
miles away.
Since 1972 the situation in Ogwashi-uku has
changed dramatically. It now has more standpipes
for water, electricity under the rural electrification
scheme, more offices, shops, and schools, and at
least one bank. Traditional life still predominates in
both towns, although signs of social change are
everywhere more evident in Ogwashi-uku. Small-
scale, nonmechanized farming is the main occupa-
tion in both towns. What is not consumed locally is
sold, and considerable trade exists at intervillage, in-
terstate, and even interregional levels. This is espe-
cially true for Ogwashi-uku, which has a large and
successful market and several adequate roads linking
it with other towns. In both towns farming is an oc-
cupation mainly for men, with the participation of
women being confined mainly to growing such
"women's crops" as cassava, cocoyams, okra, pep-
pers, and pumpkins. Among the Igbos on the west-
ern banks of the Niger, in marked contrast to those
on the eastern side, weeding and hoeing are not con-
sidered jobs for women, and men either hoe and
weed the farms themselves or hire laborers.
Women, on the other hand, have their own oc-
cupational activities-mainly cloth and mat weaving
and processing cassava meal and palm oil. They also
engage in petty trading, which ranges from selling
foodstuffs and retailing local factory-manufactured
and imported goods to running beer and palm-wine
parlors and setting up locally as seamstresses. But
whatever work women engage in, they are expected
to share the burden of meeting the family's expenses
and to provide for a great part of the family's daily
food and other needs. This requires access to capital
or credit, sometimes at very short notice, and the
western Igbo woman is expected to obtain this capi-
tal or credit herself.
Of course, a woman's husband may provide her
with working capital in the form of money, but such
cases are the exception. The picture, more often, is
that of a woman who struggles on her own, from the
moment of her marriage, to meet her needs and those
of her family. She may continue to practice a trade
while living with her parents, or may start a small
business with some accumulated capital or with
money received from her parents. Usually this initial
capital is quite small, and she has to devise other
means of ensuring a "source pool" upon which she
can fall back in a time of financial need. One such
agency or institution to which she resorts and that
enables her to accumulate capital, albeit on a small
scale, is the contribution club.

Nature and Organization
of Contribution Clubs

The contribution club is a traditional association
whose aim is to assist its members in small-scale cap-
ital formation. Fixed payments are required at short,
regular intervals, usually on a four-day, weekly, or
fortnightly basis. The money thus collected is then
pooled and given to members, one at a time, on a ro-
tating basis.
Contribution clubs fulfill for the community
some of the functions of a bank or an insurance soci-
ety, for through such an association not only can
small-scale capital formation take place and savings
accumulate, but loans can be made or goods bought
on credit. They also offer an easy and comparatively
ready access to money in the event of a personal
crisis or misfortune. The club's success depends very
much on the existence of safeguards for ensuring that
each member continues to pay her subscription until
every member has received her "take-out" (see Ar-
dener, 1953, p. 129).
Most women whom we interviewed in both
towns-about 95 percent of the 200-belonged to at
least one contribution club. Many belonged to two or
more. The 5 percent who did not belong were mainly
from the age group 15-18 of unmarried or young
married girls, as well as a few old women. It is highly
unusual for an adult woman in these communities
not to belong to any contribution association, since
such inaction on her part could be interpreted as lazi-
ness, irresponsibility, or even as antisocial behav-
ior-any of which opinions could injure her good
The contribution club is formed through the ini-
tiative of a woman who talks to and organizes other
women. If the initiator is a woman of sufficient social
standing, her invitation to other women to join as
members will meet with a positive response. If the
initiator is not a woman of sufficient social standing,
she often persuades another woman of such standing
to assume the role of initiator. A sum is agreed upon
and members contribute this amount regularly.
Usually the sum agreed upon will not place an
undue strain on the contributor. In any case the
choice to join or not to join is still the woman's. We
discovered that most of the contribution clubs
charged 20 kobo, or 35 cents; a few charged 50 kobo,
or one naira, while one or two charged more than
one naira. Many women felt that it is better to con-
tribute a small sum at close intervals than to contrib-
ute a large amount at longer intervals. Contributing
on a weekly or fortnightly basis enables the women


to use the small profits they make from their sale of
cassava meal, palm oil, or woven mats before these
are spent in other ways.
Members are usually given their loans according
to an order agreed upon at the founding of the club.
The arrangement may depend on the order in which
members joined the association, with the founder at
the head of the line. But this order may be disrupted
in cases of proven emergency. A member who sud-
denly finds herself in dire financial straits will be
allowed to receive her take-out before those ahead of
The club's founder becomes the Nne-Otu
(mother of the association) or, more often, the Eze-
Otu (queen or president of the association) and is ad-
dressed as such when the members meet. A contri-
bution club always has an inner executive made up
of its officers, but the president makes the final rul-
ing on major issues. She is looked up to by the mem-
bers, and the club's prestige depends greatly on her
effectiveness as a leader. She must be firm but kind
and must be acknowledged by all as a person of very
high integrity. Such a person can always find a fol-
lowing, and it is not uncommon for a woman of this
stature to found more than one contribution associa-
tion. We discovered a few such women in both
towns. One woman, the first wife in a polygamous
household of five wives, had initiated four contribu-
tion associations, one of which was made up solely
of her cowives and other lineage wives.
In contrast, a thief, liar, or bad debtor cannot
successfully found a credit association regardless of
how genuine her intentions are. Other women will
assume that she needs money and will default as
soon as she has collected her take-out. A "lazy"
woman who has no visible means of income will
usually also fail in her attempt to gain a following.
But screening in a contribution association is a
two-way game. The founder and/or initiator of a
club will also carefully screen applicants. Potential
members with dubious character or those whose
credit worthiness is not beyond reproach will not be
admitted, for fear that they will default in their pay-
ments. In fact, recent investigations reveal that all
would-be members are now required to have a
surety who will be answerable to the association for
their conduct.

Male Roles in Contribution Clubs

Although we searched very hard, we did not encoun-
ter more than three wholly female contribution

associations. Women who said they initiated or be-
longed to wholly female associations actually meant
that their associations were female dominated. One
association with 60 members, for example, had 16
males, while another with 340 members estimated its
male membership at 80. This observation is true of
most of the other associations, which generally have
a female-to-male membership ratio of about 4 to 1.
Questioned on why they could not run all-female
clubs, officers of these associations said that they
needed male musicians to beat the drums, blow the
trumpet, and generally provide the music at meet-
ings. In addition, eight of the initiators interviewed
believed that the presence of males also helped to
maintain law and order ("the eyes of a man drive
away thieves").
Some clubs even had a male president in addi-
tion to the female president. The male, in some
cases, had simply been offered honorary member-
ship because his name would lend prestige to the as-
sociation. The male and female presidents joined to
run the meetings and, in all cases, seemed to work
harmoniously together.

Benefits of the Clubs

A member of a contribution association is entitled to
a take-out even before she has paid her full share of
contributions-in other words, she is given the
credit on trust. Most of the contribution clubs inves-
tigated charged a compulsory "safe money" of a few
kobo per share, in addition to the regular share con-
tribution. Any donations or monies accruing to the
association from other sources are also paid into the
association's "safe," from which such contingencies
as musical instruments for the association's use are
met and from which short-term, low-interest loans of
small sums are made to needy members.
Most of the associations charged 20 kobo per
month on a loan of 2 naira for members and 50 kobo
for sponsored nonmembers-terms that are very rea-
sonable by village money-lending standards (see Ar-
dener, 1953, p. 135). Borrowers could defer paying
back their loans for the length of the association's
contributory cycle, provided they paid the interest
accruing on the loans regularly. All loans, however,
were called in shortly before the expiration of a cycle,
to enable the association's accounts to be "straight-
ened out" before it embarked on another contribu-
tory cycle.
Apart from loans, which were repayable, most


contribution associations would also make an out-
right grant from the "safe" to any member who had
suffered a substantial financial loss or a bereavement
requiring an unexpected expenditure. In the latter
case, members also kept an all-night vigil at the
house of the bereaved to demonstrate their fellow-
It is interesting to note that the women of these
communities did not patronize such alternative sav-
ings and credit institutions as banks and post offices.
At the time of this fieldwork, the only alternative fa-
cility for saving was the post office. We found that
many women, especially in the more rural town of
Obamkpa, thought the post office's sole function was
the dispatch and delivery of mails. But even where
all the savings and credit functions of the post office
were understood, it still had limitations. Minimum
contributions were higher, one could not borrow
more than the account was worth, and paperwork
and a delay of several weeks awaited those seeking to
make emergency withdrawals.
These limitations have all acted to discourage
people from patronizing the post office for purposes
of saving. The women we interviewed were almost
all illiterate, so that the paperwork involved must
also have frightened potential savers. The contribu-
tion association also has an edge over the post office
as a status-conferring instrument in the community.
When a woman uses the post office for saving, few
people know how much she has managed to save,
and therefore her economic standing in the commu-
nity cannot be evaluated on that basis. In contrast, a
woman who belongs to many contribution associa-
tions or holds multiple shares in a few, especially
where the take-out is high, and is known never to
default on her contributions, is highly respected in
her community.

Uses of Take-Out Money

Although it was difficult to find a woman who said
she joined a contribution club with the aim of gener-
ating capital for consumption, sometimes a woman
found herself using the money for non-profit-yield-
ing purposes. She may spend part of it on food, on
clothes for the family, or in paying hospital bills. But
even such an expenditure is an investment in the
well-being of the people who benefit from it and as
such perhaps should not be considered non-profit-
Since education is the gateway in Nigeria to a fi-

nancially more secure and successful life, most of the
women wished to secure it for their children. About
80 percent of the women interviewed invested or
proposed to invest at least part of their take-out
toward their children's education. Mothers joined
contribution clubs in order to be able to pay their
children's school fees or provide them with school
uniforms and textbooks. Although the education of
one's children is an area in which Igbo parents are
expected to cooperate, with the father providing the
fees and the mother providing the other needs, it is
not uncommon to find the mother being left to carry
the whole burden. This is more often so in a polyg-
amous household, where the multiplicity of wives
and the many children make it difficult for the hus-
band to fulfill his financial obligations to his family.
A woman may, however, be forced into this role if
she has a sick husband, or indeed a lazy or irrespon-
sible one.
A good education usually ensures a steady job
for the children who benefit from it. Accordingly, in
rural Igbo communities, where at least 75 percent of
the population still lives at or below subsistence
level, mothers look to their children to provide for
them when they finish their schooling and settle
down to jobs with regular salaries. This thought of a
rosier future for a woman's children and for herself
makes such an investment a matter of unequalled
importance. Mothers will join several contribution
associations despite the hardship involved, jump
their place in line, borrow from the association's

safe, or pledge their complete take-outs in order to
pay their children's way through school. Employ-
ment opportunities are becoming increasingly rare
throughout Nigeria, especially for the not so well ed-
ucated, with the result that mothers who can no
longer provide for their children's education are
forced to use their take-outs to "buy" jobs for them.
Another area in which women invested take-
outs was trade. The women studied, unlike their
counterparts in the eastern part of the country,
handle most of the petty trading in their local mar-
kets. They buy in bulk from Onitsha, which is only
20 miles away, and retail such factory-made goods as
soaps and canned foods or such food items as rice,
beans, and dried and iced fish, sometimes compet-
ing for the meager profits in such retail trading with
male Igbo traders from the east. The difficulty with
such retail trade is that the initial capital investment
is small, and accruing profits are correspondingly
small. There is also no reserve on which women can
fall back if business goes bad. These reasons and the
fact that a woman is very often forced to take from
what she sells to feed her family all contribute to the


risks involved in such deficit financing and the diffi-
culty in making profits.
Sometimes a woman arranges to collect her take-
out from two or more contribution associations at the
same time. She is then better able to embark on the
retailing of factory-manufactured prints and other
textile materials. Such enterprises require much
more capital to set up, but they also bring in more
gain so that less is used for immediate consumption
by the family. Trading, however, is not an end in it-
self but only a means to an end. A woman might in-
vest accumulated profits in her children's education,
in building herself a house, in making household re-
pairs, or in meeting her family's food and other ex-
We discovered a new area of priority in Og-
washi-uku, which has a big local market. A call by
the Omu (the town's female ruler) for the women to
help improve the appearance and the general
hygiene of the marketplace has led many women to
spend a good part of their take-outs to erect concrete
market stalls, with corrugated iron roofs in place of
the usual thatched ones. This is a worthwhile invest-
ment for the women since much of their time away
from home is spent in their market stalls.
Many women invested their take-outs in the
cloth-weaving cooperative, the only cooperative ven-
ture in Ogwashi-uku at the time of our fieldwork. On
the payment of a stipulated sum, a member was en-
titled to cheaper materials, with which she was able
to carry out her weaving craft. As a member of the
cooperative, she was entitled to free instruction and
supervision organized from time to time by the in-
dustry's senior staff. Membership also guaranteed
her a sure and quick market for her finished prod-
ucts, for the cooperative undertook the marketing of
such goods.
Another area of investment was the establish-
ment of drinking bars. This enterprise was under-
taken mostly by prostitutes, although a few married
women also opened bars. They were set up either in
front of the women's homes or within a convenient
distance of government offices or industrial estab-
lishments to ensure the patronage of the workers.
We did not encounter any instance in which
women used their take-outs to purchase land for ei-
ther building or farming. At the time of the field-
work, land had not become an economic commodity
in either town, and any indigent who wanted to
build or farm was free to do so. Acquiring land for
building purposes was, however, not the ambition
of any of the women we met. A husband would al-
ways provide his wife with enough land to put up a

house if she wanted to. None of the women invested
significantly in agriculture. Some successful women
may have subsidized their husbands in the purchase
of yam seeds for planting; however, this was not evi-
dent either from the women's answers or from par-
ticipant observation. On the women, however, fell
the duty of marketing the agricultural surplus for
their husbands, for which they were sometimes,
though rarely, compensated.


The major constraint on development in the two Igbo
communities studied lies in the poverty of the rural
population. Development requires massive financial
investments. The contribution clubs have proved to
be a system through which women can invest in
their children's education, in trade ventures, and in
their family's well-being. Our study of the women
participating in such associations illustrates the posi-
tive role women can play in the development process
and in bettering their own lives, and suggests ways
in which women, with additional capital, could
make productive investments in the fight against
rural poverty.


Afigbo, A. E. 1974. "Women in Nigerian his-
tory." Unpublished manuscript, Nsukka, Nigeria.
Ardener, Shirley. 1953. "The social and eco-
nomic significance of the contribution club among a
section of the southern Ibo." In WAISER Proceedings.
Ibadan: West African Institute of Social and Eco-
nomic Research.
--- 1964. "A comparative study of rotating
credit associations." Journal of the Royal Anthro-
pological Institute 94, no. 2 (July-December):
Bascom, William R. 1952. "The Esusu: A credit
institution of the Yoruba." Journal of the Royal An-
thropological Institute: 63-69.
Boserup, Ester. 1970. Women's Role in Economic
Development. New York: St. Martin's Press.
Geertz, Clifford. 1962. "The rotating credit asso-
ciation: A 'middle rung' in development." Economic
Development and Cultural Change 10, no. 3.
Green, Margaret M. 1964. Igbo Village Affairs.
London: Frank Cass, esp. pp. 44-48.


Isong, C. 1958. "Modernization of the Esusu
credit society." Niser Proceedings, pp. 111-120.
Leith-Ross, Sylvia. 1965. African Women: A Study
of the Ibo of Nigeria. London: Routledge & Kegan
Little, Kenneth. 1965. West African Urbanization:
A Study of Voluntary Associations in Social Change.
Cambridge: Cambridge University Press, esp. pp.
Marris, Peter. 1961. Family and Social Change in an
African City. London: Routledge & Kegan Paul.
Ottenberg, Phoebe. 1959. "The changing eco-
nomic position of women among the Afikpo Ibo." In
Continuity and Change in African Cultures. Ed. William

R. Bascom and Melville J. Herskovits. Chicago: Uni-
versity of Chicago Press.
Ottenberg, Simon. 1955. "Improvement associa-
tions among the Afikpo Ibo." Africa 25, no. 1: 1-28.
Paulme, Denise, ed. 1971. Women of Tropical Af-
rica. Berkeley and Los Angeles: University of Califor-
nia Press.
Van Allen, Judith. 1972. "Sitting on a man: Colo-
nialism and the lost political institutions of Igbo
women." Canadian Journal of African Studies 6, no. 2:
1974. "Women in Africa: Modernization
means more dependency." The Center Magazine 7, no.
3 (May-June): 60-67.

Marked Preference

for Female Sterilization

in a Semirural Squatter Settlement

Leela Gulati

The following article analyzes the reasons very poor semi-
rural women in South India choose sterilization as a
means of birth control. The women, who work for mini-
mal wages in a narrowing job market and whose children
have little hope for jobs or land marriages, are choosing to
be sterilized after two or three births. Among their limited
range of contraceptive choices, female sterilization in the
nearby hospital where women go for deliveries and abor-
tion is perceived by them to be the most advantageous in
terms of the cost of medicines and transportation and time
away from jobs. The article illustrates the centrality of
economic factors in low-income women's family planning

The birth rate in rural Kerala has shown a slow but
steady decline, particularly in the last 10-12 years,
dropping from 37 per thousand in 1966 to 28 per
thousand in 1975. For rural India as a whole, the birth
rate in 1975 was 37. Although this decline cannot be
attributed to a single factor, family planning efforts
in the State of Kerala are believed to have played a
considerable role in bringing it about (Krishnan,
The family planning program in Kerala began
modestly in 1955 but gathered momentum in the
mid-1960s. Its main emphasis has been on surgical
sterilization of both men and women; as a result,
nearly 24 percent of all reproductive-age couples in
the state are now protected by sterilization. Annual
rates of vasectomies have fluctuated significantly,
whereas female sterilizations have steadily in-
creased. For the period 1970-71 to 1976- 77, however,
there were 40 percent fewer female sterilizations than
male sterilizations.
The purpose of this article is to study the impact
of the family planning program on a small squatter
settlement called Aakulum Colony, located in a vil-

lage on the outer fringe of the state's capital city,
Trivandrum. More particularly, the article explores
what form of sterilization was more acceptable
among the settlement's very low income households
and why.

The Squatter Settlement

The squatter settlement' consists of 44 households
clustered in a group of impermanent houses made of
unbaked mud bricks, mud, or thatch, perched on the
slope of a hill about one kilometer from the city limit.
The total population is around 215. Most of the work-
ing men and women living in these households eke
out their livelihood from intermittent wage employ-
ment, mostly in unskilled jobs. More than 36 percent
of the households belong to scheduled castes (a term
commonly used for the disadvantaged caste groups
listed for certain preferential treatment in a schedule
of the Indian Constitution), whereas the proportion
of scheduled castes in the state as a whole is only 8
percent. Even among the remaining households, the
overwhelming majority comes from other socially
disadvantaged castes. The proportion of households
relying on agricultural labor and construction work,
categories in which there is a considerable overlap-
ping of workers, is as high as 60 percent, compared
with 32 percent for all of Kerala. The general literacy
level of the settlement, however, is not significantly
lower than the state average.
The squatter settlement enjoys the advantage of
being close to a major city (Trivandrum) and of being

'It is in the same settlement in which I had earlier studied the system
of food rationing as it operates in the State of Kerala. See Gulati (1977).


even closer to a major hospital. A bus to the city
passes the settlement at almost hourly intervals, and
the nearby Medical College Hospital is on the bus
route. Without meaning to play down the latter ad-
vantage, however, we should add that access to hos-
pital facilities is possibly the least difficult, in terms
of the distance to be covered, in the State of Kerala as
compared with the other states of India. In more than
half of Kerala's talks (each district is subdivided
into 3-5 talks) the average area covered by a public
medical center is less than 50 square kilometers (Cen-
tre for Development Studies, 1975).
However, the squatter settlement possesses vir-
tually no other amenities. The 44 households are all
served by a single public water tap, which is not con-
veniently located. This is the only source of safe
drinking water. There are virtually no private or
public toilet facilities; no hut has an electric connec-
tion, nor are there any street lights. By contrast, in
Kerala State the proportion of villages with access to
electricity is close to 95 percent, well above the all-
India average of 36 percent.
At the nearby road junction, a public loud-
speaker installed by the Panchayat (the village-level
government body) usually relays radio programs.
Also, five of the settlement households have portable
transistor radios. The corner tea shop buys the local
newspaper, and practically everyone who goes there
for tea, breakfast, or lunch gets a chance to read it or
to hear it read. Families are visited by two women
officials from time to time. One is the village-level
worker, gramasevika, who runs the nearby midday
feeding center for children up to age 5 and whom
everyone in the settlement, the women in particular,
seems to know. The other visitor is the auxiliary
nurse/midwife from the nearby Primary Health Cen-
tre, based in Pongomood, who is charged with keep-
ing track of the family planning target couples within
its jurisdiction.
Sixty-two married women, ranging from 20 to 75
years of age, reside in the settlement's 44 house-
holds. Of these married women, only 43 are now
living with their husbands; 11 are widows and 8
are separated. Although the incidence of illiteracy
among women in the squatter settlement is the same
as in rural Kerala, work participation rates among
married women in this community are, by compari-
son, very high. Women married to skilled workers-
mostly carpenters, whose wages are generally higher
and who also get work more regularly-are virtually
the only ones not working. Women workers are en-
gaged largely in unskilled jobs in construction, agri-
culture, or other occupations, although a few are self-

Knowledge and Acceptance
of Family Planning

All married men and women in the squatter settle-
ment know about methods for limiting the number
of children. Family planning posters and radio mes-
sages talk generally of the need for smaller families.
A major source of information on methods of con-
traception is the nearby hospital. Married women
who have given birth had heard the details of family
planning at the hospital when they went for their
confinements. The majority of deliveries in the set-
tlement now take place in the hospital, making such
visits an effective and important source of informa-
tion on family planning. Women in the settlement
had also heard about family planning methods dur-
ing the periodic visits of the village-level worker and
the nurse/midwife.
Mass vasectomy camps (organized in Kerala be-
tween 1970 and 1972, in a carnival-like style) were
also instrumental in spreading information and help-
ing to remove the stigma of secrecy once attached to
family planning. In the capital city of Trivandrum, a
15-day mass vasectomy camp was organized in Janu-
ary 1972. At the time, a number of officials and public
men visited the settlement, but the most active
among them was the local village-level worker, who
worked as a motivator and visited all the houses. Al-
though the camp was organized to perform vasec-
tomies, women too became aware of the family
planning program in general and of the aim of the
camp in particular. For good or bad, however, the
most pervasive and steady medium of information
on family planning has been word of mouth. Neigh-
bors, friends, and relatives spread the word about
those who undergo an operation for sterilization.
Every adult in the settlement seems to know not only
who has undergone surgery, whether a vasectomy or
tubectomy, but also what postoperative care was
necessary in each case.2
Although settlement residents complain about
the long lines at the hospital and the rude behavior of
the doctors and other hospital staff, they do not seem
to entertain strong reservations about hospital deliv-
eries, induced abortion, tubectomy, and vasectomy.
Women expressed no fear about surgical steriliza-
tion, nor did they attach any stigma to it. With re-
gard to vasectomy, some women spoke about the

A follow-up study of vasectomy acceptors in the mass camp in Tri-
vandrum showed that nearly 68 percent ot the acceptors had prior knowl-
edge of their relatives' having undergone vasectom According to this
study, "such knowledge might have induced them to undergo the vasec-
tomy operation." See State Bureau of Economics and Statistics (1975).


frequent backaches and loss of physical stamina that
their men complain about, but none said anything
about the after-effects on women of tubectomy.
Some men spoke about their loss of virility. In one
case in which a man had undergone vasectomy, his
wife still conceived and had children. The man now
denies having had a vasectomy and his wife sup-
ports his claim. Nevertheless, everyone refers to this
case as evidence of the possibility of failure.
Awareness of alternative, temporary methods of
family planning was rather patchy. Although many
married women have heard of the intrauterine de-
vice (IUD), none has used it so far. The IUD program
was started in 1965. After an initial spurt in the num-
ber of acceptors, a decline set in during the years that
followed, probably because of the reported side
effects. For Kerala State as a whole, the number of el-
igible couples using the IUD declined from 14 per
thousand during peak acceptance to 6 per thousand
in 1973-74.
Why have so few married women chosen to use
the IUD? A woman who had undergone tubectomy
explained: "When we don't want children, it is bet-
ter to have the matter finished. Let there be a perma-
nent end to this business of having children." IUD
acceptance also meant frequent trips to the doctor.3
For the state as a whole, the ratio of couples protected
by the IUD to those protected by surgery has, in re-
cent years, been 1 to 10.
What is more interesting is that while men and
women in the settlement were ready to talk openly
about sterilization, both were reluctant to talk about
the use of condoms. A stigma seems to be attached to
condom use; perhaps they regard such use as a sign
of oversexiness. Little information could be collected
on the extent of condom use. The village worker was
not sure that any man in the settlement used the con-
dom. According to the health inspector, acceptance
of condoms, which are distributed free, was nil.4

Sterilizations in the Settlement

A total of 27 sterilizations have been performed in
the squatter settlement: 9 vasectomies and 18 tubec-

3This is supported by observations cited in Estimates Committee
(1972): "The decline was partly because of shortcomings in pre-insertion
scrutiny and counselling, faulty insertion procedures and inadequate detec-
tion and treatment of side-effects. These lapses probably were the cause of
discouraging the present users and scaring away the future acceptors."
'Even for Kerala State as a whole, the number of couples using con-
ventional contraceptives-largely condoms-is believed to have declined
from 9,650 in 1970-71 to 6,522 in 1972-73. For the subsequent period, no
estimate has been attempted of the couples using conventional contracep-
tives. See State Bureau of Economics and Statistics (1979).

tomies. In 4 cases, both husband and wife have been
sterilized. Even for the state as a whole, it is not un-
usual for the number of female operations to exceed
the number of male operations. For instance, tubec-
tomies exceeded vasectomies significantly in the
years 1973-74 and 1974-75. In Trivandrum district,
however, female sterilizations have exceeded male
sterilizations all along since 1972-73.
Of the 38 married women in the reproductive
age group-those aged 45 or below-20 are protected
by either male or female sterilization or both. Of the
remaining 18 married women, 13 have had either no
child or only one. Strictly speaking, we should ex-
clude these women when determining the number of
eligible women married but not protected by steril-
ization or any other method. If we do so, the total not
protected would be 5.
Why did these five eligible women not accept
any kind of protection? The reasons given were var-
ied. Two of the women are widows and do not plan
to remarry in the near future. A third is a young, sep-
arated woman living with her parents. The remain-
ing two women felt that they had reached the end of
their childbearing period and that contraception was
not really necessary.

