Gender issues in farming systems research and extension

Material Information

Gender issues in farming systems research and extension
Series Title:
Westview special studies in agricultural science and policy
Poats, Susan V
Schmink, Marianne
Spring, Anita
University of Florida -- Women in Agricultural Development Program
Place of Publication:
Westview Press
Publication Date:
Physical Description:
xxi, 450 p. : ill. ; 23 cm.


Subjects / Keywords:
Agricultural systems -- Research -- Congresses ( lcsh )
Women in agriculture -- Congresses ( lcsh )
Agricultural extension work -- Congresses ( lcsh )
Farms, Small -- Congresses ( lcsh )
Agricultural systems -- Research -- Congresses -- Developing countries ( lcsh )
Women in agriculture -- Congresses -- Developing countries ( lcsh )
Agricultural extension work -- Congresses -- Developing countries ( lcsh )
Farms, Small -- Congresses -- Developing countries ( lcsh )
Mujeres en la agricultura -- Congresos
Extensiâon agrâicola -- Congresos
Granjas pequeänas -- Congresos
Productividad agrâicola -- Congresos
Systáemes agricoles -- Recherche -- Congráes ( rvm )
Femmes en agriculture -- Congráes ( rvm )
Agriculture -- Vulgarisation -- Congráes ( rvm )
Petites exploitations agricoles -- Congráes ( rvm )
Agriculture -- Productivitâe -- Congráes ( rvm )
bibliography ( marcgt )
non-fiction ( marcgt )
conference publication ( marcgt )


Includes bibliographies.
General Note:
"Published in cooperation with the Women in Agricultural Development Program, University of Florida"--Page opposite t.p.
General Note:
Based on an international conference, held Feb. 26-Mar. 1, 1986 at the University of Florida.
Electronic resources created as part of a prototype UF Institutional Repository and Faculty Papers project by the University of Florida.
Statement of Responsibility:
edited by Susan V. Poats, Marianne Schmink, and Anita Spring.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact Digital Services ( with any additional information they can provide.
Resource Identifier:
17386097 ( OCLC )
87034315 ( LCCN )
0813373999 (pbk. : alk. paper) ( ISBN )


This item has the following downloads:

Full Text

Gender Issues
in Farming Systems
Research and Extension

Published in cooperation with the
Women in Agricultural Development Program,
University of Florida

Gender Issues
in Farming Systems
Research and Extension

Susan V. Poats, Marianne Schmink,
and Anita Spring

Westview Press


Westview Special Studies in Agriculture Science and Policy

Figures 28.1 and 28.2 are reprinted by permission of Lynne Rienner Publishers,
Inc., from Peter E. Hildebrand and Federico Poey, On-Farm Agronomic Trials in
Farming Research and Extension. Copyright C 1985 Lynne Rienner Publishers, Inc.

This Westview softcover edition is printed on acid-free paper and bound in softcovers
that carry the highest rating of the National Association of State Textbook
Administrators, in consultation with the Association of American Publishers and the
Book Manufacturers' Institute.

All rights reserved. No part of this publication may be reproduced or transmitted in
any form or by any means, electronic or mechanical, including photocopy, recording, or
any information storage and retrieval system, without permission in writing from the

Copyright 1988 by Westview Press, Inc.

Published in 1988 in the United States of America by Westview Press, Inc; Frederick A.
Praeger, Publisher; 5500 Central Avenue, Boulder, Colorado, 80301

Library of Congress Cataloging-in-Publication Data
Gender issues in farming systems research and extension.
(Westview special studies in agriculture science
and policy)
"Based on the 1986 University of Florida
"Published in cooperation with the Women in
Agricultural Development Program, University of
1. Agricultural systems--Research--Congresses.
2. Women in agricultural--Congresses. 3. Agricultural
extension work--Congresses. 4. Farms, Small--Congresses.
5. Agricultural productivity--Congresses. I. Poats,
Susan V. II. Schmink, Marianne. III. Spring, Anita.
IV. University of Florida. Women in agricultural
Development. V. Series.
S494.5.S95G46 1988 630'.88042 87-34315
ISBN 0-8133-7399-9

Printed and bound in the United States of America

The paper used in this publication meets the requirements
of the American National Standard for Permanence of Paper
for Printed Library Materials Z39.48-1984

6 5 4 3 2 1


List of Tables and Figures........................... ix
Marianne Schmink and Susan V. Poats............. xv


7/1 Linking FSR/E and Gender: An Introduction
Susan V. Poats, Marianne Schmink, and
Anita Spring.................................... 1

2 Integration of Intra-Household Dynamics
into Farming Systems Research and Extension:
A Survey of Existing Projects
Rosalie Huisinga Norem ........................... 19

3 Gender Relations and Technological Change:
The Need for an Integrative Framework of Analysis
Alison Evans .................................... 37

4 Problems of Understanding and Communication
at the Interface of Knowledge Systems
Janice Jiggins................................... 49

5 From Recommendation Domains to Intra-Household
Dynamics and Back: Attempts at Bridging the
Gender Gap
Amalia M. Alberti ............................... 61

6 Research, Recommendation and Diffusion Domains:
A Farming Systems Approach to Targeting
Peter Wotowiec, Jr., Susan V. Poats,
and Peter E. Hildebrand ......................... 73

7 Incorporating Women into Monitoring and Evaluation
in Farming Systems Research and Extension
Jonice Louden ................................... 87

8 A Comparison of Rural Women's Time Use and
Nutritional Consequences in Two Villages in Malawi
Lila E. Engberg, Jean H. Sabry,
and Susan A. Beckerson........................... 99

9 Correcting the Underestimated Frequency of the
S Head-of-Household Experience for Women Farmers
Art Hansen ....................................... 111

10 An Evaluation of Methodologies Used in Time
Allocation Research
Eva Wollenberg.................................. 127

11 Gender, Resource Management and the Rural
Landscape: Implications for Agroforestry
and Farming Systems Research
Dianne E. Rocheleau M ........................... 149


VI2 Farming Systems Research in the Eastern Caribbean:
An Attempt at Analyzing Intra-Household Dynamics
Vasantha Chase.................................. 171

13 Economic and Normative Restraints on Subsistence
Farming in Honduras
Eunice R. McCulloch and Mary Futrell.............. 183

14 Phases of Fanning Systems Research: The Relevance
of Gender in Ecuadorian Sites
Patricia Garrett and Patricio Espinosa............ 199

15 Technological Domains of Women in Mixed Farming
Systems of Andean Peasant Communities
Maria E. Fernandez .............................. 213


16 The Household Enterprise and Farming System
Research: A Case Study from Taiwan
Rita S. Gallin and Anne Ferguson.................. 223

17 The Contribution of Women to Agriculture in Taiwan
Jane E. Gleason.................................. 237

18 Gender-Differentials in the Impact of Technological
Change in Rice-Based Farming Systems in India
Bahnisikha Ghosh and Sudhin K. Mukhopadhyay....... 253

19 Women in a Crop-Livestock Farming Systems Project
in Santa Barbara, Pangasinan, Philippines
Thelma R. Paris................................... 269

20 Gender Related Aspects of Agricultural Labor
in Northwestern Syria
Andr6e Rassam and Dennis Tully..................... 287


21 The Women's Program of the Gambian Mixed
Farming Project
Margaret Norem, Sandra Russo, Marie Sambou,
and Melanie Marlett............................... 303

<22 intra-Household Dynamics and State Policies as
SConstraints on Food Production: Results of a 1985
Agroeconomic Survey in Cameroon
Jeanne Koopman Henn......................... .. 315

.23 Intra-Household Gender Issues in Farming Systems
in Tanzania, zambia, and Malawi
Jean M. Due...................................... 331

24 Institutional and Policy Parameters Affecting Gender
Issues in Farming Systems Research in Tanzania
Manasse Timmy Mtoi .............................. 345

25 The Impact of Modern Changes in the Chitemene
Farming System in the Northern Province of Zambia
Mary N. Tembo and Elizabeth Chola Phiri........... 361

26 A Diagnostic Survey of Female-Headed Households
in the Central Province of Zambia
Robert E. Hudgens ............................... 373


27 The Gender Factor and Technology Options for
Zambia's Subsistence Farming Systems
Alistair J. Sutherland........................... 389

28 Using Male Research and Extension Personnel to
Target Women Farmers
Anita Spring ................................... 407