Explaining the Preference for Female

What are the possible factors contributing to the
preference for female over male sterilization? Any
answer offered on the basis of our study of the squat-
ter settlement has its limitations. All the same, since
tubectomies exceed vasectomies in Trivandrum Dis-
trict, the results of our study may help us to under-
stand a phenomenon that is much more widely

In-Hospital Deliveries

The most important factor contributing to this pref-
erence for female surgery has probably been the
great shift of deliveries in the squatter settlement
from one's own residence to the hospital. Out of a
total of 225 deliveries recorded among the 62 married
women in the squatter settlement, 51 percent took
place in-hospital.
While for women aged 46 and above, 30 percent
of the deliveries took place in-hospital, the corre-
sponding figure was 71 percent for women aged
31-45 and 84 percent for those aged 30 and below.
Actually, the proportion of hospital deliveries would


be higher except for the tendency to wait for labor
pains to start before leaving for the hospital. This is
due largely to the experience of many that the hospi-
tal sends them back home if childbirth is not immi-
nent. As a result, some women have delivered at
home while arrangements were being made to shift
them to the hospital.
The stated preference for in-hospital delivery is
indeed 100 percent among all married women of re-
productive age. They feel that hospital deliveries are
not only safer but also more convenient. Deliveries
in the huts are risky, inconvenient, and expensive.
In this context, it is relevant to recall that one of the
major problems of the settlement is an inadequate
supply of safe drinking water, as well as inadequate
room inside most huts. Younger women make no se-
cret of greater faith in doctors at the hospital than in
local midwives and old women. Also, if women go to
the hospital for delivery, any unforeseen complica-
tions can easily be attended to by the staff.
The principal problem in going to the hospital is
transportation. A taxi must be hired for about 12
rupees to drive a woman there. Even so, a midwife
charges between 5 and 10 Rs. per delivery, depend-
ing on the time taken and the complications arising
in each case. In addition, money is spent on herbs
and special food used during home delivery. All of
these expenses often equal the cost of going to the
hospital in a taxi and buying a few medicines pre-
scribed by a doctor. Much more importantly, all
women who have been to the hospital for deliveries
believe that they had far fewer postpartum complica-
tions and could return to work much sooner.
Of the total of 18 female sterilizations, 13 were
undergone after delivery and 5 after induced abor-
tion-all in-hospital. Nine out of the 13 postdelivery
tubectomies were undergone after 3 or fewer births.
The general tendency, clearly, is not to exceed a third
live delivery. Four out of 5 induced abortions were
undergone after more than 2 births, and all 5 abor-
tion cases were followed by tubectomies. It can safely
be said, therefore, that an unplanned third or fourth
delivery tends to drive a woman not only to abortion
but also thereafter to surgery.
According to the married women in the settle-
ment who have undergone surgery after delivery or
abortion, there was no question of asking their hus-
bands to undergo sterilization instead. Women go to
the hospital anyway, either for confinement or for
abortion. In either case, if they undergo sterilization
at the same time, no additional hospitalization is re-
quired. Nor is it necessary to take additional rest or
to incur any additional expense on transportation,
medicines, or special diet (Velayudhan, 1965). On the
other hand, men have to go to the hospital for the

specific purpose of sterilization and generally take at
least a month's rest from work thereafter, although
the prescribed rest period is only ten days. Usually,
men also ask for a special high-protein diet for sev-
eral weeks after the operation. On top of this, men
are widely known to complain that they cannot work
as well as they could before the operation." Women
in the squatter settlement, particularly those who are
working, would therefore prefer to undergo surgery

Preference for Smaller Families

After how many children does either partner decide
to undergo sterilization? To answer this, let us look
at the 23 couples in the squatter settlement who have
undergone sterilization. Only 1 was sterilized after
one child, while 6 opted for sterilization after two
children and 8 after three children. As many as 8 un-
derwent surgery after four or more children.
Let us now relate the age distribution of the 23
women protected by sterilization (male or female) to
the number of their surviving children. In the
younger age group of 30 and below, 5 out of 11 pro-
tected women have only two surviving children; for
protected women between 31 and 45, 2 out of 9 have
two or fewer surviving children. Also, while 9 out of
11 women in the younger age group have three or
fewer surviving children, the corresponding number
in the next higher age group is 5 out of 9.
It should be reasonable to infer a tendency
among the younger age groups in the settlement to
opt for fewer children and therefore a small-sized
family. This could be connected with the younger
age group's confidence in the survival of their chil-
dren (Kulkarni, 1975). In this connection, it is rele-
vant to note that infant mortality in Kerala State as a
whole seems to have declined drastically over the
past 15-20 years." In the settlement, of the 21 married
women aged 30 and below, 19 experienced no child
While only 3 men or women were sterilized with
no male child, 14 accepted sterilization only after
they had borne one male child and 4 after two male
children, regardless of the number of female chil-
dren. It should be added that out of the 14 couples
protected by sterilization after one male child, 7 had
two female children and 4 had three or more female

5According to the follow-up study of vasectomy acceptors (see foot-
note 2), while almost one-third of the acceptors had complaints about their
operation, about half the complaints of weakness were received from fewer
than 10 percent of the complainants.
6Infant mortality decline in Kerala has been particularly noticeable in
recent years. See Krishnan (1976).


children. What seems interesting to us is that being
assured of one male child was important for almost
two-thirds of the couples in the settlement who are
now protected.7


It would appear from this study of a small, semirural
squatter settlement in Kerala State that, with the
spread of family planning awareness among married
men and women in low-income, socially backward
households, something effective can be done to limit
family size and that there has developed a clear pref-
erence for surgery over other methods of contracep-
tion. Between male and female sterilization, the
preference seems decidedly in favor of the latter.
The pronounced shift of deliveries from the
house to the hospital appears to have played an im-
portant role in influencing the trend in favor of fe-
male sterilization. It appears that doctors and their
hospital assistants play an important role in motivat-
ing women to combine sterilizations with delivery or
induced abortion. This arrangement also means that
no extra period of rest for working women is re-
quired. At the same time, the decision to undergo
sterilization may be connected, quite understanda-
bly, with the confidence gained in the survival of
children. This is reflected in a larger proportion of

'A similar conclusion was reached by a study undertaken by the State
Family Planning Bureau. Their main finding was that the acceptors were
generally desirous of having a male child. See State Family Planning Bureau
(1973). Clearly, the existence of a significant son preference in Kerala State
goes against the common belief that, due to such factors as a prevalent ma-
triarchal system and the spread of female education, the preference for male
children may have declined. See Krishnan (1976), who seemingly subscribes
to this belief.

younger women undergoing the operation after hav-
ing two or three children. However, while the ac-
ceptable family size appears to be declining, having
at least one son seems to be a widespread preference.


Centre for Development Studies, Trivandrum.
1975. "Health indicators and demographic trends."
In Poverty, Unemployment and Development Policy: A
Case Study of Selected Issues with Reference to Kerala.
New York: United Nations, pp. 138-140.
Estimates Committee 1971-72, New Delhi. 1972.
Family Planning Programme.
Gulati, Leela. 1977. "Rationing in a Peri urban
community: Case study of a squatter habitat." Eco-
nomic and Political Weekly (19 March).
Krishnan, T. N. 1976. "Demographic transition
in Kerala, facts and factors." Economic and Political
Weekly, Special Number (August).
Kulkarni, G. A. 1975. "Relation between infant
mortality and fertility." Sample Registration Bulletin
State Bureau of Economics and Statistics, Triv-
andrum. 1975. Demographic Research Centre Paper No.
--. 1979. Statistics for Planning.
State Family Planning Bureau, Directorate of
Health Services, Kerala, Trivandrum. 1973. "A study
of factors motivating the individual to accept vasec-
Velayudhan, G. 1965. "How to achieve popula-
tion control." In Population Growth in Kerala. Ed. R.
S. Kurup and K. A. George. State Bureau of Econom-
ics and Statistics, Trivandrum.

Women in the Household Economy:

Managing Multiple Roles

Achola Pala Okeyo

The following article describes the strategies of a group of
rural Kenyan market women for coordinating their vari-
ous family responsibilities: subsistence production, in-
come generation, child rearing, and household mainte-
nance. By approaching the study of these women through
their income-generating role, which is essential to the per-
formance of their other roles, this analysis provides an
important n tree for understanding all aspects of women's

Kenya is primarily an agricultural country: a large
percentage of its population depends on agriculture
as small producers, market traders (i.e., own-account
employment), and wage earners in agriculture and
related industry. Fishing also provides a limited in-
come for a small number of people living on Kenya's
Indian Ocean Coast and around the inland lakes and
Rural women represent a major constituency in
socioeconomic and demographic considerations. Of
Kenya's 88 percent rural population, 52 percent are
women. Also, women account for 55 percent of the
total adult population over age 17.' It is further esti-
mated that 90 percent of rural women are engaged
full time in farm work on small holdings, compared
with 60 percent of the men.-
Although women account for a large percentage
of small farmers, their opportunities in wage em-

'The colonial era introduced a monetized economy in which men
and women were absorbed quite differently. While men were drawn into
the cash economy through school education (and therefore white-collar
jobs), colonial armies, and plantation work as wage laborers, women re-
mained in the nonmonetized subsistence sector of the economy. This ex-
plains, at least in part, why women came to form a large part of a stable
rural population, since men had to migrate as seasonal laborers into
towns and plantation areas.
"See Kenya, Ministry of Finance and Economic Planning, Central
Bureau of Statistics (1978).

ployment (including formal agricultural employ-
ment) remain severely limited. In the period be-
tween 1963 (the date of Kenya's independence) and
1971 women accounted for less than 15 percent of the
paid labor force.: By 1976, 13 years after indepen-
dence, this figure had risen only to 16 percent. Such
disparities are explainable, at least in part, by the
structure of access to education. Literacy is an impor-
tant factor in determining job category, remunera-
tion, and level of entry into wage employment.
Estimates show that around 79 percent of rural
women aged 15 and over cannot read or write; this is
twice as high as the national figure of male illiteracy.4
Despite these disparities rural women must still
identify and pursue sources of cash income over
which they have some measure of control, in order to
meet their financial responsibilities, the most impor-
tant of which is the maintenance of their families and
themselves. Informal-sector employment, such as
market trade, is one of the main sources of cash in-
come to which rural women look. Figures are not
available on how many rural Kenyan women are en-
gaged in full- or part-time market trade; however,
observation shows that women account for 70-80
percent of persons engaged in transactions in the

The Activity of Luo Market Women

Joluo (Luo people) are the second largest community
in Kenya.' They number over one million and live
mainly in the Nyanza Province in West Kenya. Their

'See Kenya, Ministry of Finance and Planning, Central Bureau of
Statistics (1972 and 1973).
'See footnote 2.
'Data for this paper were collected as part of a larger study on social
change and women's position in Luoland; see also Pala (1977). The main


settlements border on the low-lying plains around
Lake Victoria and stretch into the plateau areas
around the lake basin. The Luo traditional economy
was largely pastoral, but over the last 25 years it has
become more sedentary and agricultural, with cattle
still playing an economic role. In addition to their
nurturing and reproductive roles, women perform
important agricultural tasks. They are responsible for
much of the sowing, weeding, harvesting, and al-
most all storage and processing of agricultural and
animal products.
A combination of internal and external factors
(especially colonization) has created a number of
changes in Luo social and economic structure.
Among these, formal school education probably ac-
counts for some of the most significant changes in
Luo society, because education determines acquisi-
tion of new skills and access to salaried employment
opportunities, particularly in the urban areas and
rural administrative centers.
Because of differential access to formal school-
ing, more men find work in the urban/formal sector,
while women predominate in the countryside as
subsistence farmers and petty traders.
In rural Luoland the marketing of food is largely
carried out by women." They perform this activity ei-
ther as small producers or as market women. Small
producers sell produce from their own farms at the
local market to raise immediate cash for buying such
household and personal necessities as salt, soap,
sugar, paraffin, clothing, matches, meat, and cook-
ing oil. Market women engage in the regular buying
and selling of agricultural produce and/or fish in and
between rural and urban centers within one district
or between two or more districts.7 Unlike small pro-
ducers, market women do not sell produce from their
own fields; they act as intermediaries between small
producers and consumers. They also travel farther
than the nearest market and therefore cover substan-
tially greater distances, sometimes as much as 100
kilometers in a round trip. Astute market women try
to buy produce during times of bumper harvests of a

instrument of data collection was a questionnaire containing 14 questions
aimed at assessing the participation of women in the marketing of agri-
cultural produce and fish in rural market centers in South Nyanza District
in Kenya. Some of the research topics investigated were: the historical or-
igins of Luo market women as an occupational group in the countryside;
types of markets visited and conditions of work in food marketing;
sources of initial capital; the relationship between household and trading
responsibilities and how they are managed; income and expenditure pat-
terns of market women; and obstacles experienced in the course of work.
As Luo and Kenyan herself, the author has also drawn information from
observations and life experiences. The sample of 84 market traders was
drawn from two sublocations and consisted of 54 women specializing in
the marketing of agricultural produce and 30 specializing in fish trade.

particular crop and hoard it until a food shortage af-
fects a part of the district. Shortages tend to occur in
the weeding season, a few months before the millet
or maize harvest.
Women who specialize in the marketing of fish
also buy from areas of high supply and sell to those
of high demand. However, whereas the market
woman who trades in agricultural produce plays the
role of an entrepreneur who makes a profit on pro-
duce bought at low cost from the small producers,
women who trade in fish buy at initial high prices,
which affect the profits they can make on sales. A
greater profit accrues to fishermen and owners of
fishing boats and nets, who set the price of fish. The
market women buy a fresh catch from fishermen on
the shores of Lake Victoria and sometimes sell it
fresh at the nearest market. Because of the great dis-
tances to markets and the slow mode of pedestrian
transport, however, they are often forced to smoke
the fish and make two or more visits to the market
before they are able to sell one load of fish and buy a
new supply. Men also trade in fish and have the ad-
vantage of bicycles, which enable them to travel to
distant markets to sell fresh fish at twice the price of
smoked. It is therefore important to point out that
market women who specialize in the sale of agricul-
tural produce are likely to do a more successful busi-
ness than fish traders, because they can travel long
distances on the bus without undue spoilage of their
merchandise (see the section on income and expend-
iture patterns, below).
On a typical market day an observer will see two
main types of market women: the first, called johand
abedo (literally, the person who sits and exchanges),
can best be described as a retail trader. She arrives at
the marketplace by 10 A.M. for markets that start in the
morning and by 2 P.M. for markets that open in the
afternoon. Normally she has a stall or a designated
place where vendors sit according to the type of mer-
chandise. She sits there for many hours (between

"The basis of this specialization seems to derive from the role of
women in subsistence farming and in nineteenth-century barter trade.
Evidence from oral traditions on the premonetary period, especially
1800-1900, suggests that women were engaged in bartering of agricul-
tural and dairy products at border/fringe markets (see Pala, 1977). During
the same period, men bartered cattle and have continued to be cattle
traders. The current pattern of commercialization of the labor and occupa-
tional skills of men and women in the rural subsistence sphere follows
closely the precolonial forms of specialization, economic cooperation, and
exchange, as well as the importance of the domestic group as a resource
'In Kenya a district is a geopolitical entity that was first set up by
the colonial administration. The second largest administrative unit, it
was designed to coincide with indigenous structural boundaries within
and/or between linguistic groups and may be designated as rural or ur-
ban depending on its physical location. Several districts constitute a
province. There are eight provinces in Kenya.


four and eight per market day), selling produce in
small measures of ondong (a tiny straw basket the size
of a tumbler) or in cans of the same size. If she is a
fish trader she will sell medium-size fish in whole
quantities and large fish by the piece. Small fish are
sold in groups of three or by the handful. Agricul-
tural produce, unlike fish, is usually bulked at home
and brought to the market in small portions, entail-
ing at least three to five trips back and forth before
one set of grains is sold and a new load bought. Such
a slow process of sale seems necessitated by the com-
bined effect of distance from the home to the market
and between marketing centers, limitations on carry-
ing capacity by headloads, and the lack of storage fa-
cilities in the marketing area.
The second type of market woman engages in
wholesale trade. She sells produce in larger meas-
ures, usually in paraffin cans weighing 50 pounds,
and spends a relatively short time at the market. Be-
cause wholesale traders sell produce in bulk over a
long period of time, they develop clients with stand-
ing agreements for transactions. Other types of mar-
keting women appear seasonally to sell grains to
licensed beer club owners or secure a place at such a
club to brew and sell beer themselves. These women
are sometimes wives or sisters of a beer club owner
who manage to earn their own cash while providing
the club owner with labor and liquor for customers,
as well as a small fee for use of the premises.

Conditions of Work among Luo
Market Women

Occupational Socialization

One of the most informative characteristics of an oc-
cupational group is the factors that contribute to its
decision to pursue a certain specialization within the
limits of a set of realizable options. These factors are
situational, biographical, and economic and are es-
sentially rooted in the historical and socioeconomic
constraints within which such a group operates-for
example, whether they live in rural or urban areas;
whether their family type is nuclear or extended;
skills acquired; lifestyles and income levels of the
family. I refer to the role of such factors as occupa-
tional socialization. In asking Luo women why and
how they became market women, I was making the
assumption that a pattern could be discerned in their
responses that would confirm the hypothesis that

such choices are neither accidental nor mysterious
success stories. Rather, they derive from a combina-
tion of objective socioeconomic factors that can be
described and analyzed.
While the range of cash-earning activities for
women in rural Luoland remains fairly limited, the
interviews show that specific socioeconomic or situa-
tional features in the lives of rural women played an
important part in directing the choice to become
market women. For instance, source of initial capital,
advice and encouragement to start, and support to
continue trade are crucial for market women.
To the question, "Where did you get initial capi-
tal to start the trade?" 32 of the 54 agricultural pro-
duce traders named relatives as the source, 11 said
they earned the initial capital from their own wages,
and 11 earned the money by selling their own pro-
duce. Of the 30 fish traders, 16 named relatives as the
source, 10 said they made the money from selling
their own produce, and only 4 named their own
wages. Thus it seems that the foremost factor en-
abling a market woman to begin her trade is the ca-
pability of her kinsfolk to mobilize the necessary
initial capital." The second major factor is the avail-
ability of surplus produce from her own field, which
can help her undertake a bigger venture. It would
seem that women trading in agricultural produce rely
to a greater extent on support from relatives than do
fish traders. Although the difference in amounts re-
quired is not significant, agricultural traders need
more initial cash.
On the question of advice and encouragement,
37 of the produce traders and 17 of the fish traders
said they were supported by relatives; 15 of the pro-
duce traders and one of the fish traders said they
were advised by a friend. The salience of immediate
kin in the initial support of the market woman is
quite striking in terms both of capital and advice and
of encouragement to pursue the trade.
The main reason reported by market women for
starting in trade was lack of cash. Cash is needed to
meet household expenses and sometimes school fees
for children. Although women can rely on kinsfolk to
assist them in raising the cash needed to start a busi-
ness, they seek alternatives that would reduce their
financial dependence on relatives. Therefore they
prefer to pursue a job that guarantees a steady, inde-
pendent income for domestic and personal expenses.

"The most frequently named relatives were brother and sister, fol-
lowed by husband and husband's mother "Own wages" here refers to
money earned as a hired agricultural laborer or from sale of cotton from
one's own field


Organizational Mechanisms for Work

To ensure continued operation of their individual
businesses, market women resort to collective work.
They quickly learn that the pursuit of specialized
commercial activity imposes an additional constraint
on their time and labor, which are already substan-
tially committed to farming, housekeeping, provid-
ing for the family, and child rearing. Women often
experience the severity of labor shortage during peak
agricultural activity (especially weeding and harvest-
ing, drying, and storing of produce). This is par-
ticularly acute when such agricultural requirements
coincide with illness (personal or in the family),
death, social visits, recovery period after childbirth,
and other domestic commitments. For this reason
women make contingency plans to offset labor and
time shortages by emphasizing among themselves
the principle of collective work and mutual obligation
and cooperation.
Market women rely on both agricultural and
marketing work groups, organizing them in such a
way that they can participate in both groups without
too much loss of time for either activity. Agricultural
work groups are based upon labor exchanges be-
tween group members. If a member is unable to at-
tend, she may send a substitute or the money to hire
a substitute for the time she cannot be present. Labor
exchange, however, does not automatically guaran-
tee group members access to one another's crops.
Similarly, marketing work groups engage in labor
exchanges that carry no right of access to profits
made by individuals.
Marketing work groups provide traveling com-
panionship and the possibility of increasing profit
margins through division of labor in which desig-
nated persons travel on certain days while others
manage the storage of produce and identify sources
of produce. From the field data it is clear that fish
traders tend to be traveling companions only, while
grain traders tend to form a cooperative in the sense
of sharing services in order to minimize individual
costs of transportation and storage and thus increase
Although the basic principle of these groups is
economic cooperation, they achieve very specific ob-
jectives. Grain traders, for instance, work in groups
of two to six women. Over the years the groups
achieve a more or less cohesive structure with the fol-
lowing functions: (a) to guarantee group members
fair play in the trade; (b) to regulate market prices by
placing a ceiling on how much to pay the small pro-
ducers and at what price to sell; (c) to guarantee

members the flexibility to substitute one another's
labor and time without loss to the enterprise; and (d)
to provide a supportive base for one another's eco-
nomic and social welfare."
Fish traders use their group support differently.
They tend to operate more as individual units rather
than as a cooperative. However, they usually occupy
a designated area of the market, and in this way con-
trol one another's sale prices. They also tend to travel
together, but rarely engage one another to do their
business if they have to be away. Again, it should be
noted that this organizational strategy differs from
that employed by grain traders largely because of the
nature of the merchandise. It is relatively easy for
grain traders to rely on one another to transport their
produce from one market to another because it can
be loaded on a bus. In contrast, fish is prone to spoil-
age and crushing, and fish traders are reluctant to ask
another woman to take charge of their fish from the
lakeshore to the market and back home. Because the
methods of fish preservation are relatively under-
developed, women fish traders still represent a less
well organized and poorly remunerated occupational
group in the countryside."'

Effects of Domestic Responsibilities
on Commercial Pursuits

Recognizing that women have substantial respon-
sibilities on the farm and in the household, we next
explore how such obligations are related to and affect
their ability to pursue commercial (i.e., professional)
activities. To answer this question, I posed the ques-
tion: "When you are away on business, who takes
charge of caring for your husband and small chil-
dren, minding the livestock and chickens, cleaning
the house, and fetching water?" The assumption un-
derlying this question was, of course, that for a

"Market women are subject to a number of legal regulations that
govern the licensing and movement of produce both in quantity and in
quality from one place to another. Quite often they are arrested and fined,
and occasionally jailed, for failure to comply with such regulations. In
these circumstances group support is important: members can produce
money to bribe the police or lobby with the chief or magistrate to release
the arrested women or reduce police harassment. The groups are often
quite effective in achieving their intentions.
"'Under an improved system of storage and transportation, the
Kisumu municipal fish market has shown that the fish trade can be quite
lucrative. In this market, equipped with refrigeration facilities and a sink
with fresh running water in each stall, women earn in one day almost
three times what the rural "headload" traders make in a week.


TABLE 1 Substitution of labor for performing market women's household and related tasks
Substitute labor


Child care





Water portage 4 38 -7 33c 2
Cooking 4 46 -17 15d 2
Cleaning 7 37 11 26e 3
Tending livestock 11 32 23 8 6 4

aThis computation includes 9 respondents whc cite the babysitter (japidi) as the substitute labor. Babysitting is always performed by children under
age 14 and therefore counts as work done by children.
b9 respondents have grown-up children, and 3 have no children
c22 said they fetch water themselves; 11 have water tanks and therefore no longer have to fetch water.
d6 respondents are widows and therefore do not have a husband who has to be fed in their absence; 9 do their own cooking.
e These respondents said they have to do their own housecleaning.

woman to pursue a career outside the home, she
must be able to find some way of substituting her
own labor in agriculture, the household, and her ca-
reer so as to make all her pursuits at least manageable
if not remunerative. '
The most striking aspect of the women's re-
sponses is that, for every category of work, the labor
of older children is the most frequently substituted
for adult women's labor when women are away on
business (see Table 1). The contribution of children
is quite high even for a task like livestock tending,
which is ordinarily regarded as a man's respon-
sibility. A small number of women rely on hired la-
bor, while a slightly higher number depend on the
help of relatives. The low salience of the husband's
role in housework reflects the customary pattern in
which men neither cook, fetch water, nor care for
small children. Therefore, the presence of a husband
at home does not necessarily alleviate the burden of
work for market women, except with respect to ani-
mal husbandry. The presence of older children,
however, enables rural women to pursue an income-
generating activity, such as food marketing, for long
periods away from home.
The low salience of cowives" in substituting for

"Working "outside the home" here refers strictly to long-distance
travel that occurs regularly and requires time that must be substituted in
some specific way in order to maintain the home and farm. Ordinarily,
women farm and sell produce outside the home However, the distance
between farm and market is usually short (1-4 kilometers), enabling a
woman to combine her routine household affairs and her marketing and
agricultural activities within an integrated daily schedule. Of course, she
still relies on help from children and other family members.

market women's housework can be interpreted as a
reaffirmation of the independence of cowives in
providing for their households and in the daily care
and maintenance of their own houses. Luo women
do not ordinarily rely on a cowife to perform domes-
tic duties, especially food preparation, child care, or
housecleaning. They do, however, assist one another
in supervising children who remain behind to as-
sume household duties. It is believed that one of the
foremost strengths of the wife's role is her ability to
sustain the provisioning of her own household with-
out appearing to depend on other women. A cowife
may cook for her husband in the absence of cowives
because wives in a polygynous marriage share the
responsibility of feeding their own husband and
other males in that kinship category from the domes-
tic group.
Respondents were further asked to identify tasks
they find relatively easy to accomplish either on their
own or with the help of a suitable substitute, as well
as those for which a substitute was hardest to ar-
range. By and large, housework was not problematic,
and a substitute could readily be found. Some mar-
ket women are even able to combine such housework
as fetching water and housecleaning with their com-
mercial duties. However, a disproportionate number

"One would expect a higher salience of cowives in these tasks since
43 of the women interviewed are married in polygynous unions in which
the total number of wives ranges from 2 to 4. Only 19 of the women report
a cowife as their traveling companion on business trips. I was also told
that a cowife is rarely a traveling companion because someone (usually a
cowife) has to stay behind to oversee daily maintenance while the market
woman is away on business.



of agricultural produce traders (40) reported that farm
work was the most difficult task to delegate to any-
one, while only 4 of the fish traders reported this dif-
ficulty. On the other hand, 13 fish traders reported
procuring food as the most problematic area of labor
This discrepancy can probably be accounted for
by the fact that fish traders can more readily combine
agriculture with their work because of the nearness
of markets they visit. In fact, they have to depend
largely on agricultural production for their livelihood
because their meager trade does not provide suffi-
cient money to buy food. Women who trade in agri-
cultural produce, on the other hand, travel long
distances and are often not home to do agricultural
work.13 However, their trade is lucrative enough to
enable them to raise substantially more money with
which they can buy food at the markets they visit.
They can also ask someone who stays behind to buy
food at the nearest market. The differential in cash
availability is also observable in the pattern of hiring
labor: in nearly every case, women who report the
use of hired labor are those who trade in agricultural
A pertinent observation here is that, even
though women depend largely on their children's la-
bor while they are away on business, the availability
of money to install a water tank, buy wood for fuel,
or hire additional farm labor is important and desir-
able. For it seems that the use of labor-saving devices
in house and farm work enables women to invest
more time in trading ventures, with profitable re-

Income and Expenditure Patterns

The majority of fish traders interviewed (23) and only
9 of the agricultural produce traders earn less than 30
Kenya Shillings (K Shs 7.5 = US $1.00) per month.
All of the fish traders and only 19 of the agricultural
produce traders earn less than 100 Kenya Shillings.

"It should be noted, however, that agricultural work-especially
such peak season activities as weeding and harvesting-is more disrup-
tive for women who are long-distance traders in agricultural produce. Al-
though they rely on the group approach to speed up work in such
seasons, many of the women regretted having to give up their trading ob-
ligations for a time to be present even in a supervisory capacity at times of
weeding, harvesting, and storage. Even though the women can buy food
with cash from their trading concerns, they still rely on farm produce to
feed their families. And it is their responsibility to see that the grain
stores are not empty.

Furthermore, of the latter category 35 earn over K Shs
100. Thus the agricultural produce trader with the
highest monthly income earns at least three times the
amount earned by a fish trader with the highest in-
A large proportion of income earned in market
trade is spent on "subsistence" (food, clothing,
household goods, and health). Eighty-two respon-
dents reported using market income for personal and
family welfare, while only 2 cited school fees as a ma-
jor expenditure. The response to the question, "Why
did you become a market woman?" was invariably
stated as lack of money for household and personal
expenditure or difficulty in obtaining spending
money. Seventy-six respondents expressed the need
for cash income that they could control and spend for
household and personal expenses. Only 8 respon-
dents expressed the need for school fees as the main
incentive to become a market woman.
Three implications can be drawn from these fig-
ures and related observations on the income and ex-
penditure of market women. First, a trend in social
differentiation is now discernible among the two
groups of rural Luo market women: those who trade
in agricultural produce are better paid than fish
traders and, for this reason, are better able to acquire
more and better consumer goods in the local mar-
kets. Second, as the Luo economy continues to be-
come monetized, women retain the responsibility of
procuring food and providing for their families and
must seek ways of earning money to meet these
household cash demands. Third, as long as women
engage in some income-generating occupations,
they continue to maintain their independence- as
providers for the family. However, if they find no
opportunity to raise an independent cash income,
they must depend on their wage-earning kinsfolk
and/or husband to meet the household and personal
cash needs.
During the interviews, several women-both
market women and those who had no regular in-
come-expressed continued apprehension over
being without money. They characterized this aspect
of poverty as the lack of adequate clothing and food.
They used such Luo words as duk (nakedness), kech
(hunger or famine), and ladhruok (having too little of
anything to make ends meet, being perpetually de-
pendent on the goodwill of others). As one woman
put it, "In the past all you needed was a piece of
farmland and one cow for milk. But today, wherever
you go you need money. Tell me of a place where you
can mine money from the ground." This statement is
indeed a commentary on the changing position of


women, especially the uncertainties brought about
by the monetary economy. It ushers in a new form of
structural dependence within the domestic group
based upon the opportunity (or lack of it) to earn
cash income.


Butt, Audrey. 1952. The Nilotes of the Anglo-Egyp-
tian Sudan. Ethnographic Survey of Africa, East
Central Africa, Part 4. London: International African
Dakeyne, R. B. 1960. "The pattern of settlement
in central Nyanza in Kenya." Australian Geographer 8:
183- 191.
DuPre, Carole E. 1968. The Luo of Kenya: An An-
notated Bibliography. Washington, D.C.: Institute for
Cross-Cultural Research.
Evans-Pritchard, Edward E. 1949. "Luo tribes
and clans." Rhodes Livingstone Journal, No. 7.
1950. "Marriage customs of the Luo of
Kenya." Africa 20: 132- 142.
Fearn, Hugh. 1961. An African Economy: A Study
of the Economic Development of the Nyanza Province,
1903-1953. London: Oxford University Press.
Greenberg, Joseph H. 1966. The Classification of
African Languages. Bloomington, Ind.: Indiana Uni-
versity Press.
Gutto, S. B. O. 1976. "The status of women in
Kenya: A study of paternalism, inequality, and un-
derprivilege." Institute of Development Studies,
University of Nairobi, Discussion Paper No. 235.
Hanger, Elizabeth J., and Jon R. Moris. 1973.
"Women in the household economy." In Mwea: An
Irrigated Rice Settlement in Central Kenya. Ed. Robert
Chambers and Jon R. Moris. Munich: Weltforum
Kenya. Ministry of Finance and Economic Plan-
ning, Central Bureau of Statistics (Nairobi). 1972.
Employment and Earnings in the Modern Sector
-. 1973. Employment and Earnings in the
Modern Sector 1971.
-- 1978. Country Profile: Kenya, 1978. Nai-
robi (in cooperation with United Nations Children's
Fund, Eastern Africa Regional Office).
-- 1978. Women in Kenya. Nairobi.
Krystall, Abigail, and Achola O. Pala (eds.).
1975. "Women and education." The Kenya Education
Review 2, no. 2 (December).