- 29 The Role of Women Farmers in Choosing Species
for Agroforestry Farming Systems in Rural
Areas of Ghana
Kofi Owusu-Bempah................................ 427

OmONRIBUTIRS........................................ 444

Tables and Figures


2.1 Projects Responding to Survey .................... 22

2.2 Types of Intra-Household Data Collected by
Projects Responding to Survey .................... 25
2.3 Uses of Types of Intra-Household Data by
Projects Responding to Survey .................... 27

2.4 Most Frequently Used Methods of Data Collection
by Type of Data................................ 29
2.5 Constraints Influencing Projects................. 33

8.1 Production Activities........................... 103

8.2 Mean Time Spent in Hours by Husbands and Wives in
Two Seasons in Each Farming System ............. 105

8.3 Mean Amount of Traditional Foods Stored by Type
of Farming Household in Two Seasons.............. 107

8.4 Meals and Maize Flour Consumed and Percent of
Children with Normal Weight for Height by Type
of Farming Household and Season................. 108

9.1 Marital Changes as Expressed in Migration
of Women in and out of Rural Areas of
Northwestern Zambia, 1971 to 1972............... 116

9.2 Changes in Gender of Household Head in
Lilongwe Rural Development Project, Malawi
1980/81 to 1982................................. 119

9.3 Patterns of Changes of Gender of Household
Heads in Intensively Surveyed Sub-sample......... 121

10.1 Distribution of Time by Activity and Gender...... 139

10.2 Percent of Time Spent by Gender, Location,
and Field Types............ ......................... 141

12.1 Communities Sampled by Area Focused Study
in Mabouya Valley................................. 176

12.2 Agroecological Characteristics of
the Mabouya Valley................................ 177

12.3 Socioeconomic Characteristics of
the Mabouya Valley................................ 178

13.1 Annual Average Yields for Upland Intercropping
of Maize, Sorghum and Beans...................... 187

13.2 Recommended Daily and Yearly Intakes of Nutrients
for Average Family .............................. 190

13.3 Subsistence Activities and Nutritional
Adequacy.......................................... 191

13.4 Foods and Their Contribution to Low Cost Diet ... 194

16.1 Population of Hsin Hsing by Period and Age,
1958-79........................................ 226

16.2 Occupations of Hsin Hsing Households, 1979....... 229

16.3 Land Use of 61 Farming Households,
Hsin Hsing, 1978-79............................ 230

17.1 Income Earned from Off-Farm Employment.......... 241

17.2 Labor Usage (in hours) by Gender and Crop........ 245

17.3 Crops and Total Hectarage of Farms without
Female Family Labor............................ 248

17.4 Crops and Total Hectarage of Farms with
Female Family Labor............................ 249

17.5 Diversification Indices of 30 Farms.............. 250

18.1 Time Allocation of Male and Female Population:
Hours per Person per Day ....................... 257

18.2 Male-Female Differences in Labor Input in
Rice Cultivation............................... 258

18.3 Time Allocation by Male and Female Population
in Economic and Home Production Activities:
Bhandarkona (hours per person per day)........... 262

18.4 Labor Use Functions: Household Survey Data (Nadia):
Pooled (Log Linear)............................ 265

18.5 Labor Use Functions: FMS Data (Hooghly and Birbhum)
(Log Linear)
Dependent Variable: Share of Women in Total
Labor Use....................................... 266

19.1 Percentages of Men, Women, and Children in Sample
Households Providing Different Sources of Labor by
Specific Activity in Rice Production............. 275

19.2 Percentages of Men, Women, and Children Providing
Different Sources of Labor by Specific Activities
in Mungbean Production ......................... 276

19.3 Percentages of Men, Women, and Children Providing
Different Sources of Labor by Specific Activities in
Cowpea Production............................... 277

19.4 Sample Households Using Different
Sources of Labor by Specific Activities
in Squash Production........................... 278

19.5 Percentages of Men, Women, and Children in Sample
Households Providing Different Sources of
Labor in Specific Activities in Cattle
and Carabao Production ......................... 279

19.6 Percentages of Men, Women, and Children within
Households Using Different Sources of
Labor by Specific Activities in Poultry
and Swine Production........................... 281