Mbithi, P. M. 1972. "Harambee self-help: The
Kenya approach." The African Review 2, no. 1 (June):
147- 166.
Mutiso, C. G. 1971. "Mbai sya eitu: A low status
group in the centre-periphery relations." Paper pre-
sented at the annual meeting of the African Studies
Association, Denver, Colo., 3-6 November.
Ogot, Bethwell A. 1963. "British administration
in central Nyanza District 1900- 1960." Journal of Af-
rican History 4: 249-273.
1967. History of the Southern Luo. Vol. 1:
Migrations and Settlement 1500-1900. Nairobi: East
African Publishing House.
Okoth-Ogendo, W. H. 0. 1975. "Land tenure
and the transformation of peasant economies in
Kenya." Address at the International Women's Year
Tribune's Panel on the Family, Mexico City, 27 June.
Okot p'Bitek. 1971. Religion of the Central Luo.
Nairobi: East African Literature Bureau.
Ominde, Simeon H. 1968. Land and Population
Movements in Kenya. London: Heinemann.
Pala, Achola 0. 1975. "A preliminary survey of
the avenues for and constraints on women in the de-
velopment process in Kenya." Institute of Develop-
ment Studies, University of Nairobi, Discussion
Paper No. 218.
1976. "Women and development: An
overview on Kenya." Paper presented at the Ameri-
can Council on Education, Overseas Liaison Com-
mittee Seminar on Women in Rural Development,
Washington, D.C., 21-23 April.
-- 1977. "Changes in economy and ideol-
ogy: A study of Joluo of Kenya (with special refer-
ence to women)." Ph.D. dissertation, Harvard
University, Department of Anthropology.
1978. "Woman power in Kenya." Ceres,
No. 62 (March-April): 43-46. Food and Agriculture
Organization of the United Nations, Rome.
-- 1978. "Women's access to land and their
role in agriculture and decision making on the farm:
Experiences of Joluo of Kenya." Institute of Develop-
ment Studies, University of Nairobi, Discussion Pa-
per No. 263.
Southall, Aidan. 1952. Lineage Formation Among
the Luo. London: International African Institute.
Staudt, Kathleen. 1975. "Women farmers and in-
equities in agricultural services." Rural Africana, No.
29 (Winter 1975-76): 81-94.
Sutton, J. G. 1974. "The settlement of East Af-
rica." In Zamani: A Survey of East African History. Ed.
Bethwell A. Ogot. Nairobi: East African Publishing

Women's Reality:

Critical Issues for Program Design

Taherunnessa Abdullah and Sondra Zeidenstein

The following article looks at social controls on the be-
havior of rural women in Bangladesh that need to be con-
sidered in project designs if women are to be able to
respond to development programs. Rural women in many
countries are constrained by analogous social pressures,
which usually have an economic basis favoring the more
powerful and therefore are resistant to change. Since these
pressures on rural women often have a negative effect on
the goals of rural development, they need to be understood
and addressed.

Once one chooses as part of rural development policy
to initiate projects directed specifically to rural
women, whether they be agricultural extension, in-
vestment in food production, employment, educa-
tion, family planning, or health services, one must
design such projects in ways that rural women are
able to respond to. No matter what the project's long-
range objectives or what ideology or policy dictates
it, the first steps of implementation are always taken
with an acceptance of the situation as rural people
perceive it.
Underlying the approach of projects intended to
reach rural women must be an awareness that par-
ticular manifestations of rural women's behavior-
such as farming strategies, migration patterns, de-
sired number of children, food consumption pat-
terns, education of daughters-are responses to their
total socioeconomic situation. Project approaches
based on a partial view are not likely to succeed. Un-
derlying the project approach must also be an aware-
ness that women, like men, act out of self-interest.
Women are seeking, to the best of their abilities in
society as it exists, to satisfy their needs. Like rural
men, rural women are calculating in pursuit of their
goals and have little margin for risk. No one involved
in development work questions these statements
when applied to men. They are equally true for rural

women. However, we must recognize that the
cultural or socioeconomic pressures experienced by
rural women as a sex (varying with, but not negated
by, class) as they pursue security and survival are
different from those facing men.
The following description of some aspects of the
culture of rural women in Bangladesh exemplifies the
kind of information a development project needs in
order to be relevant. It is adapted from a longer study
of rural women of Bangladesh and the first steps of a
Bangladesh project to introduce opportunities for
change.' It identifies some of the pivotal factors that
control responses of Bangladeshi rural women to the
introduction of change, and that therefore need to be
taken into account in project design.


To initiate change in the condition of women in vil-
lages through a project, one must first grapple with
the concept of status or prestige, much more than
with economic class. The social ideology of status,
set by the powerful, controls the behavior of almost
all women in the village. If one is seeking to extend
resources to poor women to enable them to improve
their economic condition, one must understand how
the ideology of status limits the ways in which they
can afford to respond when new opportunities are

'The project is the Women's Programme of the Integrated Rural De-
velopment Programme (IRDP), under the ministry of Rural Development,
Local Government and Cooperatives. It began in 1974 as a component of the
Bangladesh Population Planning Project funded by the World Bank and is
coordinated with that project by the Population Control and Family Plan-
ning Department. See Abdullah and Zeidenstein (forthcoming). Other as-
pects of the culture of rural women explored in the study are: the work of
rural women; life crises; family and female strategies; cooperation and con-
flict; health and reproduction.


offered. The guidelines a project evolves to initiate
action toward long-range objectives must take into
account rural women's perspective on their situa-
tion. That perspective is not based on class.
Definitions of economic class in Bangladesh are
usually based on ownership of or control over land.
Definitions vary, but generally they distinguish four
classes: the landless, the small farmer, the middle
farmer, and the rich farmer. The landless have no
more than a homestead, sometimes rented. Small
farmers own up to about 1 hectare, middle farmers
1-3 hectares, and rich farmers over 3 hectares. Land-
less and small farmers comprise about 50 percent of
rural families; they cannot feed their families from
their own land. We have found through the project
that categorizing rural women by land ownership
does not explain variations in their response to avail-
able resources as efficiently as does the concept of
status. Because of status considerations, many of the
women among the 50 percent of poor rural families
would be unable to act to improve their situation if
opportunities were made available.
Status in rural society is frequently a basis for
power, influence, and respectability. It is usually ac-
companied by land ownership and financial means,
but may also substitute for them when a family's eco-
nomic condition has deteriorated, or is building. A
family that is newly rich may try to increase its status
by marriage alliance with a poorer family that has
it-usually a girl with status marrying a boy with
money. Status is a viable socioeconomic asset in that
it is a basis for making good marriage alliances that
will sustain or improve a family's position and for
maintaining power, influence, and control over re-
sources. Thus, a land-poor family will seek to hold
on to status as long as it can.
One traditional sign of status in rural Bangla-
desh is the behavior of the women in a family. Ap-
propriate behavior for high-status women includes
strict purdah, complete sexual division of labor, and
relative freedom from menial work. Only families in
good economic condition can afford to support such
behavior, but because it is the symbol of a family
with influence and power, most families aspire to
come as close as possible.
The image of relative idleness among its women
is projected by families that want to indicate their
status. On a trip to western Bangladesh, project staff
spoke separately to the men and women of a pros-
perous, conservative farm family. The women, in re-
sponse to questions about what they did, spoke
about their chickens, vegetables, rice processing,
and the like. The men, in response to the question
about their women, said "they cook and sew quilts."

The image of shelter for widows and other de-
pendent women is also projected as a mark of status.
Even when women are in real need, determination to
maintain status will keep them from revealing that
need by violating purdah or by doing certain kinds
of work for money. Women themselves usually rec-
ognize the value of preserving status, especially for
the sake of their children; but sometimes their need
is so great that what the family gives them in return
for limiting their possible responses to need is not
enough. Several women have spoken of this di-
One woman whose economic condition had de-
teriorated said, in response to the question "Why
don't you work?": "How can I? We were in good
condition for so long. Now, how can I do any work?"
The interviewer asked: "You can do handiwork,
can't you?" "Yes, I can, but for that capital is needed.
Where would I get the money?" The interviewer
said: "You can grind others' paddy." The woman re-
sponded: "It isn't possible. We might starve to death
but we have to maintain our status. Otherwise the
neighbors will speak ill of us."
A 25-year-old divorced woman with one son,
living at home with her "respectable" brothers (who
are seeking another marriage for her), told the inter-
viewer that she hoped to raise her son to be a "real
person." But she has no money for food or clothing.
She helps her mother with the domestic work but
wants to take a loan through the cooperative so that
she can husk paddy commercially, since she knows
how to do it. She will have to do it secretly because
her relatives and neighbors will criticize her. "Some-
times," she says, "one may know how to do work,
but not be able to get any use out of it."
There are, of course, variations in how families
and even whole villages will enforce status through
social pressure. But there is no doubt that it remains
one of the dominant values of rural society. If one as-
sumes that women in about 50 percent of rural fami-
lies are poor by class definitions and one adds to that
those who are poor and dependent because of the
loss of a husband through death, divorce, separa-
tion, or abandonment, one might expect the number
of women who would respond to any economic re-
source to be very high. But when one recognizes that
most of these women and their families are con-
cerned with protecting status-as an economic as-
set-one understands the complex problem a project
faces when it seeks to direct resources to rural
Among poor women (excluding for the moment
those who have no one but themselves to depend
on), we have experienced some basic differences in


response to resources that seem related to differences
in their status aspirations. A basic question for a
woman and her family seems to be, what do they
have to lose by behaving as if she had no aspirations
to status? Our experience indicates that women
whose families have been without status for a gener-
ation or two will act as if they have less to lose than
women whose families have been respectable in this
or the last generation. They will be more likely to
loosen purdah to take advantage of economic oppor-
tunities, to accept the opportunity to work, and to do
any kind of work than women who may be equally in
need but are restricted by status considerations. Poor
women from families that have declined quite re-
cently, though they may be widows, divorced, moth-
ers of hungry and uneducated children, and unable
to meet their basic needs from family resources, are
likely to be reluctant to break purdah or to do work
that is associated with nonstatus women-that is,
heavy menial work-for pay. They may be unwilling
or unable to take advantage of training unless it is
given in the village or to earn money in ways for
which there are ready markets, such as making mats
or processing rice. As one woman said, "We are not
the kind for making mats or selling puffed rice. Our
prestige will go." In many cases it is not just that a
woman will not do this kind of work for fear of losing
status; she may not know how to do it because of the
economic condition of her family. Rice processing,
for example, requires great skill passed on tradi-
tionally and cannot be quickly learned. Because she
is also unlikely to have been given education that
might now suit her for work having some status, she
is in a very helpless situation.
Some of these women will earn a small income
within the shelter of purdah in such respectable
ways as making fans, quilts, baskets, and even fish-
nets for commercial marketing. Ironically, they are
more likely than women without concern for status
to be exploited by family and middlemen because of
the restrictions on their behavior, their lack of expe-
rience in making money, and their dependence on
others. Although the latter women may be exploited
by society at large, they know, at least within the
limits of their work world, how to negotiate business
arrangements in their own self-interest. They can do
anything and go anywhere opportunities are of-
fered-to the city to be servants, to "food for work"
projects, to special training programs. But of course,
as soon as they get back on their feet, they will seek
to re-establish their status and return to respectability
and shelter. No matter how hard these women work
and how well they do, no one in rural society will
willingly emulate their behavior or consider it re-

Any project concerned with the integration of
women in rural development must therefore deal di-
rectly with the issue of status. It is not something to
lose sight of in efforts to reach the "poorest of the
poor," nor should the needs of women who value
status be rejected because they offend the work ethic
of planners and critics. Project administrators must
recognize that programs for rural women have to be
concerned with changing social values related to
women. They must seek ways to substitute work for
idleness as status symbols, and they cannot do so
only by addressing the poor. They must address the
more complex problem of how to enable status-
bound women to maintain and even raise their sta-
tus through regular income-producing work. It is not
easy, but it is the only viable way. It means that the
long-range policy of a program for women must be to
find ways for all women in need to be able to work,
not just those who are poorest now. The approach for
the present generations if one wants to move in that
direction must be to find various kinds of work for
women under varying social pressures and with
varying skills.


Purdah, or the veil, is a characteristic of Islam. Its
practice varies from place to place. Our concern with
purdah in Bangladesh is not with its religious signif-
icance but with the way it affects the behavior of
rural women.
The strict practice of purdah is a social and reli-
gious ideal in rural Bangladesh. Strict practice means
that a woman stays within the family compound,
which is surrounded usually by a "wall" of vegeta-
tion and sometimes has screens of woven rushes to
protect the inner courtyard. If possible she stays
within the inner courtyard. She is never seen by any
but close family males. This degree of seclusion re-
quires access to water for bathing within the com-
pound and the presence of servants. If a woman in
strict purdah must leave the compound to visit her
parents once or twice a year or for an emergency, she
wears a burkah, a loose garment that covers her from
crown to toe, and travels in a bullock cart with a
cover over each end or in a rickshaw with the front
covered, or travels at night by boat.
As a result it is costly for a rural family to main-
tain strict purdah for its women, and only the weal-
thy or those males (such as religious leaders and
school teachers or those with a "name" to preserve)
who make a special effort of time and energy can af-
ford it. Only a few compounds in a village, perhaps


five or six, can maintain such strict purdah, but what
is important to realize is that many villagers look to
these houses as the most prestigious and respected-
an honor to the village-and to this way of life for
women as the most desirable. In fact, they do not like
to see a lapse in the behavior of these women, and if
there is one it will take time before the family's pres-
tige can be restored.
Almost all village women aspire toward purdah.
Except for the very poor who have no choice but to
work in others' houses or beg or glean the fields,
most rural women practice purdah in a way that
strives to emulate the ideal, yet has flexibility to
allow for certain economic realities of the family sit-
uation. These women may move discreetly within a
cluster of compounds-usually of related families-
when men are in the fields or at night. Beyond these
compounds, reachable by sheltered paths, may be
open fields and public roads, which, using Mer-
nissi's terminology, are "male space" (1975). Women
do not usually traverse male space except on visits to
their parental homes. At such times they wear bur-
kahs if they can afford them. If not, they may cover
their faces and bodies with asari arranged like a bur-
kah and shield themselves with an umbrella. Those
who cannot afford these symbols of modesty cover
their heads and faces as much as possible with part
of the sari they are wearing. It can be seen, then, why
rural Bangladeshi women are so dependent on men.
Maintenance of purdah means that women cannot
have access to the world that lies beyond the bound-
aries of the compound, except through intermedi-
aries: young children for small matters; husbands,
fathers, brothers, and grown sons for whatever they
need from outside. Most women do not visit the
marketplace, the center of economic, social, and po-
litical activity. They do not go to the mosque, the
center of religious and social activity, nor to the
fields, the center of agricultural activity. They do not
go to school past puberty, even if they can afford it, if
doing so involves being with males or walking
beyond permissible boundaries. They do not have
access to the fruits of their labor nor the chance to la-
bor when in need. They do not go to the centers
where medical and family planning services are
available, and they cannot see the families to whom
they send their daughters in marriage.
It is commonplace to say that women's work is
complementary or auxiliary to that of men-that
women help men. But that terminology is not useful.
To better understand the situation of women, we
must understand that women's work does not have
direct access to the marketplace. At every turn it
comes up against the boundaries past which male in-
termediaries are necessary if it is to find a market or

social value. Women without men simply cannot get
their money's worth or their rights. This is why the
greatest need in a woman's life is male support-not
necessarily because he earns her food, but because
he is a middleman for her production.2

Women and Income

The Islamic ideal as it concerns the family economy is
that the male provides for and protects the female.3
The obligation that he undertakes as sole authority is
that of sole provider. Bangladeshi rural women's
work-unlike that of African women or, to some ex-
tent, Indonesian women-does not involve, as a
norm, any economic autonomy. Women say, "If men
give us food, we can eat; if they give us money, we
can spend." Men say, "We provide." Perhaps that is
the reason, in addition to the universal tendency to
overlook the struggles of the very poor, we have to
look very closely at rural women's lives before we
notice that many women are earning small amounts
of income, out of various kinds of need.
To design programs that women will be willing
and able to take advantage of, one must know as
much as possible about how and why women, in
contradiction of the ideal, are earning income and
what they are doing with it. We must not accept such
prevalent myths about women and income as:
women are not wage laborers; only destitute women
are motivated to earn income; rural women who are
not poor want money only for frivolous purposes;
male control of the purse is perceived by women to
be in the best interest of the family.
These and other myths can flourish in the ab-
sence of information about the income-oriented be-
havior of women, which is an issue of great
complexity. For one thing, the lack of attention in the
past to the existence of laboring women makes it
hard now to estimate the degree of unemployment
among them as a result of the gradual increase in ma-
chines that replace their labor. Women seek work as
laborers when they can no longer be concerned about
status. For some women this condition occurs
abruptly through the loss of male support and re-
sources. For others, it is dictated by the economic

'It is important to note that there are variations in women's experi-
ence of purdah by role and age. In spite of purdah, rural women know a
great deal about what is going on in the world as it involves their interests
'This section looks at ways in which women have access to and con-
trol over income from their labor as opposed to the routine heavy work re-
sponsibilities they have as farm wives and mothers. See the chapter on the
work of women in Abdullah and Zeidenstein (forthcoming).


condition of the whole family. However, many rural
families, in this situation of dwindling resources,
even those in which male members are working as
laborers, will tolerate great deprivations before suf-
fering the loss of status involved in allowing women
to labor.
What is the work of laboring women? The most
common way for these women to earn food in rural
areas is to sell their labor in other households, doing
rice processing and other menial tasks. There are a
variety of arrangements for doing this work (exclud-
ing exchange of labor among economic equals). One
arrangement is the formal contract negotiated for
specified periods of time-a few days, a few months,
or longer.
Women may have connections with particular
households that call them when there is work. To
guarantee being called for this work, a woman may
have to be generally available for other lighter serv-
ices, such as cleaning fish or milling rice. A less com-
mon arrangement involves year-round work and
residence. In addition, women may do rice husking
on a more commercial basis-that is, buying paddy
from the market and selling it back as rice or a more
expensive variation, puffed rice. Some capital is re-
quired for this arrangement. Village women can also
earn income by milking cows, collecting fuel, har-
vesting chilis, or performing as midwives. In some
places poor women do postharvest gleaning of what
is left in the fields, processing and selling it.
In some villages the arrangement between rich
and poor women is more feudal than commercial.
For example, women from a number of poor house-
holds may be at the call of a rich household for vari-
ous services as needed, often without having to be
asked. Such work includes helping with rice husking
and milling, fetching water, cooking at feasts, help-
ing in special food preparations, and being available
to cear the courtyard of drying rice when rain threat-
ens suddenly. For some of these services there will
be no immediate return; for others, some of the food
being processed or prepared is given to them. But
throughout the year, sustenance and support are
given in the form of food distributed on religious oc-
casions such as Shabi Bharat and Eidul Azzah, or in
the form of saris and relief goods distributed during
natural calamities or at election time, or because of
promises made to Allah for prayers granted, or at
baby-naming ceremonies or death anniversaries.
The rich family may also provide small loans without
credit or publicity, or, in times of scarcity, a place to
pawn one's valuables for small amounts of credit.
(They can be "borrowed" back for, say, a son-in-
law's visit and hopefully some time reclaimed.) The

rich family may distribute used clothing, provide
special foods for unexpected occasions, lend furni-
ture, pots and pans, and even ornaments for wed-
dings, and provide countless other loans of assis-
tance for which the labor of the poorer women is the
"payment." There may be less loss of status in such
an arrangement than in a strictly commercial one.
There are other traditional ways for poor rural
women to earn small amounts of food or money
without having to work in others' houses, which, be-
cause it involves heavy manual labor and violation of
purdah, is one of the least desirable ways to earn
food. Women, within the shelter of the compound,
may be earning for themselves as older widows or for
their children as heads of families. Or, contrary to
the accepted view of women, their contribution may
be counted on as one of the several sources of income
a rural family depends on to survive. In some parts of
the country, women with some capital buy goats or
cows, which they give to poorer women to raise. The
poorer women get the first year's milk and issue from
the animals and the second year return the cow or
goat and its new offspring. Women with enough re-
sources to buy a few chickens and keep them alive
and healthy earn small amounts of income from sell-
ing eggs, chicks, and chickens on a fairly regular ba-
sis or simply keep them as an asset that can be
converted to cash when a need arises. Milk, also, is a
marketable asset, in either the village or the local
market. Women with cows decide how much of the
available milk will be used for the family's food and
how much is to be sold. Again, the amounts of in-
come possible from such transactions are quite small.
Small amounts of other foods bring in a few takas for
women-vegetables, fruit, dates, syrup. Or women
may use them as barter, along with rice, in exchange
for a variety of small household items and trinkets
that tradesmen bring directly to the village. It is im-
portant to remember-because rural women con-
sider it important-that if women market their
produce outside the neighborhood, they are always
dependent on others as intermediaries in the trans-
action. In other words, they do not automatically
control the income their work produces. Whether
they do control it depends on many factors, but espe-
cially on whether there is an intermediary and, if so,
who he or she is.
Poor rural women also use their traditional skills
in handiwork to bring in meager amounts of money.
This kind of work can be done in the security of pur-
dah if there are males or children to help out, but at
the same time women have no direct involvement in
the financial transactions. Women may make quilts
on contract for other women who provide the raw


materials. In some areas women make fishnets, spin
thread, or make bidis (a kind of cigar) by contract to
middlemen who supply the raw materials, including
in some cases the spinning wheel.
Women who have greater resources or who have
only liquid resources have other, though not always
more productive, ways to make money. They may
keep small shops in their homes, or they may stock
rice and other commodities until the prices go up in
seasons of scarcity. (They may buy these com-
modities directly from other women in the village or
through intermediaries from the market or save them
from their own store.) Some women lend rice for in-
terest; or they lend money for interest, sometimes in-
vesting in local businesses, sometimes in land as
surety, which they then cultivate through laborers or
the "owner" as tenant. In one village near a cottage
weaving industry, women of all classes invest money
in the weavers, getting perhaps 15 takas a month in-
terest for an investment of 100 takas. Similarly,
women from weaving families may lend money to
farmers in sowing season and get rice as interest. As
one woman said, whatever money they have should
not be idle; it should be put to work.
It is important to emphasize that many of these
income-producing activities are being carried on dis-
creetly. They are not a secret (except in a few cases)
but they are not something to be flaunted. Thus, un-
less one has gained the confidence of rural women,
one may not see how many women are involved in
earning income or are interested in doing so. A num-
ber of rural women with whom we have discussed
income-generating activities have indicated that
they were not interested in earning money if it
would have to come through the hands of male rela-
tives. In interviews, we have heard of specific at-
tempts women have made to conceal income-
generating activities from their husbands. (Women
usually told us what other women have done.) For ex-
ample, one woman stocked rice in another woman's
house, while another woman had a neighbor raise a
goat for her. One woman secretly sold rice from her
own storage, while another has opened a "pan" bus-
iness with her young son and has told him to keep
their earnings a secret from her husband. Most
women say that they hide their savings in holes in
the bamboo, in the roof, or under piles of cloth.
What are the reasons for secrecy? One is the de-
sire to save. Women have said that if their husbands
knew there was extra money around, they would
spend it and work a few days less if they were day
laborers or give women less for expenses if they were
wage earners. Some husbands, if asked to market an
item for their wives, keep the money for themselves.

Women indicate that this is a matter of the character
and behavior of individual husbands. They say that
if husbands are "good," women can trust them to
buy and sell what they request. But if the husbands
are not "good," women try to sell through other
women or through vendors. Some husbands, they
say, are so bad they will take whatever they can.
How are women spending the small amounts of
income they do control? Women in very poor fami-
lies and those who are heads of households usually
spend what they earn on immediate survival-that
is, daily food. Women with more money spend it on
such expenses as small household necessities like oil
or soap; school fees and other educational expenses
for their sons; emergency needs like medicine; re-
leasing land from mortgages, leasing land, buying
land or a house; gifts for their married daughters;
support for their widowed mothers; special luxuries
and treats for their sons; ornaments (i.e., personal
assets) for themselves. With their own money,
women are able to meet family emergencies if they
arise and to provide for family security in coopera-
tion with their husband or perhaps, once in a while,
in spite of him. They are able to spend money on ob-
ligations women especially feel (e.g., for their
mother's or daughters' well-being), without having
to ask permission from the husband; they are able to
provide for their own security when it seems at risk
(e.g., if divorce seems imminent or a cowife's chil-
dren are getting more of the family finances); and
they are able to increase their influence over deci-
sions in the family.

Social Change without Program

In almost all our interviews with rural women over
age 30, the women offered unsolicited comments on
changes they had experienced in their lifetime.
Mostly these changes were related to age at marriage,
practice of purdah, education for girls, and the dow-
ry system. These are some points of direct impact on
rural women's lives of larger forces at work in the
economy-rapidly growing population, increasing
fragmentation and alienation of landholdings, diver-
sification to sources of income other than land. As
Seneratne (1975) carefully describes, changes in the
rural resource base produce changes in social prac-
tice to take advantage of them; belief systems are
concomitantly revised to support the practices of
those who control the most resources. Thus, new


patterns in education, marriage, and practice of pur-
dah should, if they seem to be successful strategies
for deployment of women, produce changes in belief
systems, mainly criteria for status related to women's
In relation to age of marriage, most older women
recalled having been married at least one or two
years before the onset of menstruation. They say that
early marriage was considered conducive to a good
adjustment, for the new daughter-in-law did not
have time to develop her own personality and was
still malleable. Besides, it was considered a waste of
money to keep a daughter in the parental home,
where she would require food and clothing but do
little or no work. Now, our respondents say, the ap-
propriate age for marriage is the onset of puberty.
However, they also indicate that a number of village
girls past that age are still unmarried, something
they consider a recent social problem. It is obvious
that there is still no strong rural belief system to jus-
tify later marriages, since relatives of unmarried girls
who are clearly 16 or 17 will consistently report their
ages as 12 or 13. These unmarried young women are a
social embarrassment-there is no appropriate role
for them. If they are going to school, their families
will justify their being unmarried in the name of ed-
ucation, but in most cases will take them out of
school if a good marriage offer is presented. Their
families might welcome new strategies that would
relieve the embarrassment connected with later mar-
riage and give them some positive justification for
having their daughters still at home.
Another change that women indicate is the re-
laxation of purdah. They continue to emphasize the
importance of purdah but recognize that it is being
interpreted in new ways. Some condemn this relaxa-
tion while others support it, depending, no doubt,
on whether advantages are derived from old or new
ways. Some prosperous older women who have had
to relax purdah to manage their households effi-
ciently say that they have done so "for the sake of
work." Parents and husbands of girls past puberty
who are going to secondary school or who have taken
some of the new jobs that involve contact with men
talk about "inner purdah" being as valid as seclusion
or the burkah. (In the past, higher education for most
girls was said to be impossible because it would
"hamper purdah"; now the definition of purdah is
being changed.) Although strict interpretations of
purdah are still the operative norm in most villages,
the reasons for these deviations among respectable
families indicate that purdah might be relaxed for
more women if the tradeoff were really advanta-
geous. If young women from families that are per-

ceived as "respectable" relax purdah restrictions to
take advantage of resources that are considered
valuable, they may set a standard in the village for
the interpretation of purdah that will influence oth-
ers. The fact that dependent women from well-off ur-
ban families have in recent years taken secretarial
training and become secretaries has given that job
status, so that other women who are more in need
and concerned with respectability can also do such
work. History indicates that change is possible, but
one must be very sensitive to how it is possible if one
wants to work toward further relaxation of purdah
Another significant change reported by rural
women concerns the education of daughters. Secular
education for rural Muslim males is a fairly recent
phenomenon, perhaps accepted for the past 30 years.
Primary education (up to class five) as a desirable
standard (not necessarily an attainment) for rural
girls seems to have become acceptable over the last
10-15 years. There are indications that secondary ed-
ucation for rural girls, though still very negligible, is
on the increase, with higher numbers enrolled in
areas with girls' rather than only coeducational high
schools. Such girls' schools, however, are very rare.
Very few women over 35 whom we interviewed
had more than a year or two of primary school. Most
were illiterate. They gave a variety of explanations
for their lack of schooling. Basically, they said, it was
not customary to educate girls in those days. Their
fathers and mothers believed that educated girls
would not obey their husbands; their characters
would be spoiled; they would not adjust well in mar-
riage. Their job was not to learn Bengali, but to learn
Allah's kalam (words or sayings). That would serve
them in the next world. If they made their husbands
happy and maintained the family, they would reach
All of these women report that they have given
some education to their daughters or have wanted to
but could not afford it. Even the few who state that
they do not believe in educating their daughters say
they have nevertheless done so out of social pres-
sure. In these cases, some education for daughters
seems to have become a form of behavior associated
with status. This seems to be a situation in which the
advantages of new behavior are so widely accepted
among those who set rural standards that the old be-
lief system that education for girls is bad has been al-
most completely undermined. It is likely therefore
that the main obstacle to some primary education for
more rural girls is lack of resources, both in the
family and the national budget, and not social sanc-


The change that elicits most concern among
women from families concerned with status is the
shift from bride price to dowry. Most older women
mentioned that in their day the groom gave to the
bride's side; the bride's side gave little in return. The
typical comment was, "In our time the bride's father
did not give a compulsory present in the marriage.
Whatever he gave out of his own will was enough.
Fathers of those days did not have to suffer in giving
daughters in marriage. Nowadays you have to give a
dowry to the boy-a cycle, watch, transistor. But in
our time the groom's side needed to give lots of gold
ornaments to the girl." Probably the dowry system is
not yet an issue for those who do not care about edu-
cated husbands for their daughters or for those few
families whose status is considered a sufficient dow-
ry. As Shirley Lindenbaum (1975) has pointed out,
the new emphasis on dowry rather than bride price
is the response to a shift away from land as the pri-
mary basis of status to the accumulation of money as
an alternative or concomitant basis.
Many marital difficulties are being blamed on
the problem of dowry. Brides of 14 and 15 are being
divorced or not even claimed from their parents'
home because promised dowries have not been paid.
Educated girls who marry into rural households are
being especially abused by their mothers-in-law.
Daughters are likely to be considered more of a bur-
den under these circumstances-that is, more costly
when resources are scarcer-than when they married
earlier and were valued for their ability to do farm
work. Families facing these dilemmas might be
ready for alternative strategies.
From many indications this time of change in
Bangladesh is likely to affect the situation of women.
Population growth, war, and inflation have created a
strong awareness among rural people of the dwin-
dling of traditional resources and a search for new re-
sources. The burden of dependent women on family
resources is becoming heavier. Unsuccessful strate-
gies to deploy women more advantageously in re-
gard to new resources may be increasing the number
of dependent women in status-conscious families as
well as the number of partially educated women who
are stranded in transition between rural and modern
life. The number of families that are both status-con-
scious and in need is increasing. Certainly the num-
ber of unemployed poor women is increasing with
the mechanization of the heavy work of rice process-
Change in the social situation of rural women
could be very rapid. Both the high-status families,
who control village values, and the heads of families,
who control family behavior, have fewer resources to

offer in return for the dependent, secluded position
of women. Women's eyes have been opened by the
atrocities of the recent Liberation War to an aware-
ness of how insecure their situation really is. Once
belief systems begin to shift, the change over the
next generations will be very rapid because the num-
bers of young people will greatly outweigh the num-
bers of elderly.
These changes that rural women have cited as
important in their lives have occurred without policy
intervention. In a sense population growth has
loosened traditional social controls because they are
no longer efficient for those who most benefit from
them; there is a transition to new social controls. If
change continues without significant intervention,
one can predict what will happen to rural women.
Very poor women will take advantage of any avail-
able opportunity, whether it be projects like earth-
lifting in exchange for food, brick-smashing to build
roads, or migration to the cities. Elite women will
gain status from higher education as they do in the
cities, and certain kinds of jobs for those who remain
in the rural areas will not only be acceptable but will
have some status attached. The response reported by
most projects involving rural women is that after ini-
tial hesitation there was little problem in hiring
women (usually under age 30) for a variety of jobs, all
of which provide income. (As in every culture,
white-collar work is more respectable than blue-col-
lar.) Education and jobs will increasingly become a
respectable family strategy for rural girls and will
quickly exhaust the job market. The majority of rural
women will be unable to take advantage of this new
standard for prestige set by a small elite and will be
left without resources, without an alternative to tra-
ditional behavior, and with even less perceived so-
cial value than before, when farm work was consid-
ered important.