20.1 Contribution of Males and Females as Percentages of
the Total Time Spent in On-Farm Agricultural
Production....................................... 291

20.2 Contribution of Males and Females as Percentages
of Hours Spent in Legume and
Cereal Production............................... 292

20.3 Contributions of Males and Females
as Percentages of Hours Spent in Summer Crop
and Tree Crop Production ....................... 293

20.4 Mean Hours of Labor in Cereal Harvest:
Differences Between Farms Using Mechanical and
Manual Techniques.............................. 294

20.5 Regression Results on Labor Variables............ 298

21.1 Maize/Cowpea Yields of Participating Villages.... 310

22.1 Men's and Women's Average Labor Hours............ 320

23.1 Percentage of Labor Days Contributed by Females
by Operation by Crop, Sampled Families,
Kilosa, Tanzania, 1980 ......................... 335

23.2 Comparison of Crop Acreage, Income, and
Extension Visits Between Contact, Non-Contact,
and Female-Headed Households, Tanga Region,
Tanzania, 1985................................. 342

24.1 Monthly Labor Requirement for Farm Operations
for Major Crops in Eastern Uluguru Mountains..... 347

24.2 Tasks Performed by Gender, by Crop,
and by Activity in the Farming Systems of
Eastern Uluguru Mountains, 1984.................. 350

24.3 Estimation of Effective Adult-Equivalent
Work-Days Available per Month for Farm Work..... 352

24.4 Labor Requirements and Yield on the
Alternative System of Production................. 353

24.5 Expected Returns and Standard Deviation
of the Two Systems of Production................. 354

25.1 Adequate Nutritional Status of Children.......... 368

25.2 Percentage of Female-Headed Households
by Province.................................... 369

26.1 Main Characteristics of Male and Female-Headed
Households in TRD#3 (Mkushi District)............ 378


26.2 Comparison of Female-Headed Households in
TRD#3 (Mkushi District) by Presence of Males
in the Household, Oxen Ownership, and Size of
Acreage Cultivated ................ ............ 380

26.3 Comparison of Statistics on Female-Headed
Households in Three Recommendation Domains
in the Central Province......................... 384

28.1 Graduates of Malawi Agricultural Training
Institutions: Bunda College of Agriculture,
Colby College of Agriculture, and Thuchila
Farm Institute (Farm Home Assistants),
by Year and Sex................................. 410

28.2 Type of Extension Contacts for Male Household
Heads (MHH), Female Household Heads (FHH), and
Wives from the NSSA Extension Survey, Malawi
(in percentages), 1980-81....................... 414

28.3 Maize Yields from Farmer-Managed On-Farm Trials
Phalombe, Malawi, 1981-82....................... 417

29.1 Farmers' Household Type and the Average Number of
Forest Products They Know By Age Group........... 431

29.2 Goals of Farmers in the Forest Savannah
Transitional Zone of Ghana ..................... 433

29.3 Agricultural Systems (Farm Types) in the Forest
Transitional Zone of Ghana ..................... 434

29.4 Types of Deforestation and Land Degradation
Activities in the Forest Savannah
Transitional Zone of Ghana...................... 436


3.1 A Conceptual Map for Looking at the Farm-Household
System from a Gender Perspective................ 46

11.1 Pananao Sierra,Dominican Republic................ 154

11.2 Fakot Village, Bhaintain Watershed, India........ 158

11.3 Sources of Livelihood (Cash & Kind)
in Fakot....................................... 159

11.4 Misamfu, N.E. Zambia............................. 162

12.1 Agroecological Transect of the Mabouya Valley
of St. Lucia.................................... 175

13.1 Conceptual Model of Subsistence Farming-
Nutrition System................................ 188

14.1 Phases of Farming Systems Research: A Synthesis
of INIAP and ICTA Procedures.................... 201

18.1 Allocation of Time to Different Activities:
Male-Female Differential........................ 260

18.2 Gender Differential in Allocation of Time
to Different Activities by Socioeconomic
Groups and by Technology......................... 263

24.1 Conceptual Framework of Institutional,
Cultural and Policy Factors Affecting
Intra-Household Dynamics in FSR................. 348