The Project's Approach to Change

To see how the project referred to above is address-
ing the social and economic pressures constraining
rural women, we must note its main features. Paral-
leling the rural development approach for men and
operating through the same institution, it offers rural
women an opportunity to organize village-based in-
stitutions open to all adult women who purchase a
share. Membership requires women to deposit sav-
ings regularly, attend weekly village meetings, and
send five representatives every week to a training


and development center several miles from the vil-
lage. The project entitles women to credit for eco-
nomic production; access to modern training,
supplies, and services for upgrading economic activ-
ity; and improved family planning and health pro-
The project cannot enter a village without the
approval of village leaders. But once leaders have
sanctioned a credit/production institution for rural
women in their villages, individual members of the
group can act to meet their economic needs as they
could not have done alone because of status consid-
erations. Since credit is given for individual enter-
prises, women of different socioeconomic back-
grounds, skills, levels of education, and ages are able
to use it to pursue the economic ventures that they
consider appropriate to their present situation. For
example, some are raising livestock, processing rice,
or making pottery, while others are becoming tailors,
paramedics, or literacy teachers.
The fact that the group depends for its existence
(including new economic resources) on five repre-
sentatives traveling to a training center each week,
and that obtaining new training opportunities in val-
ued skills requires leaving the village to go long dis-
tances, sanctions a loosening of purdah and may
even give such opportunities some status. The exist-
ence of the institution in the village offers alternatives
and possible status as leaders to young unmarried or
divorced women in their parental homes who, with-
out such an institution, would have no alternative to
total dependence. At the same time, it provides their
families some relief from social embarrassment that
would further devalue their daughters and from the
added economic burden of a grown daughter at
Because the project addresses the interests of
women and their families in increased economic re-
sources without loss of the socioeconomic asset of re-
spectability, women are able to respond to and take
advantage of them. At the same time, the thrust of
the project is toward changing what is accepted as re-
spectable for women. That is, many women can join
and benefit from this project without leaving the vil-
lage, but, to the extent the project succeeds, it works

toward combating the tradition that status is ac-
corded for idleness and strict purdah and toward in-
creasing the value of women.
Of course, the project has a long way to go and
faces many problems as it develops. But the direction
in which it is moving seems appropriate to the situa-
tion as rural people perceive it. Although one might
have anticipated resistance, since villages do not re-
veal their potential for change until new resources
are introduced, the response of village men and
women has been positive and pressure for expansion
has been steady. However, further progress in di-
recting resources to women depends on how sub-
stantive the advantages are in competing with an
old, though weakened, code. And that, of course, de-
pends on how much planners are willing to invest in
development through women.


Abdullah, Taherunnessa, and Sondra Zeiden-
stein. Village Women of Bangladesh: Prospects for
Change. Oxford: Pergamon Press (forthcoming, 1980).
Lindenbaum, Shirley. 1975. "The value of
women." In Bengal in the 19th and 20th Century. Occa-
sional Papers, South Asia Series. East Lansing:
Michigan State University Asian Studies Center.
Mernissi, Fatima. 1975. Beyond the Veil:
Male-Female Dynamics in a Modern Muslim Society.
Cambridge, Mass.: Schenkman.
Seneratne, S. P. F. 1975. "Micro studies, employ-
ment and strategies of development." Unpublished.

ACKNOWLEDGMENTS This study was commis-
sioned by the Rural Employment Policies Branch
within the framework of the ILO World Employment
Programme. Additional financial support was pro-
vided by the Australian government and John D.
Rockefeller 3rd.

From Research to Policy:

Rural Women in India

Vina Mazumdar

The following article describes the new awareness of rural
women's role in development in India that has resulted
from studying rural women directly. It indicates the pos-
sibility of moving from research on rural women to the
formation of policy and the design of programs that take
into account the reality of rural women's lives.

Comprehensive research on rural women is a recent
phenomenon in India, as in most other countries. It
has led to a more realistic understanding of the roles
of rural women and an awareness of the impact of
their declining status on rural and national develop-
ment. In India, the next step has been to translate
such findings into recommendations for national
policy as reflected in the Sixth Five Year Plan. Al-
though not the final step-which would be success-
ful implementation of policy suggestions-it is a
critical one in linking research and action. This arti-
cle provides a country-level example of how research
on women can inform policymaking by analyzing
defects in existing policy and programs and suggest-
ing improvements on the basis of new knowledge
and understanding of the problems.

Research Findings

In 1975 the first major attempt to review and evaluate
data on various aspects of women's status and the
change in women's roles, rights, and opportunities
due to planned development was published as "The
Report of the Committee on the Status of Women in
India." Among their key findings, they noted that
the ratio of women to men has been steadily declin-
ing, thus widening the gap between the sexes from
2.3 million in 1901 to nearly 20 million in 1971. Be-

tween 1951 and 1971, the number of women workers
in agriculture declined from 31 to 25 million, while
the number of men workers increased by 34 million.
In the nonagricultural sector, women workers de-
clined from 9 million to 6 million, while men in-
creased from 33 to 48 million. The total number of
men workers increased by 27 percent while women
suffered a decline of 12 percent, reducing their ratio
in the work force to 210 per thousand men.
The last two censuses revealed a preponderance
of women in internal migration, at a ratio of 2,310 fe-
males per thousand males. Women constitute the
larger number of rural-to-rural and urban-to-rural
migrants, while men constitute the larger number in
rural-to-urban and urban-to-urban migrations. This
large-scale migration of women, mostly within short
and medium distances, is due, apart from marriage,
to severe under- and unemployment.
Identifying an interrelationship between wom-
en's low life expectancy, higher mortality, declining
participation in paid employment, and increasing
migration, the committee concluded that these de-
mographic trends were indicators of "a process of
change that is moving in a direction opposite to the
goals of our society and its plan for development." In
the committee's view, these trends represented an
intensifying devaluation of women.
Responding to the committee's conclusion that
"changes in the status of women will be a long-term
aspect of our social process and will require continu-
ous examination and assessment by persons inter-
ested in social change," the Indian Council for Social
Science Research adopted a program of sponsored re-
search to generate and analyze data on significant
patterns of social and economic organization affect-
ing women's position in the long run, and to develop
new perspectives in the social sciences, particularly
through clarification of such concepts as the family,
household, women's work, economic activity, and




productivity. The restrictive and ambiguous use of
these concepts has resulted in a general underassess-
ment of women's contributions to society.
The program emphasized the study of women in
the poor (or less visible) sections of Indian society
with a view to understanding the regional and sec-
tional differences in women's roles, status, and prob-
lems and the differential impact of development on
their lives. This emphasis was designed to correct
the bias of previous research and policies, which
were primarily influenced by the experiences of
women in the elite classes.

Influencing the Planning Process

The results of research undertaken by this program
during 1975-77 were summarized in a short memo-
randum, "Critical issues on the status of women:
Employment, health, education: Suggested priorities
for action," by the committee of social scientists
(which supervises the program) in 1977. The memo-
randum highlighted the alarming trends in the con-
dition of women, and recommended urgent policy
intervention by the government, particularly in the
areas of employment, health, and education. The
publication, which was sent to the concerned gov-
ernment ministries and to the Planning Commission
for consideration, aroused considerable interest in
the press, academic circles, and some organized
women's groups. It was also distributed among vari-
ous government agencies by the women's bureau,
established by the government in 1975 to initiate and
coordinate policies for women's welfare and devel-
The publication of this document coincided with
the decision of the government of India to begin the
Sixth Five Year Plan in 1978. It was therefore used to
create awareness of the problems of women in devel-
opment among various working groups, which were
appointed to study issues of special concern to devel-
opment policy, and to provide information and rec-
ommendations to the Planning Commission. For the
Sixth Five Year Plan several working groups on
women were specifically established as part of the
planning exercise. This paper summarizes the rec-
ommendations of three of these working groups.
While designing the National Adult Education
Program, the Ministry of Education appointed a spe-
cial working group to advise on measures necessary
to make the NAEP more attractive, appropriate, and
accessible to rural women and to poor working
women in urban areas, who together comprise two-

thirds of the NAEP's target population. A working
group on Village Level Organizations emerged as
part of the government's heavy emphasis on rural
development. The government had concluded that
the 60,000 village women's clubs, created under ear-
lier development programs, were not always an
effective instrument for rural mobilization. A work-
ing group on employment was convened by the
Planning Commission member responsible for em-
ployment policies, stemming from his conviction
that the status of women had undergone a significant
decline and that mass poverty could not be attacked
without expanding women's earning capacity.
Each working group consisted of 10-20 mem-
bers drawn from researchers, fieldworkers, represen-
tatives of the government organizations charged
with implementation, and usually a member of the
Secretariat of the Planning Commission. Each group
had to translate data, reports, and all materials used
for deliberation into realistic and politically and so-
cially acceptable recommendations for governmental
action within a five-year timetable.

Adult Education

The objectives of the National Adult Education Pro-
gram are to promote (a) literacy, (b) improvement of
functionality, and (c) development of social aware-
ness among the adult illiterate population, the major-
ity of whom are in rural areas. Within this context,
the working group on adult education programs for
women recommends a special approach to women,
because earlier policies to extend educational oppor-
tunities, both formal and informal, to the general
population have failed to bridge the growing knowl-
edge gap between the sexes. The reasons for this fail-
ure lie partly in discriminatory attitudes toward
women and partly in the inadequate realization
among planners and administrators of women's mul-
tiple roles in society. Both formal and informal edu-
cation emphasizes the importance of household arts
in women's training, ignoring the fact that women
form a substantial and integral segment of the labor
force in agriculture, industry, and services. Failure to
strengthen women's productive and economic
roles-particularly in rural areas-has also contrib-
uted to the gradual erosion of their economic oppor-
tunities, increased unemployment, and mass pover-
ty. It has burdened them with exacting labor for in-
adequate returns, in addition to their housework and
child care. The absence of leisure for either education
or entertainment is common among poor women,
both rural and urban.


To be meaningful to these overburdened wom-
en, an educational program must address itself first
to increasing their earning power. Health, nutrition,
child care, and family planning, currently featured in
informal educational programs for women, can be
included because they help to strengthen some of
women's natural roles. But it is equally important to
promote the understanding of women as individuals
with basic rights to dignity and autonomy and not
merely as instruments of production and reproduc-
tion. The NAEP should therefore seek to make
women and men more conscious of their rights and
responsibilities, of the laws governing women's sta-
tus, and of the various manifest and concealed causes
of women's oppression; assist women to achieve
economic viability through acquisition of literacy
and other necessary skills and resources; provide
them with access to knowledge in such areas as
health, child care, nutrition, and family planning;
and assist them in forming their own groups for
learning and productive activity and in strengthen-
ing their participation in the development process.
Most rural women have traditionally been inte-
grally involved in the production and distribution of
goods and services in agriculture, livestock rearing,
dairy farming, fishing, and other tasks in the pri-
mary sector; in cottage or household industry of all
types; and in traditional services (e.g., washing; re-
tail distribution of various agricultural and industrial
products in local markets). This employment, of
course, is invariably in addition to the work women
perform in caring for their families.
The social hierarchy of rural society has, how-
ever, imposed different types of constraints on dif-
ferent groups. While the economic roles of upper-
class women are confined to processing and storage
of agricultural products within the home and to the
feeding and health care of the farm workers, the
landless lower-class women engage in wage labor
and suffer from overextended work days, poverty,
malnutrition, and perpetual insecurity. Agriculture,
household industry, and local services, which have
traditionally formed their main sources of livelihood,
are affected by the spread of modernization, causing
displacement of many rural women from traditional
occupations. The increasing gap in knowledge af-
fects both categories of rural women, although they
differ from each other in other characteristics.
Educational programs for rural women, there-
fore, must avoid harvesting and sowing seasons,
provide recreational activities and child care arrange-
ments, and adjust the timing of classes so that they
do not interfere with other responsibilities. Skill
training must include productive, managerial, or-

ganizational, and participatory skills and should be
linked to similar programs being promoted by major
development agencies seeking to affect agriculture
and rural industrialization. Although literacy is in-
dispensable, it should be preceded by promoting
consciousness among the learners through discus-
sions of issues that affect their lives. It is not neces-
sary to discriminate between adult women and
young girls, since rural society does not differentiate
between different age groups in the same manner as
in urban and upper-class families.

Rural Development

The report of the Working Group on Development of
Village Level Organizations of Rural Women was
published in June 1978. The working group's man-
date included a review of the objectives and func-
tions of the mahila mandals, or women's clubs, and of
youth clubs created as part of the participatory in-
frastructure for rural development.
According to the working group's report, there
was a suspicion among those interested in rural de-
velopment that the activities and benefits of many of
the 60,000 women's clubs were not reaching the rural
poor, who are too busy coping with a hand-to-mouth
existence. The best way to make full use of women's
enormous potential is to help them achieve greater
productivity, by improving their present skills and
developing new ones for generating alternative and
new sources of employment. In view of the growing
body of evidence that men and women are not equal
beneficiaries of development, the working group rec-
ommended that the government opt for concrete and
systematic steps in favor of women. One of these
steps is to help women come together as a group to
improve their productivity or employment status
and to work toward self-reliance.
The women's clubs have suffered from a lack of
clear objectives and of attention to women's multiple
roles within and outside the home; difficulties in
identifying target groups and inability to reach the
poorest rural areas; inadequate training in produc-
tive skills, resulting in the program's being regarded
as nonproductive by both government agencies and
local women; a multiplicity of programs and agen-
cies, resulting in uncoordinated diffusion of re-
sources, inadequate coverage, and absence of an
integrated policy or approach; lack of properly
qualified field staff and inadequate provision of su-
pervisory staff; failure to mobilize the effective and
sustained support of men and the absence of appro-
priate linkages with local institutions (e.g., the Vil-


lage Councils, or Panchayats), leading to lack of
recognition or support from these bodies; and
rigidity in program design and structure, leaving lit-
tle room for local initiative or identification of needs.
The program has viewed rural women as a ho-
mogeneous group whose primary role is homemak-
ing. It has emphasized the training of better-off
women in home management, while the needy or
weaker groups, particularly the workers, have been
served only through feeding and similar programs.
Craft classes, home science education (including
training in child care, nutrition, and "home econom-
ics"-which sometimes extends to methods of pre-
paring new food crops for family consumption), and
nursery schools (balwadis) have been utilized mostly
by those who have the time and the means to use the
The working group recommended a new pro-
gram, which would be an integral part of the Sixth
Five Year Plan strategy. The strategy emphasizes
comprehensive area planning and organizations of
the rural poor as essential for rural development.
These organizations are intended to ensure the
poor's access to development resources and to func-
tion as pressure lobbies. Village-level organizations
of women that enable them to undertake viable proj-
ects and measures to strengthen their economic posi-
tion would also raise their general status and draw
them into roles of public leadership and into the
mainstream of development. Specifically, the pro-
gram would reduce under- and unemployment of
rural women and provide them with a basis for par-
ticipation and training in income-generating activi-
ties and in all developmental activities; promote self-
reliance and collective action for the betterment of
the home, family, and community; facilitate better
management of resources; improve conditions for
bearing and rearing children; and provide forums to
enable women to participate freely and fully in deci-
sions that affect their lives and the community.
A single type of women's organization cannot
meet the needs of all groups because of existing ine-
qualities and differences in the interests of rural
women. The working group thus encouraged a flexi-
ble approach with room for several types of organiza-
tions-for example, cooperatives, trade unions, and
registered societies-to grow in accordance with lo-
cal needs.


The Planning Commission established the Working
Group on Employment of Women to examine ways
to increase full- and part-time employment of

women in the organized and unorganized sectors in
both rural and urban areas; identify groups of unor-
ganized self-employed women and suggest ways to
strengthen their employment by resolving difficul-
ties of marketing, availability of raw materials, and
the like, and by eliminating middle men wherever
possible; initiate viable pilot projects among groups
of women to generate economic activity; and organ-
ize women into associations or unions.
The working group noted that the problem of
women's employment is characterized by (a) their
inability to reach for services and assistance pro-
grams offered by government and quasi-government
institutions; (b) a lack of awareness among these in-
stitutions of the need to promote women's employ-
ment; (c) the tendency of economically powerful
organizations to obtain financial and other assistance
in the name of women but, once obtained, to divert
it to other areas of investment; (d) the fact that tech-
nological modernization in several industries has not
resulted in the protection and expansion of women's
employment opportunities, nor in increasing wom-
en's skills, training opportunities, or upward mobil-
The Draft Sixth Five Year Plan has noted the ex-
istence of sectoral imbalances between men and
women in available opportunities for regular em-
ployment, training, and promotion. The plan has
also emphasized the need to expand employment op-
portunities for women and provide special programs
to prepare them for such opportunities.
The Draft Plan identified four strategies for in-
creasing employment of women-namely: (1) diver-
sification and expansion of educational and training
opportunities; (2) manpower budgeting of the female
labor force in all comprehensive area-development
plans and designing of appropriate programs to offer
a variety of training and work opportunities; (3) pro-
motion of self-employment and small-industry em-
ployment among women by ensuring them a
reasonable share of credit and other inputs; and (4) a
higher rate of investment in "women-preferred" in-
dustries and occupations.
The working group wholly endorsed the first
three strategies. With regard to the last, however,
they cautioned that a policy of increased investment
in the so-called women-preferred industries would
be a mixed blessing. A look at the nature of these
jobs shows that they include generally unorganized
activities in which women receive low wages. Work-
ing conditions are bad in such jobs as domestic serv-
ices, sanitation, and such forest-based occupations
as gathering fodder and firewood. The working
group warned that expansion of these activities


would mean trapping more and more women into
physically exhausting jobs, often at below subsist-
ence-level wages. On the other hand, they recom-
mended additional investment in such activities as
manufacturing dairy products, canning and preserv-
ing fruits and vegetables, and rearing silk worms.
Furthermore, the group's report sought an increase
in the proportion of women to total employment in
these industries. In dealing with the less desirable
industries, however, it should be ensured that the
proportion of women to total employment in these
activities does not decline from the current levels,
while steady efforts are made to improve wages,
working conditions, and regulation of working
In view of the lack of interest of such promo-
tional organizations as credit institutions in the
problems of women's development, the working
group recommended that specialized agencies be es-
tablished to identify, promote, and assist individual
women and women's groups in undertaking income-
generating activities. The specialized agencies would
create a better understanding of women's economic
and other needs among various government agencies
so that the awareness gap between the aid givers and
the potential beneficiaries may be gradually reduced.
It has long been assumed that development pro-
grams automatically benefit both men and women.
Experience has revealed, however, that unless there
is a special plan for women with specific earmarking
of funds in sectoral plans, women will not benefit.
The working group proposed that utilization of
funds earmarked for women's programs be strictly
monitored. To achieve this, the government must
create adequate and properly trained machinery and
cadres in each of its program areas. It must adopt a
well-defined policy for women's development,
clearly stating the economic and social objectives so
that they receive continuous attention and support.
Suggested programs and approaches must take
into account the employment issues and training
needs relevant to the situation of rural women. For
example, institutional training may prove inacces-
sible to the majority of rural women, because of the
illiteracy and relative immobility caused by social
and economic constraints. Training programs for
rural women may have to be provided by mobile
units sponsored by specialized agencies (Small-scale
Industries Service Institutes, Village Industries Com-
missions, Agricultural Extension Departments, and
others) or by other training units located nearby (In-
dustrial Training Institutes, District Industries Cen-
ters, Farmers Training Centers, among others) after
identifying suitable trades on the basis of local skills,

viability, availability of markets and raw materials,
investment priorities, and the like. Such agencies
should be associated with the project for a period of
time to provide needed assistance in follow-up train-
ing, supervision, and monitoring. This will call for
some reorientation of the staff of existing training in-
stitutions to make them responsive to local employ-
ment needs, so that new trades can be developed. In
order to ensure continuity in training, rural women
should be trained both as organizers and as instruc-
tors in locally viable trades.
Since the agricultural sector employs the largest
number of rural women, the working group consid-
ered it necessary to ensure proper training facilities
for such women to improve their skills and demand a
better wage, as well as to improve their productive
capacity. The existing agencies engaged in the train-
ing of farmers should be properly equipped to as-
sume the training needs of rural women in agricul-
tural and allied sectors.
To prevent displacement of women's labor
through the introduction of new technology in all
sectors of the economy, planners must change the ex-
isting structure of fiscal law, which tends to support
capital-intensive rather than labor-intensive technol-
ogy. Capital-intensive technology also tends to in-
crease demand on such scarce national resources as
power. The working group therefore recommended
special studies to examine the employment impact
and relative cost efficiency of alternative technologies
in both the agricultural and the industrial sectors.
The working group has observed that expansion
of employment and strengthening women's deci-
sion-making roles would also contribute to the adop-
tion of the small-family norm and improvement of
women's status, already accepted as national objec-

Prospects for Implementation

The recommendations of the three working groups
were delivered in 1978. The Planning Commission's
review of the Working Group on Employment has
been generally favorable, especially to the recom-
mendation that a reasonable share of the funds in
each sector (e.g., agriculture, industry, social serv-
ices, banking) be used to develop programs to ex-
pand women's development and access to employ-
ment. The recommendations were endorsed strongly
by the National Conference on Women and Develop-
ment, sponsored by the government of India in 1979.


The Ministry of Education has published the recom-
mendations of the Working Group on Adult Educa-
tion. The recommendations of the Working Group
on Village Level Organizations are still under discus-
The success of any of these recommendations
will depend upon (a) adequate communication of the
thrust of the policies through all levels of govern-
ment; (b) debate and discussion on a national scale,
including informing women of the existence of bene-
ficial development programs; and (c) the capacity of
the governmental machinery to train large numbers
of functionaries with the substantive skills necessary

to render services to women, the sensitivity to coop-
erate with them, and an acceptance of the ideology of
women's participation.
This sensitive acknowledgment of women's is-
sues by top policymakers in India is in part attribut-
able to research demonstrating the linkages between
mass poverty, unemployment, and rural inequality,
on the one hand, and the condition of women, on the
other, as well as to the constitutional commitment to
equality of opportunity and status for women. Suc-
cessful implementation, however, will depend on or-
ganized pressure from women at all levels of society.
Such organized pressure is still highly inadequate.

Part II

Approaches to Learning:
A Sharing
of Experience


Judith Bruce

Tools for learning about rural women have been lack-
ing. Many of the methodologies now in use were de-
veloped to describe the male experience and role in
rural culture. Women's role and experience when re-
ported on was often heavily filtered through both
outside and male views of them. Normative views of
women and lack of direct contact with them have
generated distorted information on their contribu-
tion in national-level statistics and in anthropologi-
cal literature.
The task at hand is to develop, quickly, a usable
and factual information base on which to plan rural
development programs that reach and benefit
women. "Approaches to Learning: A Sharing of Ex-
perience," the second part of this special issue of
Studies in Family Planning, is a forum for thoughts,
methodologies, and learning tools in process. Those
studying and working with women have few oppor-
tunities to learn from one another; the production
and presentation of "papers" is sometimes too time-
consuming and formal a means of communication.
Further, it often excludes both as contributors and
as listeners those deeply involved in development
work. Thus we have asked a variety of people who
have undertaken research on and work with women
to share their experience in several areas: How does
one gain access to rural women? What do rural
women know and how do they report on it? What
kind of work do they do and how can it be meas-
ured? How do they perceive themselves and their
Useful information on rural women and its
proper interpretation acknowledges at the outset that
rural women are a heterogeneous group. Most of the

contributions here discuss rural women in terms of
(1) past and present roles-the generation they be-
long to, (2) class, (3) age, (4) participation in subsist-
ence or market-oriented agriculture, and (5) partici-
pation in on- or off-farm employment. Making such
distinctions is not only essential to competent infor-
mation gathering, but is the only basis on which pro-
grams and policies can be planned.
The specific ideas on conducting an interview,
the questions one might ask, and the methodological
techniques described are provided for contempla-
tion, modification, and, where appropriate, use by
those doing development work (research or action).
Which of these will be useful will depend very much
upon the setting in which work is being undertaken.
One must consider the socioeconomic conditions,
the past history of development efforts, the skills
available, the purpose of the work, and the budget.
The contributions here cover a wide range of settings
and possibilities, many of them low-cost.
We have asked the contributors to feel free to
write as if to colleagues and friends. We have asked
for and gotten the openness and, at times, the tenta-
tiveness that characterize the learning process. Many
contributors describe a limitation in their viewpoint
or a problem that initially interfered with getting full
information and the adjustment that was made to
overcome the problem.
Most contributions cover more than one of the
issues listed above. Although there are multiple con-
tributions for some countries, the problem being
studied is the real focus of each article. Readers are
not expected to read this section in its entirety, but to
be selective, using the following outline:


Authors Site of Report Title and Subject Matter

Group I: Methodological Issues

Pudjiwati Sajogyo Indonesia "Studying Rural Women in West Java"
Endang L. Hastuti Developing information on labor utilization pat-
Syarifah Surkati terns, household income, technologies utilization,
Winati Wigna decision making, and economic authority. Market
Krisnawati Suryanata oriented and home production.
Benjamin White
Carmen Diana Deere Colombia and Peru "Measuring Rural Women's Work and Class Posi-
Magdalena Le6n de Leal tion"
Sexual division of labor in peasant households. Mar-
ket oriented and home production.
Nancy Lee Peluso Indonesia "Collecting Data on Women's Employment in Rural
Off-farm employment.

Group II: Time Use and Project Planning
Brenda Gael McSweeney Upper Volta "Collection and Analysis of Data on Rural Women's
Time Use"
Framework and methodology. Application of find-
Vivian Havens Gillespie Nicaragua "Rural Women's Time Use"
Survey of time use of three types of women. Off-
farm employment and women's roles in market-
ing. Benefits of time use methodology.

Group III: Instruments for Learning
Audrey Chapman Smock Kenya "Measuring Rural Women's Economic Roles and
Contributions in Kenya"
The Division of Labour Module of the Central Bu-
reau of Statistics. A national sample survey instru-
of Statistics. A national sample survey instrument.
Nadia H. Youssef Indonesia "Learning about Women through Household Sur-
Coralie Turbitt veys: An Experimental Module"
An experimental module for application in an annual
survey of women's economic activities.
Michele Goldzieher Shedlin Mexico "Assessment of Body Concepts and Beliefs Regard-
ing Reproductive Physiology"
Women's perceptions of their bodies as determined
by their culture.
Taherunnessa Abdullah Bangladesh "Project-Oriented Research on Aspects of Women's
Sondra Zeidenstein Knowledge and Experience"
Question sets used to understand rural women's cul-
ture and develop information for rural action proj-
ects that reach women.
Achola Pala Okeyo Africa "Research Priorities: Women in Africa"
A set of hypotheses about women's access to re-
sources, experience of development, and culture,
to guide policy research.


Authors Site of Report Title and Subject Matter

Group IV: What Rural Women Know
Marsha Safai Iran "Circumventing Problems of Accessibility to Rural
Muslim Women"
Measuring the contribution to agricultural produc-
tion using interviews.
Gudrun Martius-von Harder Bangladesh "How and What Rural Women Know: Experiences
in Bangladesh"
Rural women's terms of description. Demands made
by production and agriculture on women's labor.
Joan P. Mencher India "Women in Rice Cultivation: Some Research Tools"
K. Saradamoni Participatory strategies for data collection.
Janaki Panicker
Hanna Papanek Indonesia "Research on Women by Women: Interviewer Selec-
tion and Training in Indonesia"
Interviewer selection and training. Experiences of the
research team.

Group V: The Voice of Rural Women

Leela Gulati India "Profile of a Female Agricultural Laborer"
Husband-wife differences in income-earning op-
portunities and patterns of spending.
Mary Elmendorf Mexico "Anita: A Mayan Peasant Woman Copes"
Family and economic roles of women in a village
seeking change.

Group I: Methodological Issues

Studying Rural Women

in West Java

Pudjiwati Sajogyo, Endang L. Hastuti,
Syarifah Surkati, Winati Wigna,
Krisnawati Suryanata, and Benjamin White

This note describes some methods used and some
problems encountered in a research project on "Rural
Household Economies and the Role of Women,"
which has been under way since mid-1977. The main
aim of this research is to describe the problems of
rural women in the household, in the labor market,
and in society and to achieve a better understanding
of the causes of these problems, as a contribution to
identifying policies and programs at the national, re-
gional, and local level that can remove some of the
barriers to their solution. We describe the main types
of information the study hopes to provide, various
initial assumptions that resulted in the selection of a
combination of research methods, some of our expe-
riences in the application of these methods, and fi-
nally some of our hopes and expectations regarding
the eventual utilization of research results. After a
year of data collection we are now in the midst of
data analysis and preparation of the first draft of the
main research report. These final stages of the re-
search process are an appropriate time for reflection
on both our successes and our failures in achieving
our objectives, and we hope our experiences can be
of use to those contemplating similar research in the

Types of Information Sought
The main types of information our research hopes to
provide include:
1. The labor-utilization patterns of rural women
within the household's division of labor, including
both work that directly provides income in cash or
kind and such activities as housework, child care,
and the like, which (for want of a better term) we call
"home production."
2. Household income, consumption, and ex-
penditure patterns in sufficient detail to relate them
to labor inputs and the household's control of land
and other resources.
3. Types of technology used in both income-
earning and home production work.
4. The role of women in decision making in the
household and in society, seen in the context of the
division of power and authority between the sexes
and between classes.
5. The nature and frequency of women's and
other household members' involvement in various
formal and informal institutions and relationships in



society. These include interhousehold relationships
involving exchanges of goods and labor; group and
community involvements, whether ceremonial and
religious or social and economic; and interaction
with the many village-level government institutions,
programs, and services in rural Java.
6. The aspirations of rural women and the bar-
riers to their achievement.

Assumptions Underlying
the Research Design

Our study covers a number of topics, but its main
focus is the desire for a better understanding of
women's work. In economies where the household is
the main unit of production, consumption, repro-
duction, and socioeconomic interaction, women's
activities cannot be understood in isolation but re-
quire analysis of the division of labor among all
household members. We therefore built our research
design around labor-utilization or time-allocation
analysis, which involves recording all activities per-
formed by respondents within a given reference pe-
riod based on respondents' recall, with repeated
interviews among the same sample of households at
regular periods throughout a complete 12-month ag-
ricultural cycle. It was clear to us that the accuracy of
the data depends greatly on the length of reference
period used. In any research of this kind, a compro-
mise has to be made between the desire to achieve
representativeness (requiring large sample sizes,
long reference periods, and frequent observations),
the demands for accuracy in the data (which is better
achieved with small samples, short reference peri-
ods, and the careful building of close relations be-
tween enumerators and respondents), and the re-
searchers' limitations on available time, personnel,
and research funds. It was already clear, however,
that if we were to obtain a reasonably accurate pic-
ture of the time devoted to home production work,
the research design should incorporate use of short
reference periods, which do not strain the respon-
dents' memory.'
We also assumed that rural women could not be
treated as a homogeneous group and that the study
of differences in their activities and problems (par-
ticularly those based on class, age, and education)

'For an introduction to the method of village-level labor-utilization
research, see John Connell and Michael Lipton, Assessing Village Labour Sit-
uations in Developing Countries (New Delhi: Oxford University Press, 1977).

was as important as the study of common char-
acteristics. It was thus necessary to incorporate suffi-
cient variation (particularly in socioeconomic status)
in the sample to permit analysis of these differences.
However, numerous tables showing statistical pat-
terns of time allocation, income sources, expendi-
tures, interhousehold interactions, and community
involvement are of little use without the kinds of in-
formation that provide some understanding of why
these patterns occur. This information must be de-
rived from the so-called qualitative techniques (al-
though they may also involve counting) of extended
interviews, case studies, observation, casual conver-
sations, and personal involvement in the commu-
nity. If questionnaires and quantitative data provide
the "bones" in research of this kind, qualitative par-
ticipant-observation methods provide the "flesh."
The desire to combine quantitative and qualita-
tive methods, and to document not only work and
incomes but also interactions and relationships, led
to the choice of a community-study approach rather
than a more widely scattered sample of respondents
covering a whole region. We selected two villages in
West Java in which some of us had already done re-
search on other topics.
One problem that became more clearly defined
as fieldwork and discussions continued is to distin-
guish between norm and reality. Researchers, re-
spondents, and the readers of research reports are all
influenced by their preconceptions and values con-
cerning both the actual roles of women and the roles
considered appropriate by them or for them. Norms
influence the perceptions of researchers and respon-
dents not only about what ought to occur, but also
about what actually does occur. To give some simple
examples, questions about a "normal" day's work or
diet, the "general" patterns of decision making in
the respondent's household, or his/her relations
with other households or institutions may yield re-
sponses far removed from reality, even when the re-
searcher has developed a close relationship of trust
with the respondent. Our awareness of this problem
led us to concentrate as much as possible on recent,
specific events rather than on general questions on
the same topics: not, "What do you normally eat?,"
"How do you usually spend the day?," "Who usually
makes decisions about x, y, and z?," "Do you attend
village meetings?," "How did you decide how much
fertilizer to buy and whether your child should con-
tinue school?," but rather, "What did you eat to-
day?," "What did you do today?," "Did you and
your husband discuss it?," "Did you agree?,"
"Whose view eventually prevailed?," "Did you at-
tend any village meetings last month?," and so on.


Research Methods Used

Since we had already conducted other research in the
two villages (which are about 11/2 and 6 hours' dis-
tance from Bogor), a basic household census and so-
cioeconomic survey were already available. Samples
of 60 households in each village were selected using
the landholding data from these earlier surveys, to
provide approximately equal groups of households
in each of three landholding classes: landless/near
landless, small farmers, and medium/large farmers.2
For one year (November 1977-October 1978) the
four members of the research team lived in the two
villages for ten days each month. During these ten
days all the sample households were interviewed,
with the help of local enumerators, using a long ques-
tionnaire divided into four parts. The first part covers
the activities of every household member age 6 and
above during two reference periods: (1) the 24 hours
preceding the interview, for which all work and non-
work activities and the time spent in each are re-
corded sequentially, thus providing the complete
"story" of a single day's activity, and (2) the entire
month since the last interview, with questions re-
stricted to work directly productive of income in cash
and kind and to a few other activities, such as com-
munal and exchange labor and school attendance.
The second part records for each household the
sources and amounts of income and items (and their
quantities and value) consumed or purchased by the
household, again using both the 24-hour and 30-day
reference periods; this is the longest and most diffi-
cult part of the interview. A third part records (using
the 30-day reference period only, since equivalent
information is covered in the 24-hour time-allocation
questions) the involvement of each individual in var-
ious kinds of interhousehold/community interac-
tions: communal/reciprocal labor, attendance and
contributions at neighbors'/relatives' ceremonies,
visits, recreation, meetings of various groups/associ-
ations. We include in this section some other infor-
mation on health and agricultural problems and the
efforts made to deal with them. The fourth part of the
interview concerns decision making and uses a dif-
ferent approach. No questionnaire is used, but the

'The stratification of households was subsequently revised, for two
reasons. First, we have had to make many corrections in the original land-
holding data. Second, a number of landless/near-landless households
(particularly in the village closer to Bogor) have access to pensions, sal-
aries, remittances, and other incomes much larger than their landhold-
ings would suggest. Our new criteria combine landholdings and levels of
nonagricultural income.

enumerators and research team use an interview
guide, which simply lists various areas of socioeco-
nomic life both inside and outside the household (di-
vided into eight main categories: production, con-
sumption, marketing, socialization, decisions on
family size, education, relations between house-
holds, and political participation). We wish to know
how decisions are made in all these areas and par-
ticularly how authority is divided between women
and men. The enumerators and researchers do not
ask all these questions at the same time or in the
same way, but attempt throughout the year to con-
struct a cumulative picture of decision-making pat-
terns. They seek opportunities for discussion with
men and women separately and with husband and
wife together, asking how decisions are generally
made and referring to concrete examples of recent
events. The enumerators and researchers have dis-
cussed their results and impressions throughout the
year, with the hope that the influence of norms can
be progressively reduced to provide the closest pos-
sible picture of how decisions are actually made.
One unusual aspect of this research is the use of
the two different reference periods. The 24-hour re-

call provides an acceptable level of detail on the
kinds of activities respondents cannot be expected to
remember over an entire month, particularly those
that are irregular, of short duration, or frequently in-
terrupted by other activities. This applies especially
to work in and around the house (handicrafts, repair
of tools, splitting firewood, and all home production
work). The 30-day recall provides a picture of the
whole month's activities, income, and expenditures,
although at the cost of much detail. This combination
also allows us to make a methodological contribution
by comparing results of these two methods with the
same respondents and enumerators, to allow estima-
tion of the levels of error involved when the longer
reference period is used.
The interviews last between one and two hours,
depending mainly on household size and the diver-
sity of activities and income sources. The interview,
although long and tedious, is fairly straightforward
and has been increasingly left to the local enumera-
tors, with the research team checking and discussing
each questionnaire on the following day with the
enumerators. The enumerators are mostly women (9
out of 11), from the middle and upper-middle (but
not the upper, "power-elite") class; 7 are school
teachers, 1 a social worker, 1 a student, and 2 are
housewives. They conduct one or two interviews
daily during the ten-day period, and each is respon-
sible for the same group of households every month


(10-12 households for each enumerator). Interviews
generally take place during the evening when all
household members are usually home. They are
often conducted in the kitchen so as not to interrupt
respondents in their work, and the atmosphere is
generally informal. When a household member is ab-
sent, the others can often provide the information on
his/her activities, although it is sometimes necessary
to arrange a second visit.
During their ten days in the village each month,
the researchers divide their time among several ac-
tivities: editing and discussing the enumerators'
work (this requires considerable time, even though
each researcher has to edit only three or four inter-
views each day); spending several hours each day in
informal, extended interviews, participant observa-
tion of daily life and various special events, collect-
ing basic life histories of selected respondents, and
writing daily fieldnotes based on these observations;
and occasionally joining the enumerators in their
regular interviews. On return to Bogor these field-
notes are typed and copies distributed to members of
the research team.
The system of repeated visits to the same sample
allows the occasional addition of specific questions
that do not need to be asked more than once: for ex-
ample, short questionnaires on marital/fertility his-
tory, division of land and other assets between male
and female heirs, and various details of household

Methods of Fieldwork, Analysis,
and Presentation

Anyone contemplating small-scale research of the
kind we have described should remember that the
amount of work and expense involved is no less than
that required by many conventional surveys with
much larger samples. We have spent one year study-
ing only 120 households in two villages, but the
amount of data involved is enormous. If each lengthy
household interview were divided into sections-1
for the household-level information on income and
consumption and 4 per household (on average) for
the individual-level data-each subdivided again
into daily and monthly recall, and the whole process
performed 12 times during the year, each of these
sections is equivalent in scale to many conventional
"one-shot" questionnaires, and the total effort is
therefore comparable to a conventional survey of

about 15,500 (120 x 5 x 2 x 12) respondents.: To col-
lect and process this small-scale information, 4 re-
searchers, 11 enumerators, 1 programmer, and 8 data
transferrers have worked mostly full time for over a
year. It is therefore necessary to be aware of the
heavy demands on time and resources and not to be
overambitious. A small-scale approach does not re-
duce time or expense; its main advantage is rather to
allow greater detail and reliability in data collection
and the opportunity to place this material in its
societal context by incorporating more anthropologi-
cal, participant-observation, and community-study
Reliability, of course, is not guaranteed by
working on a small scale. As in all research, the real-
ity we are looking for (events, attributes, attitudes,
or relationships) is not directly accessible to us, but
(as the term "data" implies) is "given" to us, never
perfectly, by the respondent. Reliability thus de-
pends entirely on the nature of the relationship be-
tween researcher and respondent, and on whether
that relationship assists or deters respondents in re-
porting the experiences to which they alone have di-
rect access. In our research, a three-way relationship
between researchers, local enumerators, and respon-
dents is involved. Our enumerators were conscien-
tious and reliable and became personally interested
in the research. The fact that they were local resi-
dents, interviewing neighbors with whom they also
interact in daily life, was in many respects an advan-
tage. Any unusual response could more easily be
spotted, questioned, and explained, and respon-
dents were also more willing to tell enumerators than
the researchers about feelings of suspicion or bore-
dom occasioned by our repeated monthly questions.
We tried to be completely open when responding to
questions about the purpose of the study; the enu-
merators were invited to our Bogor office for a week-
end to see what happens to the data they collect. This
has helped to increase their interest and that of the
respondents, to whom they later recounted their ex-
periences. Respondents were not given any formal
recompense for our regular intrusions into their daily
lives; the researchers on several occasions brought
small personal gifts: photographs of each household,
seeds for the garden, soap, tea, cigarettes, exercise
books for the children, and so on.
Even with the best intentions among respon-

'Paradoxically, if we had carried out a one-shot sample survey of
15,500 respondents scattered throughout West Java (or Indonesia), our re-
search would probably have appeared far more impressive and convinc-
ing to policymakers, even though we do not think such a study would
have provided them with much reliable or relevant information.


dents and researchers and the best possible personal
relationships between them, reliability remains an
elusive goal. This applies equally to the straightfor-
ward reporting of uncontroversial events and to the
more sensitive areas of decision making and authori-
ty. This leaves the researcher with serious problems
of interpretation, as the following examples may in-
Our trial analysis of one month's data for time
allocation found that estimates of time spent in in-
come-producing activity (excluding home produc-
tion) are generally between 30 percent and 60 percent
higher when based on 24-hour rather than 30-day re-
call.4 We have no reason to assume there were any
deliberate omissions in the 30-day recall or deliber-
ate additions in the 24-hour recall. We are inclined to
believe the 24-hour recall comes closer to reality, but
we cannot say whether either of these conflicting re-
sults is "accurate." The two methods produce very
different pictures of labor utilization; we will present
both in our reports, with the various reasons for pre-
ferring one over the other, but what policymaker
wants to be bothered with such details? On the other
hand, how many other reports of labor-utilization
surveys, carried out with less care and using a single
method, have reported their data as "facts," without
any warning of their probable unreliability?5
The question of men's and women's decision-
making and authority patterns in the household and
in society is clearly more complex, since, like all mat-
ters concerning the allocation of power and
authority, it is likely to be sensitive, to involve con-
flicts, and to be surrounded by private and public
norms. These norms may influence informants' per-
ceptions and responses concerning not only what
ought to occur, but also what does occur. Responses
to apparently direct questions about how decisions
are made may reflect a number of "levels of concep-
tion"6-the prevailing expressed norms, the respon-
dent's perception of the interviewer's norms, or the
respondent's privately held norms of appropriate be-
havior-or they may express precisely what we wish
to know: the respondent's account, as accurate as
possible, of what has actually occurred in his/her ex-
perience. Our approach was intended to allow us to

'The discrepancies are different for each sex and class in each vil-
lage, although in all cases the 24-hour estimates are higher. In the 12
groups the discrepancies range from 12 to 132 percent, and lie in the 30-60
percent range for 7 groups. After analysis of the entire year's data, we will
report in more detail on these differences, including identification of the
types of activities in which the greatest discrepancies occur.
'See B. White, "Population, involution and employment in rural
Java," Development and Change 7 (1976): 267-290; and G. Hart, Labor Allo-
cation Strategies in Rural Javanese Households (Ph.D. dissertation, Cornell
University, 1978), Appendix A.

gradually peel away the various layers of "norms" in
the search for actual patterns of decision making and
authority in the sample households (without dis-
carding the layers of "norms" since we are also inter-
ested in them and in their influence on behavior).
Our experience with this approach has been
mixed. On the one hand, we feel that in many cases,
through informal discussions with husband and wife
both separately and together and through hearing of
and directly observing specific events, we have
achieved a realistic picture, although it is unavoida-
bly an impressionistic one. On the other hand, in
many other cases we feel that we still do not really
know how decisions are made. We have, for all
cases, information in the form: "According to Mrs.
Kartika (or Mr. and Mrs. Kartika), decisions about
production expenditures are made jointly (or, are
made by Mr. Kartika alone, made jointly but mainly
by Mr. Kartika, and so on)." However, this informa-
tion is often inconsistent with various other less sys-
tematic data from or about the same individuals,
which we can use to gradually derive conclusions but
which is not available for the whole sample or in the
same detail for all of them. Some examples from our
fieldnotes may help to illustrate the problem.' Read-
ers who have not attempted this kind of research
may want to consider how they would interpret the
following scattered items of "data," all from or about
our female respondents:

1. "I have three regular male laborers for hoe-
ing and six regular women for transplanting. .. I
give them priority in harvesting before hiring others.
They often borrow money and pay me back with la-
bor. Also, I often ask them to help around the house,
fixing a leak in the roof, cooking when there are
guests, and doing other work ....
"After the harvest was destroyed by plant-hop-
pers, I owed $25 to the Bimas [agricultural credit]
program, but I've paid more than half of it back in
installments. I think the Bimas program should be
continued, but what to do when the harvest fails?

'The phrase is from Robert R. Jay, Javanese Villagers: Social Relations
in Modjokuto (Cambridge, Mass.: MIT Press, 1969), pp. 26-28. He illus-
trates the problem by describing the three types of responses he received
to the seemingly simple inquiry, "In this village, whom does a host invite
to a slametan [ceremonial meal]?" The various responses (depending on
who was present at the interview, and his own relations with the inform-
ant) differed not only from one another, but also from his own observa-
tions at the slametans that he was able to attend.
'We have intentionally selected examples that contrast sharply with
the conventional picture of rural women's roles; in many of our other ex-
amples, women and men conform more closely to the norm both in atti-
tude and behavior. The point is merely that it is difficult to obtain a
systematic picture of these variations when the available information is


"I don't dare to borrow any more, when I still
have a debt."
2. "How many people do you work for each planting
"Fifteen. But I'm not happy with this life of farm
work; it's too insecure and the agricultural prices and
wages are getting lower compared with the price of
things we buy from the towns. If there was a factory
here, I'd work there."
3. "I got married a lot, ten times altogether.
. My husbands told me they didn't want me to
work as a trader, but I didn't listen. I built this house
two years ago, sold some land to pay for it, it was all
my own decision."
4. "When you work at Sanip's house, how much do
you get paid?"
"Ah, that Sanip has no kindness; he thinks we're
no different from animals. We just get fed each day,
nothing else."
5. "I borrowed $5 for trading capital, now it's
increased to $9. I also borrowed $160 from our hamlet
cooperative to buy some land. Just recently I built a
brick house, which cost me $730. So now I have to
find $13 every month to pay my debts. Women are
more persevering than men in such things."
6. "I was married at 19, pregnant three months
later; the second child came three years after that.
Since then I've prevented any more; long before the
Family Planning Program I was going to the local
midwife for massage whenever my menstruation was
late. Then I took the pill and tried an IUD, but they
made me ill. I used many local medicines, too. My
husband said he'd get a vasectomy, but the doctor
wouldn't do it because he was too young and has
only two children. So I think I'll ask for the operation;
I'll tell him my health is too bad to have any more
7. "My main worry now is the paddy, what to
do about all these pests and diseases. I'm glad we
have the hamlet cooperative. I got a loan to buy a
sheep and pay for my cultivation costs. The coopera-
tive's paddy store started with some rice, which we
women saved every day in small amounts. Now
there are six tons in store, and $290 in cash that we
can borrow. I'm not a member, my husband is."
8. Bu Icah got very angry when she found that
the paddy seed she had been sold was not IR 36 but
IR 32 [two different high-yielding varieties]. She
knew only when it began to ripen much later than the
rest. She told everyone she met that she'd been
cheated by Iding [the seller]: "That's what the rich do;
they don't help poor people, just cheat them."

9. Bu Ninah made improvements to the house
with no help from her husband. He wasn't involved
at all in designing the windows and door, buying the
materials, finding the money to pay the craftsmen.
He just sat in the house idling, not even helping her
oversee the workers.
10. "I've gone to the village secretary and asked
him to pass a request to the village headman that
women should be encouraged to attend village meet-
ings and to speak their view. The secretary said noth-
ing, but didn't pass on my request."

Each of these scraps of information and many
others like them tell us something about the roles of
individual women in decision making, managing a
farm, recruiting and organizing laborers, entering
formal and informal credit relations, choosing an
occupation, family planning, and making major deci-
sions about investment. They also show individual
women holding and expressing clear views about
various government programs, class relations, re-
sponsiveness of village officials, and so on. If we
could provide systematic evidence that significant
numbers of rural women are greatly involved in deci-
sions relating to production, investment, and house-
hold and community welfare, it would have con-
siderable implications for many government pro-
grams, which treat such matters as men's affairs. The
problem is that, even with our small sample of 120
households, this concrete information has not been
forthcoming in many cases. We feel (but cannot dem-
onstrate) that many responses still reflect norms,
since these responses are often quite inconsistent
with more concrete data in the cases where both are
available. How, then, do we present such data? Most
readers and particularly policymakers want "proof"
and "hard facts," not a few selected cases, however
informative they may be. In preparing our final re-
port we have not yet discovered an appropriate way
of presenting these data. Since the report is to be di-
rected mainly to policymakers, its results and conclu-
sions should be as clear and unambiguous as
possible. On the other hand, to satisfy our own
standards as researchers, we wish to describe hon-
estly the problems in analyzing and interpreting the
different types of information we have collected, at
the same time indicating our personal feelings based
on long experience in the villages about what we
sense to be the correct interpretation.
Some months ago, before fieldwork was com-
pleted, we invited representatives of various govern-
ment agencies to a small seminar at which we
presented some preliminary impressions and conclu-
sions of our research and what we saw to be their


policy implications. This was an extremely useful ex-
perience.8 By formulating and discussing our prelim-
inary conclusions, we identified many areas in
which we lacked necessary information, while there
was still the opportunity to collect it or to modify our
plans for analysis. Discussions with government offi-
cials made us more aware of the limited range of pol-
icy alternatives available to tackle the problems we
set before them, and in general gave us a more pessi-

'Not least because it stimulated interest in our research. Shortly af-
ter this seminar, the newly appointed Junior Minister for Women's Af-
fairs agreed to provide funds for a national seminar on women in rural
development, to be based on presentation and discussion of our research
and its implications for both researchers and policymakers, and planned
for early 1980.

mistic view of the potential impact of research
on government policies directed toward rural

ACKNOWLEDGMENTS Our research, under the
sponsorship of FAO/SIDA and still in progress, is
being conducted jointly by two institutions: the
Rural Dynamics Study, Agro-Economic Survey of In-
donesia (Project Leader: Dr. Rudolf Sinaga) and the
Institute of Rural Sociological Research, Bogor Agri-
cultural University (Director: Professor Sajogyo).

Measuring Rural Women's Work

and Class Position

Carmen Diana Deere
Magdalena Le6n de Leal

This note considers some of the methodological
problems of measuring rural women's economic par-
ticipation by sample survey. It draws upon our expe-
rience in researching the economic roles of rural
women in a national-level study in Colombia and in
a regional-level study in Peru.' The specific objective
of the sample surveys was to quantify the existing
sexual division of labor among the peasantry. The
unit of analysis was the rural household, and the
focus of measurement was the division of labor by
sex in such activities as daily maintenance, house-
hold production, and incoming-generating activities
pursued outside the household.
The theoretical framework upon which both

'The theoretical and methodological designs of the two projects are
elaborated in the following: Magdalena Le6n de Leal and Carmen Diana
Deere, "Estudio de la mujer rural y el desarrollo del capitalism en el agro
Colombiano," Demografla y Economia 12, no. 1 (1978); Magdalena Le6n de
Leal and Carmen Diana Deere, "Planteamientos te6ricos y metodol6gicos
para el studio de la mujer rural y el process de desarrollo del capitalism
agrario," in Informe Final de la Investigacidn "Acerca del trabajo de la mujer en el
sector rural Colombiano," ed. Magdalena Le6n de Leal (Bogota: ACEP, forth-
coming, 1980); and Carmen Diana Deere, "The development of capitalism in
agriculture and the division of labor by sex: A study of the northern Peru-
vian Sierra," Ph.D. thesis, Department of Agricultural and Resource Eco-
nomics, University of California, Berkeley, 1978.

studies were based sought to integrate the analysis of
the sexual division of labor in rural areas with the
broader analysis of agrarian change. Specifically, the
studies sought to isolate the effect of the process of
capitalist development in rural areas on peasant
households and rural women's work. We conceptual-
ize capitalist development as the development of a
wage labor force and of capitalist units of production
that rely on the purchase of wage labor. Our key con-
struct for identifying this process is the social differ-
entiation of the peasantry, through which peasant
households become either proletarian, selling their
labor power, or petty capitalist, employing wage la-
bor in the productive process. The basis for social
differentiation is the unequal access to the means of
production among direct producers in rural areas.
Cross-sectional quantitative analysis is required
to relate the process of rural class formation, based
on differing access to the means of production
among peasant households, to differences in the sex-
ual division of labor in productive and reproductive
activities. The quantification of the sexual division of
labor at a given moment in time also allows historical
analysis of different socioeconomic processes of
change to inform regional differences in women's


work. In addition, the analysis of different forms of
land tenure and labor market structures and of the
varying options that rural men and women face can
be related to differences in the sexual division of la-
bor and in family reproduction strategies, exhibited
in family structure and composition as well as fertil-
In the following, we relate some of the problems
that we experienced in the elaboration of the sample
survey questionnaire, and in the design of the sam-
ple survey itself.2

The Survey Questionnaire

We consider it imperative that the design of a sample
survey questionnaire be based upon prior fieldwork
in the designated regions. This is so not only because
of possible sociocultural differences among regions,
but also because the concrete study of different his-
torical processes should amend and enrich the hy-
potheses that guide the measurement effort. In the
Peruvian study, the questionnaire was formulated
after nine months of participant observation and
open-ended interviews in peasant communities and
agrarian enterprises. In the Colombian study, the de-
sign of the survey questionnaire was undertaken af-
ter a one-year qualitative research effort. This
consisted of three months of intensive fieldwork in
each of four regions of the country by teams of two
researchers.3 The fieldwork was followed by detailed
historical analysis of the regions based on secondary
sources. Finally, the analysis of the primary and sec-
ondary data resulted in the preparation of regional
monographs on the historical process of agrarian
change and of changes in the sexual division of la-
bor. This prior research effort provided the basis for
the design of the quantitative research stage.
An example of the importance of prior fieldwork
and analysis in informing both the hypotheses under
study and the most appropriate formulation of the
variables to be measured is provided by the concep-

'Many of the points in the following section that refer to the Colom-
bia project are elaborated in Le6n de Leal and Deere, "Planteamien-
tos .," cited in note 1; and Carmen Diana Deere, Jane Humphries, and
Magdalena Le6n de Leal, "Class and historical analysis for the study of
women and economic change" (Geneva: Role of Women and Demographic
Change Research Program, International Labour Office, mimeo, March
:'In the initial stage of fieldwork in the Colombia project, open-ended
questionnaires were utilized in the interviews carried out with peasant
households and with rural employers and wage workers. The open-ended
questionnaires for peasant households were revised and reformulated con-
siderably in the construction of the closed questionnaires for the sample

tualization of peasant household activity. The peas-
ant household may be conceptualized as a unit of
production and reproduction; yet the household and
the unit of direct production may not be coter-
minous. In the Peruvian case, the questionnaire re-
flected a conceptualization of the peasant household
as an undifferentiated unit of direct production and
of reproduction of labor power. To distinguish be-
tween the kinds of activities carried out by various
household members, the following delineation was
used: household maintenance activities required for
daily reproduction; activities geared toward direct
production of use or exchange values (agricultural
production, agricultural processing and transforma-
tion, animal raising, artisan production); and activi-
ties carried out external to the household, detailed
according to the relations of production or distribu-
tion (labor market participation, reciprocal labor ex-
change, and petty trade). In the design of the
Colombian questionnaire, we found it appropriate to
conceptualize the household as the particular arena
of domestic maintenance activities, separate from the
unit of economic exploitation, the farm. This is partly
due to the forms of usufruct of land that have devel-
oped historically, so that the household itself is
sometimes spatially separated from the unit of eco-
nomic exploitation. Further, there was not necessar-
ily a direct correspondence between the household
as a unit of reproduction of labor power and the com-
position of the labor force that carries out the produc-
tive activities. The different forms and organization
of production must be reflected in the design of the
questionnaire if data are to be captured accurately. In
the Colombian case, the household questionnaire
distinguished between three types of activities:
activities carried out physically within the house-
hold (domestic chores, artisan production, country
stores); activities associated with the unit of eco-
nomic exploitation (agricultural work, animal rais-
ing, product transformation and marketing); and
activities carried out spatially separate from the
household (wage work, labor exchange, and so on).
A separate questionnaire on agricultural production
was then filled out for each economic unit pertaining
to the household.
A second problem in the design of the question-
naire concerns the actual measurement of participa-
tion of household members in the myriad of
activities. Participation can be measured in terms of
time actually spent in an activity, of who generally is
charged with the activity, or of the average intensity
of participation. The most accurate measurement of
participation is one based on the actual time dedi-
cated to the activity. Obviously, to be most accurate,


this measure would require a day-by-day accounting
of the tasks in which each member of the family par-
ticipated and the amount of time required by each,
usually referred to as the time-allocation method. In
a sample survey to be carried out only once, the time
dedicated by various household members to the
series of activities is necessarily based on recall.
The recall method suffers from various handi-
caps. First, it requires the respondent to construct an
average measure of the time usually dedicated to the
activity. Here the reliability of the measure depends
on whether time is culturally relevant. Second, the
accuracy of the measure greatly depends on the time
unit of the analysis. The average amount of time
spent on daily activities is much more reliably meas-
ured than the average amount of time spent on activ-
ities that are engaged in only sporadically or season-
ally. Measurement of the average amount of time
spent in agricultural work is particularly difficult
given the seasonality of the activity and the varying
intensity of participation at different times of the
In the Peruvian questionnaire, two forms of time
allocation were measured by recall. First, household
members were asked the average amount of time
spent on a detailed listing of activities, based on the
relevant time frame (whether the activities are usu-
ally carried out on a daily, weekly, or monthly basis).
Second, a detailed participation schematic was con-
structed for agricultural work. This schematic re-
quired that the informant recall the actual persons
who participated in each agricultural task and the
number of days so employed. Given the extremely
small size of peasants' plots in the area, and the fact
that agricultural production absorbs a minimal
amount of labor, the recall method based on who ac-
tually participated in each activity was successful.
This method revealed that women were active agri-
cultural participants, whereas the first form of meas-
urement suggested that women dedicated a minimal
amount of time on average to agricultural pursuits
during the year.4
The success of the recall method for measuring
either actual participation in terms of labor days or

'Results of the different measures of women's agricultural participa-
tion in the Peruvian case are reported in Carmen Diana Deere, "La division
por sexo del trabajo agricola: Un studio de la Sierra Norte del Peru," Es-
tudios de Poblaci6n 2, no. 9 (September 1978). The Colombia results are re-
ported in Magdalena Le6n de Leal and Carmen Diana Deere, "Proletariza-
ci6n y el trabajo agricola en la economic parcelaria: La division del trabajo
por sexo en dos regions de Colombia" (paper presented to the Conference
on Women in the Labor Force, Rio de Janeiro, November 1978; forthcoming
in conference proceedings).

the average amount of time dedicated to an activity
greatly depends on the researcher's familiarity with
the average amount of time required by the various
tasks; that is, it largely depends upon previous par-
ticipant observation. We do not recommend meas-
urement by recall in questionnaires to be executed
by unspecialized interviewers; rather, we suggest
that this method is most successful when the princi-
pal researcher is carrying out the interviews.
An alternative measurement of participation is
the respondent's subjective evaluation of the fre-
quency or intensity of participation in a given activ-
ity. Each household member is asked whether he or
she always, usually, or never participates in a given
activity (after it is determined whether the activity is
applicable to the household). In the Colombian study
this measurement of participation was quite success-
ful for both household maintenance activities and
agricultural work. This form of questioning is readily
comprehended by the respondent and is easily ad-
ministered by an interviewer with minimal training.
It has the added advantage of capturing the par-
ticipation of family members who "help out" in the
activity, although they are not principally charged
with carrying it out.
While the focus of the questionnaire is on the
sexual division of labor, it necessarily must be re-
lated to another series of socioeconomic or cultural
variables for analysis. The selection of these vari-
ables should be derived from the hypotheses guiding
the study; yet considerable attention must be given
to narrowing the range of inquiry. The time con-
straint on the length of the questionnaire requires
that certain choices be made in terms of the comple-
mentary variables to be included. Such variables as
the formation of family income, family occupational
histories, family structure, and fertility histories are
lengthy projects in themselves, if they are to be done
correctly. For example, while fairly accurate meas-
ures of the average wage earned can be easily ob-
tained, the estimation of agricultural income or
income from animal raising requires a detailed ac-
counting of costs as well as disposition of output.
Rarely is an aggregate estimate of income from farm
production accurate; yet a good farm income ques-
tionnaire requires two to three hours to administer.