24.2 E,Jv Relationships Generated by Traditional
and Alternative Farm Production Systems.......... 355

26.1 Busiest Months for Male and Female-Headed
Households..................................... 382

28.1 Grain Yield Response for Local Maize (L) and
CCA Composite (C) to Environment, without
Fertilizer, Phalombe Project, Malawi............. 419

28.2 Grain Yield Response of Local Maize (L) and
CCA Composite (C) to Environment, with
Fertilizer, Phalombe Project, Malawi............. 419

29.1 Diagrammatic Presentation of a Woman Farmer's
Traditional Agroforestry Farm in the Forest
Savannah Transitional Zone of Ghana.............. 432

29.2 Proposed Field Layout of On-Farm Agroforestry
Research Plots for a Typical Village in the
Forest Savannah Transitional Zone of Ghana....... 438


This book is the product of an international
conference hosted by the Women in Agricultural
Development (WIAD) Program at the University of Florida
from February 26 to March 1, 1986. The WIAD Program's
general purpose is to promote an understanding of gender
and its relevance for agricultural development processes.
Women are critical to agricultural production in many
places, but access to resources and research technologies
may be constrained by gender and this may lead to detri-
mental effects on the design and implementation of sound
agricultural programs. The Program seeks to support and
develop expertise related to the roles of women and
intra-household dynamics in agricultural production,
research, and extension in order to improve the design
and implementation of agricultural programs. The
Program's primary audience is the faculty and students of
the University of Florida, but the Conference and this
book permitted us to reach out to a wider national and
international audience.
The Women in Agricultural Development Program began
as the Women in Agriculture (WIA) group at the University
of Florida (UF) in 1983 after five years of informal
activities related to women and development. It became
an official University of Florida program in 1984 and the
name was changed to Women in Agricultural Development in
1986. During early years, several initiatives emerged
from the International Programs office and from social
scientists, especially those in the Centers for African
and Latin American Studies. Momentum built up as a
critical mass of faculty came to the UF with experience
in the WID field. A group that formed in the fall of
1983, with Anita Spring as Director and Marianne Schmink
as Co-Director, served as an ad hoc committee to promote
awareness of issues related to WID and, more

specifically, to women's role in agricultural production.
The WIAD group received support from the USAID Program
Support Grant funds administered by the International
Programs office and from other on-campus sources
including the Centers for Latin American and African
Studies and the Graduate School. Principal program
activities include a bi-weekly speaker series and a
bibliographic compilation of readings on women in
agriculture, as well as work in curriculum development
and in technical assistance.
From the beginning, the WIAD Program sought to focus
its efforts specifically on the realm of agriculture, in
recognition of the UF's strengths in this area. It took
advantage of the opportunity to interface directly with a
major USAID-funded project, the Farming Systems Support
Project (FSSP) that had a worldwide mandate for technical
assistance, training, and networking activities. As
Associate Director of the FSSP based at the UF, Susan
Poats was in a key position to facilitate that linkage.
In addition to the institutional presence of the FSSP,
the general philosophy and methodology of the "farming
systems" approach provided promising avenues for atten-
tion to women. Yet gender had not been incorporated
effectively and systematically as a variable in most
farming systems work. The WIAD Program hoped to stim-
ulate the development of conceptual and methodological
approaches that could improve the incorporation of gender
issues into farming systems activities. These ideas are
discussed in more detail in Chapter 1.
In early 1984, the WIAD Program hosted a visit to the
UF by several colleagues involved in an effort by the
Population Council to produce useful case studies of
gender issues in development projects. This meeting led
to the creation of the FSSP/Population Council Intra-
Household Dynamics Case Studies Project, co-managed by
Susan Poats and Judith Bruce of the Population Council,
with funding from USAID and from the Ford Foundation.
The FSSP created a new task force to focus on the
integration of household and family concerns into the
farming systems perspective. In 1984, Cornelia Butler
Flora (a member of both the FSSP's Technical Committee
and of the task force on the family) visited UF. She
also wrote a position paper on "Intra-Household Dynamics
in Farming Systems Research: The Basis of Whole Farm
Monitoring of Farming Systems Research and Extension."
She and other task force members were instrumental in
supporting the proposal to develop case study training
materials through the FSSP/Population Council project.