The Sample Survey Design

The most important problem in the design of a repre-
sentative sample survey is the selection of the


population to be sampled. The choice of population
must be compatible with the theoretical framework.
Since our interest was to measure the sexual division
of labor in terms of class formation, our sample had
to be representative of the different class strata in the
rural areas. Capitalist class relations are defined by
the purchase or sale of labor power. However, these
characteristics can rarely be anticipated; rather, they
are variables that are measured by the questionnaire
itself. Nonetheless, the differentiation of the peasan-
try is theoretically dependent upon the household's
access to the means of production. Insufficient access
to the means of producing the household's subsist-
ence compels proletarianization, while sufficient ac-
cess to the means of production to employ wage
labor allows a petty capitalist strata to emerge in the
rural areas. Thus access to means of production pro-
vides the most relevant proxy for measuring the
household's class position.
The measurement of access to means of produc-
tion ideally should be both quantitative and qualita-
tive. The amount of land held in property as well as
in usufruct, the quality of the land, the stock of tools,
equipment, and the number of animals, as well as
their productivity (or age), should be taken into ac-
count. This information is, again, rarely available
and must result from the survey itself. To define the
universe of the sample, one must take the most read-
ily available information, access to land, as a proxy
for the class configuration of the rural area. While
land held in property is the most adequate measure
of social differentiation, land held in usufruct (i.e.,
land held in property as well as in other forms of ten-
ure) more closely determines the generation of in-
come and the division of labor by sex in productive
activities in a given time period.
The choice of the population to be sampled is
also greatly constrained by the available data base.
This is a particular problem in rural areas, where cost
constraints due to the spatial distribution of holdings
do not permit more traditional techniques of deter-
mining the universe, such as blocking. In the Colom-
bian case, three possible sources of information were
available on the composition of the rural areas. The
1973 population census was based on the household,
while the 1970 agricultural census was based on the
units of exploitation; but neither census was avail-
able in disaggregated form to private institutions to
allow identification of the units to be sampled. The
third alternative was the use of a rural cadastre, or
municipal property listings, which were available
only for some of the municipalities. Since this was
the only data source that identified the agrarian

structure in terms of individuals, it was chosen as the
population for sampling.
All of the landholdings held in property in the
municipality were stratified according to size ranges.
Once the population proportion was known, it was
possible to determine the sample cell, taking into
consideration sufficiently large cell sizes to allow sta-
tistical manipulation, the statistical representation of
the sample of the population as a whole, and the cost
constraint on the size of sample. Municipal property
listings are a juridical classification and do not corre-
spond directly to the economic unit or, in many
cases, to one particular household. This is par-
ticularly problematic as regards large properties, in
which a single juridical unit may be constituted by
several economic units under different arrange-
ments, as well as by numerous resident households.
The Colombian procedure was to utilize the property
listings to arrive at the household: all constituted
households that resided within the administrative
property were interviewed.'
The Peruvian sample survey was constructed as
a follow-up to a large peasant household income sur-
vey." The parent survey was representative of the
distribution of landholdings in property. The follow-
up survey was thus designed to be representative of
the parent survey, which greatly facilitated the
choice of the population to be sampled as well as the
execution of the sample survey.

We have attempted to illustrate some of the con-
ceptual, methodological, and technical problems in-
volved in the quantification of women's economic
participation in rural areas of developing nations.
We stress again the importance of building quantita-
tive work on a solid qualitative foundation. Quan-
titative results are extremely important for under-
standing the sexual division of labor, but they must
be gathered and analyzed in terms of the substantive
historical processes that gave rise to the differences
in class formation and in the division of labor by sex.

'In the Colombia project this procedure resulted in a sample size of
200 properties in one municipality (representing 7.2 percent of the popula-
tion) and of 150 properties in the other municipality (representing 13.1 per-
cent of the population). Questionnaires were completed for 216 and 163
households, respectively.
"The 1973 Cajamarca Income Survey was carried out by the Socio-Eco-
nomic Group of the Cajamarca-La Libertad Pilot Project under the direc-
tion of Efrain Franco and funded by the Ford Foundation. This survey
consisted of 1,500 observations of peasant households in two provinces of
the Department of Cajamarca. The follow-up survey focused on 1,050 house-
holds in one province and was constituted through random, ordered selec-
tion of 105 households for interviewing.


ACKNOWLEDGMENTS This research note draws
upon our experience with the Colombia Rural
Women Research Project carried out by the Colom-
bian Association of Population Studies under the
direction of Magdalena Le6n de Leal, and upon the
dissertation fieldwork of Carmen Diana Deere. The

Colombia project, a three-year study, was financed
by grants from the Ford Foundation and the Rocke-
feller Foundation. Ingrid Caceres, Clara Gonzales,
and Liliana Motta participated in the formulation
and execution of the sample survey reported here.
Ms. Deere's fieldwork in the northern Peruvian Si-
erra was funded by a Social Science Research Council
Dissertation Fellowship.

Collecting Data

on Women's Employment

in Rural Java

Nancy Lee Peluso

Dramatic changes in rural lifestyles and employment
have accompanied modernization. Women are par-
ticularly affected, for many of the fields of employ-
ment traditionally dominated by women are being
eliminated or altered as new technology is applied,
especially in agriculture. Competition for the re-
maining labor opportunities is high, and both male
and female laborers are grossly underpaid. A single
source of income can rarely feed, clothe, and shelter a
family, let alone finance an elementary education for
the children. Poverty in Central Java means that
everyone willing and able to work does-if and
when there are job opportunities.
The increasing pressure on the land by a steadily
growing population, the resultant masses of landless
and near-landless rural dwellers, and the decrease in
agricultural wage labor opportunities since the intro-
duction of hullers, sickles, and hand weeders (all of
which absorb less labor than traditional work meth-
ods)' are causing more people to turn to occupations
in small-scale trade and manufacturing. However,
there has been relatively little research on these sec-
tors of the labor force, especially intensive research
that can be put to practical use.
My research was carried out to determine the
economic roles of rural women working outside the
agricultural sector and to create a system of job classi-
fication for small trade and industry that would accu-
'See Rudolf S. Sinaga, "Policy implications of agricultural mecha-
nization for employment and income distribution," Bulletin of Indonesian
Economic Studies 14, no. 2 (July 1978): 102.

rately differentiate between variations in the char-
acteristics and nature of women's economic activi-
ties. I decided to focus on women working outside
the agricultural sector because extensive research has
already been done on the agricultural sector in Cen-
tral Java and the Special Region of Yogyakarta,2
where I would also be working.

Research Methodology

This research was carried out in three phases, each
applying a different methodology or seeking a spe-
cific kind of data. The three phases were participant
observation, the household survey, and the market
survey. Because family roles and relationships were
of particular interest, the family unit received special
attention; because women's economic activities were
of primary concern, the survey respondents and key
informants were chosen according to the women's

'Among the many studies on agricultural employment are the fol-
lowing: Ann Stoler, "Rice harvesting in Kali Loro: A study of class and
labor relations in rural Java," American Ethnologist 4, no. 4 (November
1977); Benjamin White, "Production and reproduction in a Javanese vil-
lage," unpublished Ph.D. thesis, Columbia University, 1976; William L.
Collier, Gunawan Wirdai, and Soentoro, "Recent changes in rice harvest-
ing methods," Bulletin of Indonesian Economic Studies 9, no. 2 (July 1973);
and M. Singarimbun and D. H. Penny, Population and Poverty in Rural
Java: Some Economic Arithmetic in Sriharjo, International Development
Monograph No. 41 (Ithaca, N.Y.: Cornell University, 1973).


Participant Observation

This phase of the research was the most important
for several reasons. First, the aim of the research was
to obtain accurate details on women's family life and
work routines and the variables that caused changes
in these. Eight women were selected for case studies
on the basis of information obtained from the hamlet
head, other traders, or through my own acquaint-
ance. I accompanied each informant for five days
(one Javanese market-cycle week), from approx-
imately one-half hour after her rising until mid-eve-
ning (9- 10 P.M.). Research hours varied according to
the demands of each occupation. For example, one
trader of earthenware pots traveled 11-18 kilometers
to two different markets on four nights of the five-
day week, leaving home at either 12:30 or 2:30 A.M.
Observation began just after she had risen to go to
market and continued until the next evening. On the
fifth day of the market cycle, she bought stock in a
morning market 10 kilometers from her home; there-
fore observation on this day began in the morning.
Most of the other women began their income-pro-
ducing work between 4 and 6 A.M., and I adjusted my
hours to their schedules. I visited each potential in-
formant, accompanied by a mutual acquaintance
(ideally a former respondent), prior to the five-day
observation period; and if she agreed to participate,
we explained the importance of her conducting her
daily activities as usual. Perhaps because I had been
in the village for seven months before the research
began and many people at least knew of me, it was
not difficult for the informants to accept this. After
the first or second day the women tended to ignore
the extra attention they received and went about
their daily business as usual.
After the initial five days of participant observa-
tion, each informant was visited once a week for up
to six months afterwards, and all changes or consis-
tencies in routine were recorded. Each woman was
also asked to record, in a marked notebook, all of the
household's monetary and reciprocal labor inputs
and outputs during one month; if the woman could
not read or write, another family member was asked
to do so.
The second great advantage of participant obser-
vation was that it allowed me to get close to the
women themselves. Within a relatively short time,
usually one or two days, they would confide their
feelings and opinions to me, often without my ask-
ing. Specific information was collected through in-
formal interviews; that is, in conversations during

slow periods at home or in the market, on the way to
or from the market, or while sitting around with the
family. Copious fieldnotes were taken on their activi-
ties, hours, the people they dealt with each day, and
their relationships to each other, in addition to re-
cording their life and work histories.
The eight case studies were followed by two
months of participant observation in the local mar-
ketplace, a small trade center located between the
two study villages. Every day I sat and talked with
two or more traders for several hours. At the end of
the market day I went home with one of them and
spent two or three hours with her family. These vis-
its were particularly informative if older family mem-
bers lived in the same household or nearby, since
they were usually eager to talk about the market of
former times. Fifty brief case histories were collected
during these two months.
The third and most practical advantage of this
phase of research was that the case studies gave me
the insight needed to formulate questions for both
the household and market surveys. Domestic and
trading or production patterns observed in one or
more cases could be checked for their general validity
through the surveys, briefly discussed below.

Household Survey

The second phase of the research, a 200-household
survey, was carried out in the two adjacent villages
from which the case studies were chosen. House-
holds were selected according to the woman's pri-
mary occupation; thus our sample was biased in that
only families in which the woman 's primary occupa-
tion was outside the agricultural sector were inter-
viewed. Neither the village administrative offices
nor the Department of Commerce had registers of
names, addresses, or specific occupations by sex.
Therefore we compiled the sample by going through
lists of households with each of the 28 hamlet heads.
The sample was almost equally divided between
women working outside the home (such as in the
marketplace) and those who did income-producing
work at home; all respondents were married, aged
20-49 years, and had at least one child.
In addition to standard demographic data, we
collected responses to questions on division of labor
in the household, decision making in the family, the
woman's time allocation, and her occupation at key
points in her life (i.e., before marriage, just after
marriage, after the first child, and after many chil-


dren). This occupational data were perhaps the most
important data collected in the survey. The last ques-
tion on occupation at a specific point in the woman's
life was worded so as to determine whether there
had been any occupational change between the first
child and the time of the survey. The question was
occasionally misinterpreted, however, and might
have been worded more accurately.

Market Survey

The third and final phase of the research was a sur-
vey conducted in the local market. Of the usual 350
regular daily traders, 125 of the women, or one-third
of all the female traders selling each commodity,
were questioned on their daily trading routines,
home preparation of their merchandise, and their
trading histories in detail, including what they had
ever sold, in which markets, and why they had
changed if this was the case. Questions were also
asked on secondary occupations, land ownership,
participation in the rice harvest, and basic demo-
graphic information.
This survey was carried out at the end of the re-
search, by which time I had already collected quan-
titative data on when and qualitative data on why
women changed occupations, merchandise, or loca-
tion of work, as well as on their mobility and par-
ticipation in traditional and modern community
activities, such as harvest labor and the government
credit program for small-scale traders. Preliminary
analysis of key questions has been made, although
the final results have not been completely tabulated.

Research Analysis

The combination of these three approaches produced
a complete set of detailed and statistically supported
data that would not have resulted from the use of one
method alone. We now offer an abbreviated version
of one of the case studies to illustrate the type of
qualitative information that was checked against sur-
vey data. Indeed, the occupational categories and ta-
bles illustrating patterns of choice and changes of
occupation might have been calculated from survey
results alone. However, the strength of the frame-
work used and the reasoning behind it derives from
the detailed life stories that I collected during the
year and a half that I lived in the village. The logic
inherent in the decision-making process could not
have been understood without close contact with the
people and participation in the village society.

A Trade History

Bu Joyo sells tape, a snack made from cassava. She
has been manufacturing and selling it for about eight
years, since a few years before her fourth child was
born. She began trading in the marketplace as a teen-
ager, selling betel-nut leaves on her own in a market
near her aunt's house. After about a year of selling on
her own, she began to accompany her mother to the
city to buy salt, which they sold in the mountain
market near home. When the government began
controlling the price of salt, she switched to buying
unhulled rice. This she took home, pounded by
hand, and resold in a roadside market. She was mar-
ried by then, and when she had her first child, she
left the baby boy at her mother's house and picked
him up on the way home. Later, when rice hullers
became common in the village and pounding rice
was no longer practical or profitable, she changed
wares again. Long-distance selling was out of the
question, for it would require too many hours away
from home; she had two young children by that
time, and a third was on the way. She began to buy
mlinjo seeds and pound them into emping, a delicacy
that is fried crisp and eaten as a snack or with meals.
She pounded the seeds at home and could mind her
young children while working. Once every five days,
she took the emping to the district market and sold
them wholesale to a regular buyer. She did this until
the price of mlinjo seeds rose sharply, after which
she changed to making tape. This process entails
boiling cassava and mixing it with yeast, then pack-
ing it into an air-tight container and storing it over-
night to ferment. The soft, tangy-sweet tape is used
in iced drinks, fried in batter, or eaten plain as a
snack. Until her last child was born, she sold her
product in a marketplace an hour's walk from home.
She left home at 4:00 A.M. and walked the four kilome-
ters alone, carrying about 15 kilograms of tape in a
basket slung on her back. At 8:00 or so, when the
market was over, she walked another three km. to
the central Regency market and bought 20 more kg.
of cassava for the next day's market. She walked the
final four km. home. She explained that, after the
baby was born, she felt too tired to carry the child
with her to so many places. The market was not near
her mother's village, nor were her other children able
to mind the nursing baby during her normally long
hours away from home. She temporarily stopped
selling in the marketplace and instead delivered or-
ders to a group of nearby food stalls. She spent a few
extra hours at home each day, bagging tape in 5-,
10-, and 25-rupiah size plastic bags. At 4:00 A.M. she
fed her child and left him with his 12-year-old


brother while she made deliveries, which took about
an hour and a half. Her son went to market in the af-
ternoon to buy cassava. After the baby was weaned,
she returned to selling in the market because her
profits were higher and she liked the contact with
other traders. Being in the market frequently allowed
her to take advantage of unpredictable opportunities
to make a few extra rupiah, such as by selling fresh
vegetables brought in by local farm women. She has
been a trader all her life and is always seeking any
opportunity to add to the family income.

Occupational Classification

The goal of this research was to provide detailed in-
formation on women's employment outside the
agricultural sector. Cases such as that of Bu Joyo,
combined with survey results, revealed clear trading
patterns and groups of commodities traded in a simi-
lar manner. I was thus able to draw up a detailed
classification system for occupations in the fields of
small trade and home industry for the area of re-
The universality of these occupational catego-
ries,:' even if applied only to Java, cannot be guaran-
teed, because region-specific cultural and social
variables make generalization on such a complex
matter difficult. However, further work on this prob-
lem led to the formation of a series of questions that
could be included on any large- or small-scale census
or labor force survey seeking information on the
characteristics of the labor force in a specific region.

Choice and Change of Occupation

By using the occupational categories devised for this
region, comparisons could be made of patterns
shown by women engaging in various nonagri-
cultural income-producing activities, primarily in
trade. Cross-tabulations with marital status and the
age and number of children showed definite trends in
occupational choice according to the life-cycle stage

'For a complete discussion of these categories, see the chapter on
occupational classification in Nancy Lee Peluso, The Economic Roles of
Rural Women Working Outside the Agricultural Sector: The Case of Tlogoadi,
Slemnan, Special Region of Yogyakarta (Yogyakarta: Population Studies Cen-
ter, Gadjah Mada University, forthcoming, 1979). The census questions
themselves are found in Nancy Lee Peluso, Putting People into Boxes or Put-
ting Boxes Around the People. Approaches to Designing Occupational Categories
for Java (Yogyakarta: Population Studies Center, Gadjah Mada Univer-
sity, March, 1979)

of the woman's family. This was of special interest
because such trends went virtually unnoticed, partly
due to the lack of a relevant system of classifying
employment and types of trade.
Besides environmental changes incurred by de-
velopment over which an individual trader has no
control, other endogenous variables influence or
limit a woman's choice of occupation or the type of
goods traded. These variables include access to capi-
tal, work experience, trade-specific or product-spe-
cific knowledge, total number of hours required
inside or outside the home, physical labor inputs,
distance to markets or other selling outlets, and
potential capital returns per hour per day.
Simultaneously, these factors are dependent on
the development stage of the woman's family, par-
ticularly her marital status and the number and age
of her children. If the demands of the woman's fam-
ily conflict with those of her current occupation, ad-
justments are made in hours or work location, or she
changes goods or occupations altogether. Bu Joyo's
case is a clear illustration of this.


The practical value of the job classification system
and of findings on patterns of occupational choice is
that they can be used in planning regional develop-
ment projects geared to the utilization of rural
women's labor. The series of survey questions for
persons employed in small trade or industry will
allow for data comparison with other parts of Java,
Indonesia, and possibly other developing countries.

ACKNOWLEDGMENTS With the exception of the
household survey, this research was sponsored and
funded by the Ford Foundation, Jakarta, Indonesia,
to which I express my deepest appreciation. The
household survey was carried out with Dra. Partini
and Drs. Sutaryo, who at the time were assistant lec-
turers in the Department of Sociology at Gadjah
Mada University, Yogyakarta. It was funded and
sponsored by the University's Population Institute.
Dr. Alice Dewey was in Yogyakarta during my re-
search; to her I owe an irrepayable debt for her ad-
vice and willingness to exchange ideas on all aspects
of the research. The ideas on occupational classifi-
cation were developed with the assistance of several


colleagues and friends at the Population Institute
(now the Population Studies Center), particularly Dr.
Terence Hull, Dr. Valerie Hull, Marie-Claire Mal-
ingreau, and Patrick Guinness. Seminars and other
academic facilities were made accessible by Dr.

Masri Singarimbun, whose moral support through-
out the project was also much appreciated. And of
course to the women who opened their lives to me
and made this study possible, I cannot sufficiently
express my gratitude.

Group II: Time Use

and Project Planning

Collection and Analysis of Data

on Rural Women's Time Use

Brenda Gael McSweeney

The Upper Volta/UNESCO/UNDP Project for Equal
Access of Women and Girls to Education, a ten-year
systematic program to promote educational opportu-
nity for girls and women and to increase their
contribution to the economic and social develop-
ment of the country,' was unique in Africa. The
project, one of 37 women-related activities under-
taken with the assistance of UNESCO during the
decade 1965-75, was aimed primarily at creating the
preconditions for educating women in remote rural
areas and at designing education programs that
would contribute to rural development." It was multi-
sectoral and was planned from the outset to last ten
Launched in 1967, the project assembled data on
obstacles preventing full access of girls and women
to education and initiated experimental programs in
several pilot zones to overcome these barriers and
hence to augment available educational possibil-
ities.4 Initial operations, based upon findings of so-
ciological studies undertaken in three regions of the
country with different levels of economic prosperity,
climatic conditions, and ethnic groups, sought to

'Republique de Haute-Volta, "Plan d'operation. Projet exper-
imental pour I'egalite d'acces des femmes et des leunes filles a
1' education," p. 11.
"1Rapporteur's report on UNESCO's contribution towards improv-
ing the status of women," submitted by the United States Executive
Board Member, Paris, May 1975, p. 46.
:'Women, Education, Equality: A Decade of Experiment (Paris: The UN-
ESCO Press, 1975), p. 12.
'Republique de Haute-Volta, cited in footnote 1, pp. 1-2.

lighten women's workloads and to improve basic
health conditions and standards of living. Labor-sav-
ing technologies were introduced: mechanical grain
mills, carts, and readily accessible water wells. It was
thought that women could allocate a portion of the
time thus saved to such educational activities as
learning modern agricultural methods, health and
civic education, and professional training. Knowl-
edge was to be disseminated through animatrices, or
dynamic village women, and traditional midwives
who were designated by the villagers themselves to
attend special courses that would enhance their roles
as leaders in change. Functional literacy classes and
radio clubs were organized in the villages. The proj-
ect also introduced activities to increase women's
revenues, such as collective fields, the receipts from
which were managed by the women.

Collecting Data
on Rural Women's Time Use

A major objective of the data collection was the gen-
eration of precise information on women's time
allocation. Considerable accuracy was necessary to
permit testing of hypotheses concerning women's la-
bor inputs and the availability of free time to women
in comparison with their male counterparts; to exam-
ine women's extrafamilial and intrafamilial alloca-

___ -1


tion of time and the implications of such time use for
development policies and programs;5 to record activ-
ities in which women's participation was substan-
tial, in order to assess whether the project's training
programs and technologies were addressing the ap-
propriate problem areas (e.g., should priority be
given to increasing efficiency and easing the burdens
of sowing, weeding, threshing, hulling, or grind-
ing?);' to analyze whether time pressures7 were in
fact an obstacle to women's access to education; and
to identify patterns of free time in order to improve
scheduling of such activities as radio listening
groups and literacy classes. Another objective was to
obtain information on variables influencing time-use
patterns and behavior.
In order to meet these various requirements as
efficiently as possible, research resources were allo-
cated to a combination of overview and intensive
survey techniques. Target and comparable control
villages were selected in the three zones reached by
the project, and, following the preparation of village
monographs, initial questionnaires were admin-
istered verbally to all women to gather basic personal
and demographic data, as well as data on schooling
of children, availability and utilization of tech-
nologies, work and earnings, daily activities (by re-
call), and assistance in carrying out workloads. More
detailed information was then collected from a ran-
dom sample of 30 women in each village and from
women leaders in the project villages on available re-
sources, agricultural workloads, opinions concerning
technology, and the impact of technology on time al-
located to various tasks and alternative use of time
thus saved. Parallel questions were asked of the hus-
bands of sample women, their responses to be
treated as variables possibly affecting their wives' at-
titudes and behavior.
Another major component of the data-collection
strategy was the use of time budgets. Three cross-
seasonal time budgets comprising all activities in ap-

5On this theme, see United Nations, World Conference of the Inter-
national Women's Year, Mexico City, 19 June-2 July 1975, The Role of
Women in Rural Development (E/CONF.66/BP/11), prepared by the Food
and Agriculture Organization of the United Nations, 24 March 1975, pp.
"Emmy Simmons has pointed out limitations on the usefulness of
time-allocation studies for development practitioners interested in de-
signing programs to facilitate active participation of women in develop-
ment. For example, knowing that women allocate more time to farming
than their husbands does not indicate whether they will respond to im-
proved seed varieties, nor does it furnish information on their relative
roles in agricultural decision making. See Emmy Simmons, Economic Re-
search and Women in Rural Development in Northern Nigeria (Washington,
D.C.: Overseas Liaison Committee of the American Council on Educa-

proximately the first 14 waking hours were prepared
for each of the women in the village samples and for
women leaders by means of direct observation. They
encompassed the recording in minutes of when the
activity began and ended, a description of the activ-
ity, the technique or technology used, and any assis-
tance in carrying out the activity. In the north-central
zone, which is populated by the Mossi, the majority
ethnic group, three time budgets were also prepared
for five men from each village, to permit analysis of
the sexual division of labor. In addition, a single ob-
servation apiece of five girls and boys from each vil-
lage, stratified by age, was undertaken to furnish
indications of the phasing into work. Comparison of
information on rural women's time use yielded by
the recall technique and by direct observation
showed that some 44 percent of women's work was
unaccounted for using recall.

Proposed Framework
for Analyzing Time Allocation

To quantify the sexual division of labor in rural
areas, the African Training and Research Centre for
Women (ATRCW) of the United Nations Economic
Commission for Africa has designed a framework
of informal employment indicators that measure
the comparative participation of men and women
in a number of tasks: food production, storage, and
processing; animal husbandry; marketing; brewing;
water and fuel supply; child care; cooking and clean-
ing; house building and repair; and community self-
help projects." Based upon actual time allocations to
these tasks by women and men in a minisample from
Zimtenga, a village in the north-central zone of Up-
per Volta, women assume 80 percent of this work-
load, including 61 percent of food production. Yet

tion, 1977), p. 2. In the Upper Volta survey, questionnaire response and
views expressed in group interviews furnished complementary informa-
tion relevant to central research questions.
'Philip J. Stone reviews time pressures experienced by women in
both socialist and capitalist countries, based upon the results of a multi-
national study using time-budget technique, in "On being up against the
wall: Women's time patterns in eleven countries," in Public Policy in Tem-
poral Perspective, ed. W. Michelson (The Hague: Mouton, 1979).
BFor a development of the rationale for, and a complete description
of, the set of indicators, see United Nations, Economic Commission for
Africa, The New International Economic Order: What Roles for Women?, espe-
cially the section entitled "Expediting balanced development: Measuring
and monitoring the participation of women as compared to that of men,"
pp. 26-31, plus the annex.


TABLE 1 Comparison of time allocations to rural activities, by sex

Average time
(in minutes)a
Women Men
367 202
178 186
69 4
35 108
39 6
30 19
4 2

A. Production, supply, distribution
1. Food and cash crop production
Weeding, tilling
Travel between fields
Gathering wild crops
Other crop-production
2. Domestic food storage
3. Food processing
Grinding, pounding grain
Other processing activities
4. Animal husbandry
5. Marketing
6. Brewing
7. Water supply
8. Fuel supply
B. Crafts and other professions
1. Straw work
2. Spinning cotton
3. Tailoring
4. Midwifery
5. Other crafts/professions (e.g.,
metal work, pottery, weaving
cloth, beekeeping, etc.)
C. Community
1. Community projects
2. Other community obligations

this schedule of work accounts for an average of only
two-thirds of the total occupations of this sample of
women in the first 14 hours of the day, and only one-
quarter of the activities of their male counterparts.
The Upper Volta data thus suggest the need to revise
the ATRCW framework to better capture rural activi-
ties by incorporating into the work schedule cash
crop production and crafts, and by including meas-
ures of time for personal needs and of free time.
A more comprehensive set of indicators (see
Table 1), based upon an expansion of the ATRCW's
proposed framework, permits examination of the
dominant patterns of overall time use and allows de-
termination of whether men score higher on a more
inclusive schedule of work than on the initial sched-
ule of work suggested by the ATRCW. The revised
framework in Table 1 displays the average time in

D. Household
1. Rearing, initial care of children
2. Cooking, cleaning, washing
3. House building
4. House repair
E. Personal needs
1. Rest, relaxing
2. Meals
3. Personal hygiene and other per-
sonal needs
F. Free time
1. Religion
2. Educational activities leadingg to
read, attending a UNESCO
meeting or class)
3. Media (radio, reading a book)
4. Conversation
5. Going visiting (including such so-
cial obligations as funerals)
6. Errands (including going to pur-
chase personal consumption
goods, such as kola, next door)
G. Not specified

Average time
(in minutes)a
Women Men
148 4
18 0
130 1
0 0
0 3
158 269
117 233
21 29

20 7
77 118
2 6

17 4
0 14
14 69

43 19

235 387

aBased on time budgets prepared by direct observation.
bWhen observation did not last the full 14 hours.

minutes allocated to each activity by the minisample
of women from Zimtenga in comparison with their
husbands. According to the new schedule of work,
women carry out 64 percent of the production/distri-
bution/supply tasks, 23 percent of crafts and other
professions, 97 percent of household tasks, and 23
percent of community obligations. These tasks repre-
sent 56 percent of all work performed in the first 14
hours of the day. Women's workloads after the ob-
servations ceased can also be expected to exceed
those of men, as women then generally prepare the
evening meal and wash up afterwards. A significant
fact that emerges is that expansion of the food pro-
duction category into a more comprehensive one in-
cluding cash crops results in measurements showing
a decrease from 61 percent to 49 percent of women's
participation in farming relative to that of men,

Total work (A, B, C, D)
Total personal needs and free time
0 35 (E, F)


owing to significant amounts of time allocated by
men to cotton production.


Based upon analysis of the minisample, provisional
findings are presented here concerning time pres-
sures experienced by Voltaic women and the influ-
ence of polygyny, age, and technology on time use.
These findings suggest directions for the analysis of
the entire data set.

Time Pressures and Education

The project sought to promote women's education by
encouraging the sending of more girls to school, and
by training women leaders who would in turn edu-
cate their colleagues." Women's workloads and time
patterns could be assumed to greatly affect the
achievement of these objectives. Time-budget data
showed that Mossi women have only 1.3 hours of
free time in the first 14 waking hours. Given the mag-
nitude of women's workloads and the importance of
the assistance furnished by girls in carrying out daily
tasks, it is little wonder that the project team empha-
sized the introduction of technologies to lighten the
food-processing and diverse portage tasks in an at-
tempt to create time in which women might benefit
from educational opportunities and so that they
might be more amenable to having their daughters
and other young female helpers attend school.


The total time allocated to the food production/sup-
ply/distribution tasks was higher for women with
monogynous husbands than for those in polygynous
households. For women without cowives, total
workloads, including household work, community
activities, crafts, and other professions, averaged
111/3 hours, compared with 10 hours for women with
cowives (or only 8.8 hours if exceptional project-
linked responsibilities of the head of the women's
group are excluded). Women without cowives allo-
cated substantially less of their working time to
household tasks than their colleagues with cowives.
For personal needs, they had less than 13/4 hours,
compared with 3 hours for women with cowives.

"Suzanne Lallemand, Projet d'Accs des Femmes i I'Education (Paris:
UNESCO, November 1971), p. 1.

Age, Sex, and Time Use

Overall inspection of the time budgets supports the
hypothesis that age does not strongly affect women's
time use. Young and old alike shoulder substantial
burdens. Females begin putting in more work hours
than males beginning at age seven:

Hours of Work
Girls Boys


Technologies: Food Processing
and Improved Nutrition

Availability of technology did not correlate with the
anticipated reduction in workloads.10 Of women's
food-processing activities, grinding and pounding
absorb the greatest share of the time (84 percent of
total food-processing time, an average of over 13/4
hours per day). Thus the project's choice of introduc-
ing mechanical mills is understandable. Question-
naire response revealed that Voltaic women tend to
utilize a mechanical mill mainly when tired or
pressed for time. If the women intend to prepare
meals in the evening despite fatigue, their workloads
are lightened by mill use. The villagers stated,
however, that the mill is often used to permit the
preparation of meals that might otherwise be for-
gone. In this case, women's tasks are in fact in-
creased, as cooking time will still take 1-2 hours.
Eventual impact of the mill should thus be sought,
not in time saved but rather in improved nutrition or
increased productivity of the labor force.

ACKNOWLEDGMENTS My research in the Republic
of Upper Volta was made possible through backing
from the United Nations Development Programme
and the full support of the Voltaic authorities, notably

"This corroborates earlier findings of Alexander Szalai described in
The Situation of Women in the Light of Contemporary Time-Budget Research (E/
CONF.66/BP/6), 15 April 1975, prepared for the United Nations, World
Conference of the International Women's Year, Mexico City, 19 June-2
July 1975, pp. 8, 10.


Professor Ali Lankoande, Scholastique Kompaore,
Marcel Poussi, and many other colleagues and
friends. I am grateful to Marion H. Freedman, Brigid
O'Farrell, Hanna Papanek, Rosemarie Rogers, Philip
J. Stone, Arpad von Lazar, and Robert L. West for

their valuable suggestions. I thank Judith Bruce of
the Population Council and Achola Pala Okeyo for
inspiring the initial seminar presentation upon
which the article is based, and Sondra Zeidenstein
and Carol Weiland for their editorial skills.

Rural Women's Time Use

Vivian Havens Gillespie

The time-budget method of obtaining information
on rural Nicaraguan women was chosen when a liter-
ature search revealed a virtual absence of informa-
tion on women in general and on rural women in
particular in this Central American country.
It was felt that the time-budget method would
provide development planners with the most far-
reaching and complete baseline data to assess needs
and ascertain the impact of efforts directed toward a
largely unrecognized and underutilized resource in
the development process-women. Time-budget
studies offer many advantages over a standard ques-
tionnaire methodology. Perhaps the most obvious is
that the time-use method details what a person does,
when it is done, and how much time it takes to ac-
complish. This type of information is virtually im-
possible to obtain solely through recall. Time
budgets are taken by accompanying a subject for a
period of time-in this instance a full day-and writ-
ing down what he or she does and recording the time

Types of Women Observed

We observed the basic work patterns of three types
of women: the housewife, who works in her home
and receives no financial remuneration; the potter,
who fabricates clay pots in her home and sells them
for a profit; and the factory worker, who is absent
from her home at least eight hours during the day.
Their work days reflect various accommodations to
child-care and food-preparation responsibilities.
The work day of housewives is incredibly frac-
tionated. These women constantly shift from one ac-
tivity to another, taking between 5 and 15 minutes to
complete each task.

By contrast, potters fabricating various types
and sizes of common clay cooking vessels do not
have such a fractionated day. They organize all activ-
ities around a block of time (averaging nearly 41/2
hours) set aside for potting. Because these women
work in their own homes, they are able to care for
their children themselves. As a group, they did much
less food preparation and food processing than
housewives, some of these tasks being performed co-
operatively by other female household members
(daughters, daughters-in-law, nieces).
The third group observed were women working
in a hemp-rug-making factory. Over half these
women wage earners were living with either parents
or parents-in-law, who performed the necessary
household chores and cared for the children. Perhaps
this living arrangement is due chiefly to the relative
youth of the factory workers, whose ages varied be-
tween 16 and 30 years with a median age of 25 (this is
about 10 years junior to the median of the rest of our
sample). Of those not living with relatives, some em-
ployed a servant to work in the home, while others
prepared the day's food before going to the factory,
left children with an aunt or sister while they
worked, and took the children home for lunch and
again following the afternoon shift.

Benefits of the Observation Method

The benefits of the observation method are several
and varied. In addition to sensitizing the observer to
her subject's lifestyle-of primary importance to any
development planner-this methodology also marks
the direction of further research and indicates which
questions merit asking or pondering, as well as likely


avenues of action. For example, the Nicaragua study
was not specifically looking for information on nutri-
tion; however, because we were present in the home
during food preparation and mealtimes, we ob-
served many commonly practiced customs: (1) the
husband and/or children were fed before the woman
had her meal; (2) if there was an egg or piece of meat
to be consumed, it was given to the husband or older
male children; (3) the diets of residents situated close
to the Pan American highway were more varied and
apparently more nutritious than those of residents of
more remote villages. This last observation led to the
speculation that the high incidence of malnutrition
in Nicaragua (estimated to be the primary cause of
death of rural women) may be due not only to nutri-
tional ignorance but also to the unavailability (be-
cause of poverty, isolation, or both) of more nu-
tritious foodstuffs.
The observation method also illuminates the ob-
stacles that the rural poor face in realizing a fair price
for their labors. For example, in the community of
Rio Abajo, tomato farmers were observed to sell
boxes containing as many as 80 tomatoes to a buyer
with a truck for 5 cordobas (7 cordobas = US$1) per
box. The supermarket in Managua buys a box of 80
tomatoes for 35 cordobas. With monopoly control of
the means of transporting the campesino's (rural in-
habitant's) produce to market, the middleman plays
a "take-it-or-leave-it" game and invariably wins.
An additional advantage of the observation
method is the recording of the actual sexual division
of labor. When questionnaire or interview meth-
odologies are used to gain information, the answer is
frequently the culturally held ideal, which some-
times differs from the reality. For example, the 16-
year-old daughter of a vegetable farmer said that she
had finished high school and now handled her fa-
ther's accounts; and, indeed, she was observed
doing just that. However, when a buyer for her fa-
ther's tomato crop arrived, the daughter joined the
fieldworkers and picked tomatoes for five hours
without a break for food or water. Later she admitted
to having been in the fields two weeks earlier staking
the plants. The observation method, employed over
the course of the entire agricultural cycle, would
yield sensitive data concerning the actual participa-
tion of women in agriculture. It would also illumi-

nate how the seasonal migration of rural dwellers to
work the cotton, sugar cane, and coffee harvests af-
fects women's role. If they join the migratory work
force, do they work in the fields and retain their
cooking, laundering, and child-care responsibilities?
If they stay home, what additional responsibilities
become theirs when spouse and/or older male chil-
dren are absent? Although the culturally held work
place for rural Nicaraguan women is in the home,
they may regularly deviate from the ideal to take on
agricultural fieldwork as well. Change agents would
want to be aware of the degree and nature of this ex-
tended participation.
Observed sexual divisions of labor should be
employed as a guide for development planners to tai-
lor projects, whenever possible, to the customary
way of doing things. A Peace Corps volunteer re-
counted her frustration with the women of her vil-
lage, who refused to join her in making furniture
(simple tables and benches) for their homes. The ob-
servation method easily yields information on cus-
tomary divisions of labor and shows that although
rural Nicaraguan women repair the mud walls and
floors of their homes, men build the house frames,
put on the roofs, and make the furniture. To be effec-
tive, development programs must be in tune with
the burden of roles and responsibilities of a local
Observation yields a wealth of information
about how rural women accomplish their tasks and
identifies the tools and technologies used. This in-
depth knowledge of an area's activities and of how
and with what materials they are completed and by
whom can mean the difference between acceptance
and success or rejection and failure of modernization
projects. Field observers should be encouraged to
capture this information and record it in detail for its
later review and input into policymaking decisions.

ACKNOWLEDGMENTS Research for this article was
made possible by a grant from the Agency for Inter-
national Development to the Federation of Organiza-
tions for Professional Women/International Center
for Research on Women.

Group III: Instruments

for Learning

Measuring Rural Women's

Economic Roles and Contributions

in Kenya

Audrey Chapman Smock

Around 88 percent of women in Kenya reside in rural
areas, where they make a major contribution to the
local economy. The existence of an ongoing rural
sample survey program, conducted by the Central
Bureau of Statistics, makes it more feasible to collect
data relevant to rural women's economic roles in
Kenya than in most other developing countries. The
Central Bureau of Statistics instituted the National
Integrated Sample Survey Programme in 1974, with a
permanent field staff in a carefully selected national
sample of rural and urban clusters. The key Inte-
grated Rural Survey (IRS) is an annual, ongoing so-
cioeconomic survey in which a variety of question-
naires are administered to the same households
within the national sample. The main objectives of
the IRS are to develop and refine techniques for col-
lecting reliable, representative, and relevant informa-
tion from rural areas and to ensure the rapid
processing and quick release of the data. During the
four annual rounds of the IRS, standard question-
naires have been alternated with special modules.
Data collected in the repeated questionnaires include
demographic characteristics of the household;
physical descriptions and valuations of assets;
patterns of household expenditure, consumption,
and income; and farm production inputs and
yields. Special modules have focused on such sub-
jects as literacy, nutrition, nonagricultural activi-

ties, and access to nonformal education programs.
Data from the first two rounds of the IRS, col-
lected in 1974-75 and 1975-76, confirm the major
contribution of adult females to farming, but they do
not provide an insight into the division of labor in
agricultural production or in other household tasks.
The IRS was originally conceived as a household sur-
vey. Therefore the approach was to delineate the
physical and labor inputs, the farm production
yields, and the incomes of the household unit, and
the data cannot be disaggregated to distinguish be-
tween the productivity, contributions, and resources
of various family members.
To complement and supplement existing survey
data, the Central Bureau of Statistics has decided to
undertake some special modules specifying individ-
ual rather than household data. The Division of La-
bour Module, which was administered in February
and March 1979, is one such survey.
The purpose of the Division of Labour Module is
to provide detailed data on the contributions of
household members to agricultural production and
other tasks, such as care of poultry and livestock,
food preparation and cooking, house cleaning, child
care, buying food, fetching water, and collecting fire-
wood. The questionnaire is divided into three sec-
tions: the first part elicits background data on the
respondent and the household to supplement data



Form I
Target Population: Female Household Heads or Married Females
in Agricultural Households over 20 years of age
Province_ Cluster _
1 2 3 4 5 6 7 8 District_ Date _
H.H. Code Respondents Name
Respondent's Name

1 Identity of respondent:
Female head [
Only wife of male head 2
Senior wife of male head 3
Junior wife of male head 4
Only wife of son of male head 5]
Senior wife of son of male head 6
Junior wife of son of male head 7
2 Age of respondent:
3 Respondent's highest level of formal education:
No formal schooling I
1-2 years primary 12 years s
3-4 years s(
3-4 years primary [3 3 y
m= 4+ years si
5-8 years primary 4 years
4 Respondent's marital status:
Never married Skip to
m _Question 8
Formerly married Q
Currently married 3
5 Number of years married:
Less than 1 year 1 8-15 years

1-3 years 16-25 years
4-7 years 3 More than


25 years

collected in other IRS modules; the second section
focuses on the participation of members in agricul-
tural production; and the third deals with involve-
ment in other household work. Following the usual
procedure at the Central Bureau of Statistics, the

questionnaire employs a preceded format to facilitate
data processing.
In order for the module to be suitable for incor-
poration into the National Integrated Sample Survey
Programme, its design had to meet certain specifica-

Card No.

Office Use


6 Age of respondent's husband

7 Respondent' husband's highest level of formal education

No formal schooling 1

1 2 ears primary

3-4 years primary 3

5-8 cars primary r4

1-2 vcars secondary\

3-4 s\ears secondary

4+ vears secondary

8 Number of generations in household:

One 1l Fh ree

Two 2 Four

9 Number of generations that regularly work on holding

One 1 rhree

Two 2 Four

10 Number resident in household

Females 15 and over

Females h-14 not at school

Females 6-14 at school

Males 15 and over

Males 6-14 not t school

Males 6-14 at school

11 Number whose main occupation is working on holding

Females 15 and over

Females 6-14 not at school

Female, 6-14 at school

Males 15 and over

Males 6-14 not at school

Males 6-14 at school


12 Who takes the malor share of the responsibility for preparing the land for

Adult females 1
Entire tamt Il 6
Adult males 2

Adult females and

Adult females and

Adult males and

Family and hired labour

_3 Hired labour

a ] reactor service

tions. The survey had to be replicable in all rural
areas covered by the national sample, irrespective of
ecological zone, household size, or level of economic
development. The format had to be simple and clear
to facilitate administration by enumerators and com-

prehension by rural respondents, who often have no
formal education. The organization of the questions
had to produce data that could be aggregated to the
national level as well as disaggregated to assess the
effects on the division of labor of such factors as edu-

Office Use

17' Iis


.t 01



at :C 0

C, C,





cation, size of landholding, ecological zone, occupa-
tion, and proximity to markets.
The module specifies the target population as fe-
male household heads or married women in agricul-
tural households over age 20. The length of the

questionnaire precluded administration to both males
and females within the time period allocated, and it
seemed preferable to select either males or females as
respondents to provide some consistency in report-
ing. IRS findings that a higher proportion of rural

at 0





-- w U
5 15 r. 8; t __ 3 I ; 8

S S S 5 o S


to t '_ Kt to to to ___t
S 5 S S S_ S

- --- --- --- --- --- -- -- u
a s s [ j a m

s 9 U0L
Y 3
__ __ j __ _ s __
6'=PPP e ao

__ K _ S -
a^ j __ __ _ ^ &_ x -


women than men are involved in agricultural pro-
duction suggested that females would be able to
supply more accurate descriptions of the type of work
undertaken by different family members. Enu-
merators' experience has shown also that it is

easier to locate and interview females than males.
The module elicits data on the contributions of
six groups within the household: females aged 15
and over, males 15 and over, females 6-14 not at
school, females 6-14 at school, males 6-14 not at

n r
I c
S g.

-c ~K

S 0 K



o n
I ^


school, and males 6-14 at school. Data specification
by groups rather than on an individual basis sim-
plifies both collection and analysis. Although such a
group approach eliminates the possibilities of under-
taking some types of analysis, it still permits consid-
eration of how individual, family, ecological, and
community factors affect the division of labor within
the household.
The survey seeks activity-specific responses to
questions about particular types of work on the ma-
jor agricultural crops and about various household
tasks. Previous survey experience at the Central Bu-
reau of Statistics has shown that rural respondents
are unable to answer questions dealing with time al-
located to various tasks, even when the recall period
is limited. The basic problem is not so much rec-
ollection as it is conceptualization according to time
units. Therefore time-budget methodologies, which
study the division of labor by measuring time invest-
ments, appear unsuitable for large-scale observation
of respondents for the purpose of recording time al-
located to specific tasks. Such a technique is far more
suited to intensive anthropological investigations of
small communities than to national surveys.

The activity-specific approach is simpler to con-
ceptualize and makes it easier to record data and to
compare patterns across communities and in a vari-
ety of ecological zones. In the questionnaire the agri-
cultural cycle is divided into four activities-plant-
ing, weeding, harvesting, and marketing-for each
of ten major crops. Respondents are requested to de-
tail which groups within the household (1) do not
work, (2) work regularly, or (3) work sometimes, at
each of the four stages of the agricultural cycle for
each of the ten crops. Enumerators omit all crops not
grown by the household. Respondents similarly are
asked which groups within the household (1) do not
work, (2) work regularly, or (3) work sometimes, at
a variety of such household tasks as care of poultry
and livestock, food preparation and cooking, and
fetching firewood and water.
The Central Bureau of Statistics hopes to have
data from the Division of Labour Module available for
preliminary analysis by late 1979. The intention is to
analyze the results and to supplement the analysis
with a more systematic investigation that incorpo-
rates data on the same households and communities
from other IRS modules.

Learning about Women

through Household Surveys:

An Experimental Module

Nadia H. Youssef and Coralie Turbitt

We present an experimental module recommended
to the Indonesian Central Bureau of Statistics as a
way of increasing policy-relevant information about
women's economic role in the household and labor
force. Its approach is likely to be relevant in any
country where such information is scanty. As yet
there is no indication that the module has been
adopted in Indonesia.
The module is intended to be administered
every fifth quarter for five years, thus including data
about women's activities in different seasons. It is
intended to be administered to all females in the
household aged 15 and over.' Since there is a lack of
methodology to ensure that enumerators and women
respondents acknowledge women's actual economic
'It is assumed that a normal household chart will be completed for the
entire household, listing everyone who lives there, their ages, and their re-
lationship to the household head.

role, the module is meant to be supplemented by a
series of micro-studies to determine whether the
macro-study is sensitive to the situation of women-
that is, whether the right questions are being asked
to obtain data in sufficient detail to accurately reflect
women's lives and concerns. The results of the mi-
cro-studies are expected to produce refinements in
the module ready for inclusion the second time it is
scheduled to be applied. The interaction between
macro- and micro-level data should be a continual
process to capture the dynamics of change. Such
efforts are critical if one is to achieve an interpreta-
tive understanding of the condition of women and to
influence policy planning.
The module asks questions of women under the
headings "Economics of the Household" and "Indi-
vidual Economics," to correct the bias that emerges
when the unit of analysis is the household rather


than household members. When the household is
considered the unit of analysis and questions are
asked of the male as head, the economic activities of
women are submerged. Consciously or uncon-
sciously, questions about economic activity and pro-
ductivity are geared to the male, and much misinfor-
mation about women's dependent condition is per-
petuated. The module also seeks to include informa-
tion about the economic roles of all adult females in
the household, not only the wife of the household

The Experimental Module

1. Age-
2. What ethnic group do you belong to?
3. What language is spoken in your home?
4. Can you speak Bahasa Indonesian [the lingua
franca] ?
5. Does your household own any of the following?
a. agricultural land (more than .5 hectare / less
than .5 hectare);
b. other real estate;
c. family home;
d. physical place of business (e.g., shop / boat /
other work place);
e. none of the above.
6. Have you ever had any children?
a. number of children born;
b. number of children living;
c. number of children who died;
d. age at first pregnancy;
e. age at last pregnancy.

7. Can you read and write? In what character?
8. Are you presently attending school or a training
a. If in school, what type of school (e.g., voca-
tional secondary-government; general sec-
ondary-private; primary-religious)?
b. If in a training program, what type (e.g.,
on-the-job; nonformal literacy; nonformal
home science)?
c. If not currently in school or training, have you
ever attended school or a training program?
Highest level of formal schooling; type of
school? Type of training; number of years of

9. Are there any training programs (other than for-
mal schools) that you know of in this area?
a. If yes, what is being taught?
b. Have you attended?
c. If not, why not (e.g., no time / no money / no
interest / don't feel welcome)?

Migration History
10. Are you now living in the province of your birth?
If not, where were you born (location and
11. How long have you been living in your present
town or village?
12. If you have moved from your birthplace, what
was the direction of your last move?
a. rural to rural;
b. urban to urban;
c. rural to urban;
d. urban to rural.
13. On your last move, did you move-
a. alone;
b. as part of a family move, following: spouse /
parents / adult son / adult daughter / other
family members?
14. How many times in your life have you moved
from one province to another?

Marital History
15. Current marital status: single / married / sepa-
rated / widowed / divorced / abandoned /
consensual union / polygamous marriage. (If
never married, skip to question 20.)
16. Number of years in current marital status?
17. If married, is spouse currently present in the
a. If not, how long has he been absent? For what
b. If spouse is present, within the last year has
he been absent from the household for a pe-
riod of 6 months or more? If yes, for what
18. Age at first marriage? Age at last marriage?
19. Number of previous marriages?
a. Approximate number of years spent in each
previous marriage: first / second / third /
b. number of children from each marriage: first /
second / third / fourth.

Family Headship
20. Who in this household takes responsibility for
the family in terms of-


a. providing financial support;
b. making decisions regarding important family
21. How do you define the term "head of house-
22. Do you consider yourself the "head of house-
hold"? If yes, why? If no, why not?
23. Have you ever considered yourself as head of this
household? If yes, why?
24. If previously married, have you ever considered
yourself to be the head of a household at any
previous time in your life? If yes, specify for
what period of time, for what reason.
25. Have you ever been the main provider for your
family? If yes, when was this? Why?

Household Economics
26. In the past year, have you contributed in any
way to bring money or other material neces-
sities (food, clothing) to this household? If yes,
in what form (cash / other / both)?
27. Are there other members in this household who
contribute to the support of this family? If yes,
who (family member / type of contribution-
cash, other, both)?
28. Are there any persons living away from this
household who contribute to the support of
this family?
a. none;
b. relatives (cash / other / both);
c. nonrelatives (cash / other / both).
29. Does this household receive any income from
sources not mentioned above?
30. Who is the person responsible for managing the
household finances?
31. Do you consider the family income to be: ade-
quate / not quite enough / far too little for your
household needs?
32. If additional income is needed, is there addi-
tional work available in this area? If work is
available, who in your household would be
qualified to get such work (self / other female /
other male)? If work is available, who in your
household has the time for extra work, quali-
fied or not (self / other female / other male)?

Individual Economics
33. Are you currently (this season) engaged in some
kind of economic activity? (If yes, skip to ques-
tion 34.)
a. If not economically active, are you looking for

work? If not looking for work, why not
(e.g., gave up, ill, no need, not proper sea-
son, in school, retired)?
34. Complete the following chart for all women who
are engaged in any economic activity. Com-
plete one chart for their current primary activ-
ity and for as many as two secondary activities
[maximum of three charts]. If they do other
kinds of work in addition, simply list the re-
maining activities.
A. Primary Activity [ ] Secondary Activ-
ity[ ]
B. Specify type of work. For agricultural work-
ers, is it on own farm or another's farm?
C. Method of Payment:
a. cash: amount (Rp) per (time / piece);
b. in-kind: amount per (time / piece);
c. none.
D. Is the work you are now doing permanent /
irregular/ seasonal? (How many seasons is
it available to you in a year?)
E. How many hours per day do you engage in
the above activity?
F. How many days per week are you engaged in
this work?
G. Status of worker: employer / employee /
own-account worker / unpaid family
H. Place of work: home/ farm / marketplace /
plant / no defined place.
I. Do you need capital for your work?
a. If so, where do you get it?
b. Do you require daily capital?
J. Do you need tools or machinery to do your
a. Are they your own tools or machines?
35. Do you engage in other economic activities dur-
ing the year that are different from the ones
you have just described?
a. If yes, list them by season and by primary ac-
tivity / secondary activity.
36. What would you say is your main economic ac-
tivity for the year? (If no one activity can be
named, enter "not defined.")
37. Are your current year's activities your usual
sources of income (say, over the past 5 years)?
38. What is your busiest time of the year?
39. What do you think are the best (highest paying)
economic opportunities for women like your-
40. What are the worst opportunities?


41. If you could choose, what kind of work would
you prefer?
42. If you needed to borrow money in an emergency,
where would you get it (e.g., relative / em-
ployer / friend / bank / informal local lender)?
43. Do you own, in your own name, any real estate?
If so, what?

Individual Health
44. How would you describe yourself: healthy / not
very healthy?
45. Is there a doctor or medical facility near you?
a. If so, have you ever used it for your own
health problems?
b. If you were ill, would you use it?
c. If not, why?
46. If you are, or later hope to be, the mother of chil-

dren, would you take your children to such a
medical facility if they were ill?
47. Who usually attends to you when you need med-
ical care?

ACKNOWLEDGMENTS This work was carried out
under a grant to the Federation of Organizations for
Professional Women. The final report, entitled "A
preliminary study in three countries: Kenya, Nic-
aragua, Indonesia," was funded by the Women in
Development Office of AID. The authors were influ-
enced by Eva Mueller's paper "The women's issue in
measuring household behavior," presented at a con-
ference on Women in Poverty: What Do We Know,
sponsored by the International Center for Research
on Women, 30 April-2 May 1978.