The case study project, initially launched from a
WIAD activity, generated further momentum. In February
of 1985 the project's advisory committee contacted those
individuals and projects around the world who might be
interested in developing case study materials from their
own experiences in farming systems work. A survey
questionnaire requested more detailed information from
those who responded. The results of this questionnaire
provided some indication of the current state of know-
ledge on intra-household dynamics in on-going farming
systems projects (see Chapter 2). The FSSP/Population
Council project selected eight proposals to receive
support for the development of case study materials. Yet
the more than 75 respondents who expressed interest in
the project indicated that the incorporation of intra-
household variables was a greater concern to farming
systems practitioners than had been anticipated.
Parallel to the case studies project, the WIAD
Program proposed bringing together farming systems
practitioners to a conference on the UF campus. The
focus of the event would be purposely narrow in order to
maximize the potential for improving the integration of
gender into the farming systems approach, taking
advantage of the interest aroused by the case studies
initiative. Sessions would address specific issues of
theory, method, and policy related to Farming Systems
Research and Extension (FSR/E) in developing areas,
systematically comparing African, Latin American, and
Asian experiences. Marianne Schmink volunteered to
organize the conference and Susan Poats coordinated the
format of the program. WIAD's Steering Committee served
as an advisory body and a conference task force was
appointed. Funding came from the Center for Latin
American Studies, the Center for Tropical Agriculture,
and the Center for African Studies, as well as from the
Ford Foundation and the Rockefeller Foundation. A call
for papers circulated in July of 1985 defined specific
topical areas of concern and asked potential participants
to summarize the data they would present, its relation to
FSR/E, the geographic focus, and its relevance to the
designated themes.
The response was overwhelming. Over a hundred paper
proposals arrived from all over the world, expanding the
geographic focus of the conference. In an effort to
accommodate as many participants as possible, the organ-
izers structured a comprehensive program of concurrent
sessions and explored as many sources as possible for
travel support, giving priority to visitors from develop-
ing countries. Over 40 participants received full or


partial funding to attend the Conference. There were 91
speakers in fifteen formal paper sessions, three round-
table discussions, and an after-lunch lecturer. Carmen
Diana Deere of the University of Massachusetts at Amherst
delivered the keynote address on "Rural Women and State
Policy: An Evaluation of the Decade" (later published in
revised form in her book with Magdalena Leon, Rural Women
and State Policy, Westview, 1987). Films and recent
publications on FSR/E and gender were available for
review during the Conference and several social events
facilitated a high level of information exchange and
networking. Attendance at conference sessions was
consistently high. A total of 298 persons were register-
ed or on the formal program and many others from the
campus and local area attended without registering. The
magnitude and complexity of the Conference demanded the
assistance of many volunteer helpers. Special thanks are
due to Carol Brown, Donna Epting, Jean Gearing, Janet
Hickman, Patricia Kuntz, Cindy Lewis, Greg Moreland, Bill
Reynolds, Barbara Rogers, Robin Sumner, and Darla Wilkes.
Assistance was also provided by Wharton Williams Travel,
Classique Cuisine, Renaissance Printing, Farming Systems
Support Project, International Programs, International
Food and Agricultural System (IFAS), and the College of
Liberal Arts and Sciences.
Chris Andrew, Director of the FSSP, formally opened
the Conference on behalf of the University of Florida.
In his opening remarks he said:

The intellectual capacity assembled in this
room to address gender-related issues in on-farm
research is second to none. We are fortunate
this week to have presentations by people who
have traveled from more than twenty countries,
from forty universities (eleven non-U.S.), from
three International Agricultural Research Cen-
ters, and many other national, donor, private
voluntary organizations, and private entities.
The titles in the conference program mention
twenty-two countries and we know that more will
be discussed. Nearly every region of the world
is represented.
The intra-household concept, as an integral
and dynamic part of the farm system, must con-
sider gender issues. I emphasize integral and
dynamic because we must take care not to esta-
blish artificial boundaries to accommodate
conventional simplicity. With a broad view of
the household, third world food security can be

addressed in a national framework. We need a
conceptual focus that recognizes the farm house-
hold as a critical element in successful agri-
cultural policy for research and development.
To enunciate and address the interdependent
needs of women and men in farming systems as
they interact with the bio-physical environment
is an important goal.
The critical importance of agriculture to the
vitality and strength of many of the world's
countries is widely recognized. At the same
time, the increasing diversity, complexity, and
intractability of the problems facing agricul-
tural development make it imperative that agri-
cultural research systems change and adapt to
address specific realities. One of these is the
role of women in agriculture. Another is that
gender issues are being considered together with
the farming systems approach, an agenda that
likely could not have found a platform ten years
ago. In the ebb and flow of agricultural
research, this Conference represents something
of a revolution in agriculture and in

The Conference offered an excellent opportunity for
exploring and testing new ideas and successful approaches
for incorporating gender sensitivity in agricultural
research and development. The FSSP/Population Council
project used the Conference to test a case study based on
FSR/E project activities in Zambia. The two authors of
the case, Charles Chabala and Robert Nguiru, were present
to view how the case worked in a training context.
Hilary Feldstein, Rosalie Norem, Kate Cloud, Susan Poats,
Nadine Horenstein, and Mary Rojas conducted the
abbreviated training sessions with approximately sixty
conference participants. Their detailed evaluations and
suggestions were used to improve the case. Highlighted
repeatedly during the Conference was the need to
inventory and assess the usefulness of various methodolo-
gies used in dealing with gender issues. An ad hoc
special methodology workshop held on Saturday afternoon
following the close of the formal sessions was attended
by more than seventy-five people. Janice Jiggins and
Hilary Feldstein are using this session as the basis for
developing a methodologies handbook to accompany the
FSSP/Population Council case studies.

Aside from the high level of interest in the formal
and ad hoc sessions, informal networking was intense
among those present. These interactions underscored the
remarkable level of interest in the practical aspects of
integrating attention to gender issues into agricultural
research and extension. The Conference provided a
meeting place for people with diverse disciplinary and
area expertise to learn from one another. The most
exciting outcome of the Conference was the discovery of
such a broad base of research already underway on gender
issues in agriculture.
A total of sixty-four written papers were available
in a three-volume set at the Conference. In order to
make conference materials more widely available, the WIAD
Program decided to organize a selection of the interna-
tional papers for publication as a book. Although the
quality of most of the papers presented was high, space
considerations made it impossible to publish all of them.
This volume includes less than half of the papers
presented at the Conference. In making the difficult
decision as to which papers to include, the editors used
several criteria. Most important was adherence to the
original thematic focus on farming systems work, as
outlined in the call for papers. Secondly, preference
was given to papers that presented new ideas or methodo-
logical approaches for the integration of gender into
FSR/E projects. Finally, regional balance and the
inclusion of non-U.S. authors who might not otherwise
find an audience for their work in this country was
sought. The overall goal was to produce a coherent
reader with a comparative perspective, that would be both
stimulating and helpful to people trying to implement
gender-aware farming systems projects. We hope the many
excellent papers that could not be included in this
volume will find publication outlets elsewhere.
This book, and the WIAD Program's overall effort,
have depended on the hard work and support of a dedicated
group of faculty, students, and staff from the UF. A
special thanks is due to Jean Gearing, who was respon-
sible for much of the editing, correspondence, typing,
and organizational tasks associated with the book. The
editors are also grateful to Pamela Shaw for her editor-
ial work, to Kathy Gladden for bibliographic and clerical
assistance, to Sharon Leslie and Barbara Rogers for their
assistance in artwork, and to Lana Bayles, Kenna Hughey,
Shirlene Washington, Sabrina Byron, and Dana Whitaker who
provided additional help with the word processing. The
book and the Conference that produced it are, in turn,
part of a larger effort in which the WIAD organizers have


been privileged to collaborate with many colleagues, some
of whom are mentioned above. It is hoped that the ideas
and experiences discussed by the authors of the following
chapters will help to stimulate and improve the consid-
eration of gender as a crucial issue in agricultural
production and development.

Susan V. Poats
Marianne Schmink


Linking FSR/E and Gender:
An Introduction

Susan V. Poats, Marianne Schmink, and Anita Spring

The title of this book is like a code. The two terms
"gender issues" and "farming systems research and exten-
sion" are shorthand. Each represents an extensive field of
research and practice: women and development or WID, and
farming systems or FSR/E, respectively. The two fields
have much in common. Both emerged relatively recently in
response to dissatisfaction with the results of technolo-
gical change in