Assessment of Body Concepts

and Beliefs Regarding

Reproductive Physiology

Michele Goldzieher Shedlin

The instruments presented here were developed dur-
ing 1974-76 as part of an anthropological study of
factors relating to the cultural acceptability of mod-
ern contraceptive methods and services in a tradi-
tional community in Central Mexico.' They have
since been adapted and used as tools in research,
teaching, and evaluation by a number of social scien-
tists working in Asia, Latin America, and the United
Interest in the study of body concepts was fos-
tered by Steven Polgar for many years before his
death in 1978. In addition to the research presented
here, work has been carried out by Susan Scrimshaw
and Michele Shedlin in an urban slum in Cali,
Colombia, and by Scrimshaw with pregnant His-
panic women in New York City. Research by Mary
Elmendorf is currently under way among Mayan
women in the Yucatan. Susan Philliber is using both
male and female drawings as part of an evaluation of
an adolescent sex education program in Virginia. She
also plans to use these instruments as part of an eval-
'The instruments as illustrated here are approximately 60 percent of
original size.

uation of the Young Adult Clinic at Columbia Pres-
byterian Medical Center.
The rationale for developing these instruments
was the belief that the way in which individuals ex-
perience and perceive their bodies is culturally pat-
terned. Perceptions of the structure and functioning
of the body are a reflection of culturally determined
cognitive categories. Inherent in many health care
programs, especially those designed by developed
countries for countries of the developing world, is
the erroneous assumption that the body, its proc-
esses, and modifications to it (i.e., disease, side
effects of contraceptive methods, and the like) are
universally experienced in the same way." These as-
sumptions by planners and providers of health care
have contributed to the lack of cultural acceptability,
underutilization, and lessened impact of many of
these programs. It is becoming increasingly clear that
we must understand what receiver populations per-
ceive as their health needs-perceptions that deal

'P. Manning and H. Fabrega, "The experience of self and body:
Health and illness in the Chiapas Highlands," in Phe omenological Socwl-
ogy: Issues and Applications, ed. George Psathas (New York: Wiley, 1973).


with a personal sense of well-being, as well as with
clinical diagnoses of health and illness.
In many developing countries where education
and information have not yet reached large sectors of
the population, interpretations of the body and body
processes are based upon folk beliefs. These beliefs,
which are both cause and effect of the individual and
cultural context of body concepts, often contain dis-
torted and fallacious ideas about the structure and
function of the body. Because individuals may have
little correct information, they have varying amounts
of difficulty in developing a clear map of their
bodies, especially the internal body.3
In studying the body concepts held by a particu-
lar population, we can obtain a perspective on the
culturally structured experience of the population,
especially with regard to health, illness, and re-
production. In a general sense, questions and an-
swers related to the body provide keys to under-
standing the importance and role of the body and
body parts, the interpretation of body processes, and
cognitions associated with the body. This informa-
tion can also serve to explain such related behaviors
as motivations and constraints in seeking, utilizing,
or rejecting health education, preventive medicine,
and other health services, including family planning.
In addition, the investigation of body knowl-
edge and concepts can be important in such areas as
the assessment of correct information about the body
and body processes, the identification of erroneous
beliefs and gaps in knowledge, and the identification
and understanding of attitudes about the body and
body parts. This information can be used specifically
to increase communication and assist in the educa-
tional process by providing an awareness of body
vocabulary and the cultural appropriateness of body
terms to health personnel involved in promotion and
These areas of information can be used by plan-
ners and providers of health care services in the de-
sign and delivery of more culturally appropriate
promotion, education, and services to particular
groups. The instruments presented here can also be
used in evaluating the impact of promotion and edu-
In the anthropological investigation in Central
Mexico for which they were developed, these instru-
ments assisted in identifying culture-specific percep-
tions and attitudes concerning the female reproduc-
tive cycle and in eliciting information on menstrua-
tion, conception, pregnancy, breastfeeding, and
fertility regulation. They were also used in determin-

:'See S. Fisher, Body Consciousness (New York: Jacob Aronson, 1974).

ing how body knowledge and beliefs are related to an
understanding and acceptance or rejection of modem
contraceptive methods.
The women interviewed in this investigation
were largely illiterate and were members of a tradi-
tional Indian and Mestizo community. They were
between the ages of 15 and 44 and in sexual union at
the time of interview.
The body outlines, filled in or finished by the
women, and the concepts of reproductive physiology
elicited during the drawing process, show varying
degrees of knowledge and accuracy. However, when
viewing them together, various patterns and themes
emerge. With regard to knowledge of specific inter-
nal organs, the most frequently drawn were the
stomach, intestines, heart, liver, uterus, and lungs.



Less frequently drawn were the kidneys, ovaries, the
"neck of the uterus," and the placenta. Ribs were oc-
casionally included, the only bones that were spon-
taneously indicated.
The external structures most frequently drawn
were the breasts, vagina, and umbilicus; the area of
the waist was sometimes emphasized. In general, all
internal and external structures were represented as
circles of varying sizes, usually quite small and rarely
connected to one another in any way. When a spatial
connection was perceived, such as between the
stomach and the intestines, the ovaries and the
uterus, or the heart and the liver, the organs were
usually represented as concentric or contiguous cir-
cles. Drawings of organs that occasionally varied
from the circular form were the intestines, drawn as

varying types of lines, and the heart, which was rep-
resented as a valentine.
The accurate placement of internal organs was
uncommon. The heart was located directly behind
one of the breasts, in the center of the body, or di-
rectly between the breasts. The liver, frequently
placed adjoining or even attached to the heart, had
the largest range of variation in placement. It was
drawn high in the chest near the heart, in the center
of the chest, at waist level on either side, in the cen-
ter of the body, and even at different points in the
abdomen. The kidneys, one or both, were placed in
the abdomen, with little idea as to correct placement.
The lungs were usually placed at shoulder level. In-
testines were drawn above, below, inside, and out-
side the stomach. The stomach, like the liver, also
occupied many locations but with less of a range
than the liver, being generally in the abdominal area
and placed centrally. The uterus was the organ most
frequently located correctly, and when the ovaries
were identified, they were more or less in the right
area also, although slightly high or low. The placenta
was drawn inside, outside, and near to the stomach
or uterus. The only reluctance in answering or draw-
ing was encountered when the women were asked to
draw or give names for the vagina. They denied the
existence of any names and often covered their
mouths with their rebozo (shawl), an action that re-
flects embarrassment and discomfort. Most, how-
ever, did answer eventually when encouraged and
The only other external parts the women were
encouraged to draw were the breasts, essentially to
learn the terms used. Breasts were usually repre-
sented as small circles, and nipples were rarely indi-
One of the main objectives of the drawing in-
strument was to elicit the names used for body parts,
not only for purposes of communication, but also to
obtain an idea of the relative importance and aware-
ness of the various parts as illustrated by the number
of terms that existed. The parts with multiple names
were the breasts, stomach, uterus, and vagina. The
use of five different terms for the breasts clearly re-
flects an importance and awareness among women,
who lactate and breastfeed for a large proportion of
their reproductive lives.
The names for and locations and explanations of
the structure and function of the stomach and uterus
were confused by many of the women. When the
baby was believed to grow in the stomach and not
the uterus, the stomach was located differently and
the uterus seldom drawn. Thus, the part identified as
"where the baby grows" was called not only matriz

Numero de Embarazos:


(uterus) but also estomago (stomach), barriga (belly or
pregnancy), and vientre (abdomen, belly, womb).4
The type of terms given to the vagina reflects a
lack of usage of names for the area and the embar-
rassment associated with referring to it. It is called
generally la parte (the part). Other terms occasionally
used instead were cuerpo (body), colita (little tail),
and nalga (buttock).
Explanations of why certain organs are identi-
fied and hold more importance for these women are
found in, among other things, women's disease ex-
perience and their beliefs about disease. The beating
heart is the sign of life and is believed to "move the

'Note the ambiguity between barriga and vientre and that both defi-
nitions include alternative meanings of stomach and uterus; thus the ac-
tual definition depends on usage.

blood" and to make blood. It is also believed to be
the origin of feeling and emotion and has religious
and emotional as well as physiological meaning. Cir-
rhosis of the liver is common in both men and
women due to the constant and heavy consumption
of alcohol, and problems of the liver are associated
with coronary failure. Tuberculosis is common in the
area and is known to affect the lungs; along with
other respiratory conditions it is a major cause of
morbidity and mortality, especially in the winter
months. Blood in the sputum is known to come from
the lungs. Also prevalent in this area are numerous
parasitic and gastrointestinal conditions affecting the
stomach. When discussing susceptibility to disease,
the stomach was frequently cited as the most suscep-
tible area. This belief is also related to awareness of
the intestines.
The uterus, whether or not confused with the
stomach, abdomen, or other organs, was also drawn
frequently since numerous health problems occur
among these high-parity women. The associations of
the uterus with menstruation, conception, and preg-
nancy are ever present in the women's experience
and conversation. A "fallen womb" (matriz caida) is a
common complaint.
The placenta is seen by the women during the
birth process. The waist, kidneys, and abdomen
(vientre) are cited as areas where one feels menstrual
discomfort, as well as aches and pains caused by in-
fections, changes in climate, foods, and so on.
The infrequent inclusion of internal reproduc-
tive organs other than the uterus directly reflects the
women's beliefs about reproductive physiology, es-
pecially their beliefs about conception. The majority
of the women explained that sexual contact was nec-
essary to become pregnant, but many felt that be-
tween two and ten separate acts of sexual intercourse
(el me usa-"he uses me") were necessary for a child
to begin to form, since conception occurred when the
liquid, or blood, of the woman united with the blood
of the man in her matriz, estomago, vientre, or bar-
riga and since numerous contacts were required to
accumulate enough "blood." Some felt that the
woman's liquid was always present in the uterus;
others said it was formed there during intercourse.
One woman said that the woman's ovaries produced
the liquid and that a male ovary produced the man's
liquid. This liquid, when believed to be blood, was
said to change its color to red once it had entered the
woman's uterus or had mixed with the blood of the
veins. A few women said that only the male pro-
duced a liquid and that once inside the uterus it be-
gan to grow into a child.


The instruments included here reflect the gen-
eral body shape and size of the Indian and Mestizo
women. Variations such as a torso-only (shown at
right) or a slimmer version with short hair were pre-
tested and rejected. It was important for the respon-
dents to be able to identify with the basic body
outline; however, it is possible that some groups may
respond equally well or better to an "ideal" body
type. This must be decided in each case in the field by
protesting the instrument.
One of the most useful aspects of these instru-
ments is their flexibility and adaptability. They are
easily modified to the respondent population, and
questions can be developed to meet the needs and
objectives of specific research, education, or evalua-
tion. In addition, the drawing process assists greatly
in stimulating discussion and adding depth to open-
ended questions. The instruments were not meant to
be ends in themselves, but rather tools to assist in
the interview process. The completed drawings can
be studied and evaluated individually along with the
interview and then in conjunction with the other
drawings in order to provide scope and depth to
overall conclusions about the population. When the
drawings are viewed as a group, they tend to bring
out the erroneous folk beliefs of significance, as well
as to illustrate correct notions of body structure and
functioning. They can be evaluated according to:

1. number of parts correctly mentioned;
2. number of parts correctly placed;
3. correct representation of size and relationship
to other parts;
4. correct terminology;
5. number of terms given for any one body part;
6. consistent errors in size, location, or function
of a body part; and
7. consistent omissions.

Comparisons of body concepts such as those of
respondents with and without access to information
or services, of rural and urban women, of different
age groups in the same population, of different
cultural groups in the same area, or even cross-
cultural comparisons can be useful in identifying
perceived health problems, in evaluating the impact
of services, and in guiding research related to the
cultural acceptability of health care.

ACKNOWLEDGMENTS The study described in this ar-
ticle was supported by the Task Force on Acceptabil-
ity Research in Family Planning of the Special
Programme of Research, Development, and Research
Training in Human Reproduction of the World
Health Organization and by the Agency for Interna-
tional Development. The author wishes to thank
Jesus Corral Gallardo, Estela Ortiz Romo, John F.
Marshall, Samuel Wishik, Allan Rosenfield, Jerald
Bailey, and Guadalupe Castro for their advice and
support of this research.


Project-Oriented Research on Aspects

of Women's Knowledge and Experience

Taherunnessa Abdullah and Sondra Zeidenstein

The following sets of questions were developed to
influence the content of a centrally administered
rural development project intended to direct re-
sources to women organized in village-based insitu-
tions in Bangladesh. When the project was initiated,
it was obvious that prevalent urban views of rural
women were far removed from the reality of their
lives. A project based on such views would have lit-
tle relevance to rural women or rural development
and would rapidly deteriorate. In an effort to reveal
certain features of rural women's lives as a basis for
defining relevant project content, sets of questions
were developed and administered, in interviews of
one to two hours, to women individually or in
groups in different parts of the country. Hypotheses
suggested by women's answers were confirmed or
modified by project staff posted throughout the
country and routinely in contact with rural women.
There was no attempt to conduct a rigorous scientific
study of rural women. The purpose of these ques-
tions was to find out from rural women of various so-
cioeconomic backgrounds their experience and
knowledge regarding issues the project could ad-
One critical area where information was needed
to ensure the relevance of training, supplies, and fi-
nancial support was women's involvement in agri-
culture. Questions were developed to discern
women's level of knowledge and basic practice in re-
gard to postharvest rice production, which is their
responsibility within the sexual division of labor.
Another short set of questions was developed to as-
certain their knowledge about livestock care. Neither
of these areas was initially considered relevant to a
project for rural women. Because most rural women
of Bangladesh are limited by purdah from involve-
ment in field aspects of agricultural production, there
was little awareness of what they knew about deci-
sions relating to field production. Questions were
also developed to ascertain their level of knowledge
on these matters.
The national project for rural women was struc-
turally parallel to a project for rural men, which in-
cluded provision of credit for increasing agricultural
production. For women's projects such economic
content was new; it raised questions and disapproval
on the grounds that control of money was a male role

and that extending credit to women for their eco-
nomic activities would cause conflict and undermine
the family. Two sets of questions were developed to
explore the degree of women's involvement in in-
come generation prior to project input.
Another goal of the project was to find ways to
link women in villages with family planning and
health services. To do so required an initial sense of
women's knowledge and practices in these matters.
In regard to family planning especially, it was felt
that women needed reliable information about con-
traception, since they were isolated in their villages.
To counteract the normative view that women did
not discuss their reproductive functioning and there-
fore might be "corrupted" by such an approach, it
was important to ascertain the actual situation.
Another issue that was completely obscured by
normative belief was divorce. Awareness of the
prevalence and significance of divorce in rural areas
was important in understanding rural women's be-
havior. Questions on divorce and on the village code
of good behavior for women were intended to pro-
vide a more realistic picture of the social pressures
influencing village women's behavior. Such infor-
mation was considered important in interpreting
women's responses to new resources.

Research Questions

Rice Processing
1. Who parboils and dries rice in your bari?
2. Why do you parboil rice?
3. How long do you parboil it?
4. How much heat do you use?
5. How do you know when it is finished?
6. How can you tell when the rice is dried?
7. Do you get a lot of broken rice?
8. What causes "Cheeta dhan" [spoiled rice]?
9. How do you store the rice seed?
10. Do rats and/or insects spoil the seed? Why? Why
11. Why do some seeds not germinate?
12. How much rice do you get from one maund of
13. How much koi?


14. How much morri? [Koi and morri are special rice
15. Do you get better rice from a dheki or a mill?
16. Do you use the mill? Why? Why not?
17. How much does the mill charge?
18. How much time does it take for one woman to
process 10 maunds of paddy?

Livestock Care
1. How can you tell if a cow, bullock, or goat is of
good breed?
2. What do you feed your animals?
3. How do you care for them?
4. What diseases do they contract?
5. How do you treat them?

Knowledge of Field Production
1. How much paddy land do you own?
2. Where are your plots?
3. How many crops did you grow this year? Last
4. How much yield did you get this year? Last year?
5. When were the crops sown this year? When were
they harvested?
6. How much paddy did you sell this year? At what
price? How much last year?
7. How many laborers did you hire this year? How
many days did they work?
8. Do you have irrigation?
9. How much did you pay for seed?
10. How much did you pay for fertilizer?
11. How much did you pay for insecticide?
12. How much is your land worth?

Earning and Saving
1. Do you know women who are earning money?
Tell about one.
2. What do you think of her?
3. What is her attitude toward her husband, in-
laws, family?
4. Do other villagers approve of her behavior?
5. Do villagers object to her earning?
6. Some women save money. How do they do it?
Tell me about one.
7. Where does she keep her money?
8. What does she do with her savings?
9. Does her husband know she has savings?
10. If she wants to do business, how much interest
does she charge?
11. If she lends her savings, how much interest does
she charge?
12. If she stocks paddy, how does she manage it?
13. If she lends goats, what are the arrangements?

Women's Earnings
1. Who decides how to spend the money you earn?
2. Do you give your earnings to your husband?
3. Have you bought land with your earnings? In
whose name?
4. Did you buy ornaments for your sister's mar-
riage? Did you have to ask your husband's
5. How do you spend your earnings?

1. How old were you when you had your first pe-
riod? Was it before or after marriage? Was
there a special ceremony? Describe it.
2. Did you know about menstruation before your
first period? If yes, how? If no, who explained
to you about menstruation? What did she say?
3. Can you say why we have menstruation? Why
does it start? Why does it stop?
4. Do you know anyone who didn't menstruate?
Can you have a baby before menstruation
starts or after it stops?
5. What different names are used for menstruation
in your village?
6. After how many days do you get your period? Do
you feel pain before and/or during menstrua-
tion? What other feelings do you have? How
many days does the bleeding continue? Does it
happen this way every month? If not, do you
know why?
7. How do you keep yourself clean during menstru-
8. Is there any prohibition regarding food, move-
ment, daily activities, and religious activities
during your period?

1. Do you have children? How many?
2. What is the age of the youngest? Of the oldest?
3. Did any babies die? How many? Of what? Any
miscarriages? What were the causes?
4. When you first got pregnant, how did you know
you were pregnant?
5. What words do you use for pregnancy?
6. Are you told to eat any special foods during
7. Do you have any special ceremonies during
8. Do you get any special instructions about how to
behave during pregnancy? If yes, what and
from whom?


1. How old were you when your first child was
2. How old were you when you got married?
3. How many years after marriage was your first
child born?
4. Whose house did you give birth in-your
mother's or mother-in-law's?
5. Which room in the house was the child born in-
the main room, side room, or other?
6. How long were the labor pains with your first
child? With the next child?
7. Who was present during labor? During delivery?
8. Sometimes the baby's navel or the mother's
womb gets infected after childbirth. What was
done in your case to prevent infection?
9. Did you have an infection after childbirth? If yes,
how did you take care of it? Who prescribed
the medicine?
10. Did your baby have an infection after childbirth?
If yes, how did you take care of it? Who pre-
scribed the medicine?
11. How was the baby born? Were you lying down,
sitting up, or in some other position? Can you
show us?
12. To make the baby come faster, lots of things are
done. What was done in your case?
13. Immediately after childbirth, what were you
given to eat?
14. Did you take any medicine? If yes, what?
15. Think about what you ate for the first 40 days af-
ter childbirth. What did you eat a lot of? What
were you not allowed to eat?
16. Did you get special foods to dry up the birth ca-
nal? For producing milk?
17. To dry up the baby's navel quickly, what was
18. How long after childbirth did you have to stay in
the same room?
19. What prohibitions were there about moving af-
ter childbirth? Could you move about any time
of day or night?
20. Did you nurse all your babies?
21. When did you start nursing after childbirth?
How frequently do you nurse your baby? How
long did you nurse your first baby? Your last
22. How long were you bleeding after childbirth?
23. When did your period start again after the first
child? After the last child?
24. How many more children do you want?
25. Have you ever heard of family planning?
26. Have you ever used it? What method? Did you
have problems with it?

1. Village girls know about sex before marriage.
How old are they when they know? How do
they know? Don't girls talk about the comings
and goings from the husband's room? What do
they say?
2. When you got married, did you know about sex?
Who told you?
3. What words are used among village women to
discuss sex?
4. If there are sex problems, to whom do girls talk?
5. Whom does your husband talk to about sex?

1. Many women don't want more children after
they have had 7 or 8. Perhaps they can't afford
to feed and clothe more children, or they have
some other reason. Do you know anyone who
has had an abortion? If yes, why? How? What
method was used? Did the husband or family
know of the abortion?
2. Are there doctors of any kind who give medicine
for abortion?
3. What other methods do you know of for abor-
4. What other reasons might there be for abortion?
5. What do villagers and village leaders think about
6. Do you know cases of premarital sex, of unmar-
ried mothers, of relationships with men other
than the husband?

1. How common is divorce in your village?
2. Mention some cases that you know of.
3. What were the causes?
4. Who initiated the divorce?
5. What happened to the wife?
6. What happened to the children?

Village Code
1. Who is considered to be a beautiful girl in your
village? What does she look like?
2. Who is considered to be a good girl? How does
she behave?
3. Who is considered to be a bad girl? What has she
done? Do you think she's bad?
4. What happens to a bad girl in your village?
5. What do you consider to be a good husband?
6. What is purdah? Do you keep it?
7. What does the moulvi [village religious leader]
think is a good girl? Do you believe the
moulvi? Does your husband?

Research Priorities:

Women in Africa

Achola Pala Okeyo

An Expert Meeting on Research and Data Collection
on Women and Development was held in Nairobi,
Kenya, during 4-9 December 1978 as a follow-up to a
meeting of researchers held in Nairobi in August
1975.' The meeting was convened by the United Na-
tions Children's Fund (UNICEF) and codirected by
the author and Virginia Hazzard (Senior Programme
Officer, UNICEF, Eastern Africa region). It was at-
tended by experts in research on women and devel-
opment in Africa, including a representative of the
African Training and Research Centre for Women
(ATRCW) of the United Nations Economic Commis-
sion for Africa (UNECA).
Although the discussion took place in the con-
text of African development and the situation of
women in the Eastern Africa region in particular, it
was felt that the results of these deliberations would
have a much wider application to the situation of
women in other regions with agrarian-based econo-

Objectives and Scope

The main objectives of the meeting were to develop a
conceptual and methodological guide for a baseline
study of the position of women in development. This
was to be accomplished by reviewing research on
women and development in order to evaluate its va-
lidity and effectiveness in explaining the position of
women and contributing to their advancement; for-
mulating hypotheses and indicators of the position
of women; generating questions, based on formu-
lated hypotheses, that would guide future research
designed to validate the hypotheses; and presenting
a guide for data collection that would discuss re-
search methodology, not simply in terms of tech-
niques or instruments of measurement, but by

'The following discussion is derived from the Report of the Expert
Meeting on Research and Data Collection, 4-9 Decenbel', 1978, Nairobi, Kenya,
United Nations Children's Fund, Eastern Africa Regional Office, Nairobi,
Kenya, 1978. Copies of the complete document are available from Virginia
Hazzard, UNICEF, Eastern Africa Regional Office, P.O. Box 44145, Nai-
robi, Kenya.

identifying factors that should be included in an
analysis of the position of women.

Hypotheses for Research

During the International Women's Year (1975) and
subsequent years, a number of meetings have been
held at international, national, and regional levels at
which research needs were emphasized. In most of
these meetings, research needs were identified and
recommendations for future research presented. It
was the consensus of the Nairobi meeting, however,
that the identification of research priorities was im-
portant but should build upon what is known. This
consensus was based on a review of literature on
women and development, which concluded that a
number of assumptions have been made on the sit-
uation of women, but that adequate and comparative
data are insufficient to substantiate them. It was felt
therefore that it would be useful to reformulate hy-
potheses suggested by available data and to generate
questions that would guide data collection (using
primary and secondary sources) to validate or refute
them. The group formulated hypotheses based on
the literature review, personal research experiences,
and a consideration of the minimum knowledge
needed to assess the situation and needs of women.
Although there may be a bias in the hypotheses
chosen because of research interests represented at
the-meeting, it was felt that data collected on the ba-
sis of such testable propositions would further a sci-
entific understanding of the roles and status of
women by moving from the perception of their situa-
tion (description) to an explanation of observed
trends and processes (conception).
The hypotheses and questions listed here were
formulated by the two working groups during the

Modernization of Agriculture

Hypothesis. Agrarian reform leads to increased
work loads for women, restriction of female employ-
ment to the primary phases of the production cycle


(in the manual as opposed to the mechanical opera-
tions), and devaluation of female labor.
Suggested Questions. What crops are grown?
Which are for sale; for home consumption? Who con-
trols cash crops; the production process in cash
crops; the income from cash crops? Who makes deci-
sions about land use in farming; disposal of income?
Who provides labor; what type and amount (time-
budget studies)? What is the breakdown of labor by
age and sex in farming (by crop)? What variables
govern the use of hired labor?

Hypothesis. Mechanization tends to be used by
men, and leads to increased demand on female labor
(time spent) and displacement of women by men
from their traditional agricultural roles.
Suggested Questions. What is the available tech-
nology for use on the farm and in the household?
Who uses the technology; for what purposes; with
what result? What effect does technology have on
production; on social organizations? Who deter-
mines what technology is used? What technology
purchases are made? What is the impact of farm/
home technology on women?

Hypothesis. Land reform leads to loss of land-
holding and user rights by women and to their eco-
nomic dependency.
Suggested Questions. Do women's rights to land
differ from those of men? Has land reform altered
women's rights? What was the previous landholding
pattern? How has land reform affected women's cash
income? Has land reform increased women's labor

Hypothesis. Agrarian reform leads to changing
legal rights,2 including title rights (ownership and
disposal), user rights, and disposal rights (sale,
mortgage, inheritance).
Suggested Questions. How has land reform af-
fected women's traditional land rights; their owner-
ship and user rights? What legislative measures have
been introduced to protect disposal rights? What
provisions have been made for women's inheritance
rights? What are the rights of women to land under
customary law; under statutory law? In a given situa-
tion, which law is applicable? What are women's
land rights in customary law? Are women aware of
their legal rights? Are laws regarding women's rights
being observed in practice?

2It was proposed that these changes be examined within both cus-
tomary and statutory law.

Hypothesis. Agricultural extension services tend
to exclude women.
Suggested Questions. What is the breakdown of
agricultural extension users by sex? What is the per-
centage of women farmers participating in agricul-
tural extension? What is the size of their holdings?
What are the types (curriculum content) of agricul-
tural extension activities in which women engage?
What is the sex composition of agricultural extension
staff? What are the career histories of extension

Hypothesis. The structures of marketing and in-
dustry undervalue work done by women.
Suggested Questions. What commodities do
women market? What is the scale (size) of their mar-
keting? Do they employ other persons? What market-
ing facilities (i.e., infrastructure) are available to
women producers and distributors? How does avail-
ability or unavailability of infrastructure affect
women's market activities? What is their access to lo-
cal, regional, national, and international markets?
What is women's knowledge of market conditions?
Has any market research been done relative to
women's production? How do women participate in
industrial production; in what types of industries?
What types of tasks do they perform? How do they
perform them (techniques, structure)? Do men and
women have differential productivity? Why do
women predominate in lower job categories (socio-
economic, cultural factors)? What is their access to
support facilities-for example, day care centers or
family members who can manage the housework?
What is the availability of training opportunities?
Are there any legal bars to women's participation in
industrial production?


Hypothesis. The nutritional condition of women
and children deteriorates with the introduction of
cash crops.
Suggested Questions. Does cash cropping de-
crease land area devoted to food crops? How much
cash crop income is used for food purchases? Does
family diet change with cash cropping? What factors
influence food purchasing habits (e.g., multinational
corporations, advertising, social mobility, nutrition
education)? What factors influence sales of domes-
tically produced food? What disease patterns are
associated with poor nutrition? Are there any cor-
relations between cash cropping and nutritional defi-
ciencies, by sex, age, and life cycle?



Hypothesis. Male labor migration creates chang-
ing social structures and leads to sex role restructur-
ing, psychological stress on women, female-headed
households, and changes in household decision-mak-
ing patterns.
Hypothesis. Female labor migrants tend to find
employment in low-paying and demeaning job cate-
gories in urban centers.
Suggested Questions. What is the structure and
rate of male/female migration? What is the extent of
urban-to-rural and rural-to-urban migration? How
does this vary by age and sex? To what sectors of the
labor market do female/male migrants move? What
is the impact of labor migration on family structure
and decision-making patterns?

Population and Health

Hypothesis. Increased economic opportunities
lead to reduced fertility.
Suggested Questions. What is the extent of pop-
ulation policy in the region? What is the impact of
population activities (programs)? What is the rela-
tionship between fertility and increased opportuni-
ties; responsibilities? How do increased education
and job opportunities affect fertility? What is the re-
lationship between access to health services and fer-

Women's Programs and Support Structures

Hypothesis. Programs designed especially for
women perpetuate the sexual division of labor, re-
strict women to the beginning of the production
chain, and channel women into the stagnating sector
of national and international economies (e.g., hand-
Suggested Questions. What is the breakdown by
sex of work done in community self-help schemes?
To what extent are day care services available for
working mothers? What is the cost per child? Who
supports the organization of women's groups? To
what extent are they organized spontaneously? To
what extent is there a coordinated, mutually support-
ive women's movement?

Women's Development Priorities

Hypothesis. Development research and plan-
ning have overlooked the priorities of women and
their development goals (an evaluation of women's

present situation and their perceived direction for
change is a suggested methodology).
Suggested Questions (description of an actual situa-
tion). What is good/bad about the respondent's life
and why? What could improve the respondent's life?
What are the constraints (problems)? How do women
evaluate their life situation (criteria)? What are
women's views of a good life (individual, group, and
community)? What are the factors they consider es-
sential to having a good life? What are the constraints
on having a good life? What are the potentials within
the existing situation for achieving a good life? If
there must be a choice, what would be the priorities?


Hypothesis. The socialization process reinforces
sex role definitions, affects women's aspirations, de-
termines male/female spheres of power, reinforces
male/female decision-making patterns, and deter-
mines sex preference for children and family size.
Suggested Questions. What are the prevailing
patterns of child rearing (from infancy to maturity)?
Do women see a need for any changes in the social-
ization practices that reinforce sex role definitions
and aspirations? Do women see any possibilities for
alternative social systems in which sex role defi-
nitions are changed? To what extent do women ac-
cept the possibility of such alternative socialization
practices? How would respondents feel about a sit-
uation (society) in which traditional sex roles were
interchangeable (especially with men assuming more
roles in home and child care)? Who are women's role
models? Are there cross-sex models? What forces
(individual situations, etc.) affect or determine
women's aspirations? To what extent are women
aware of the effects of their socialization practices on
the aspirations of their own children?

Power and Decision Making

Hypothesis. Modernization has led to the loss of
traditional bases of authority and decision-making
power among African women.
Suggested Questions. What are the sources of
power (social, political, economic)? To what extent
do women have access to them? Are women aware of
these sources of power? How do women contribute
to the definition of group values? What are women's
rights and obligations in the reproductive processes
and functions? Are women aware of their power
potential and how this might be used in their favor?

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs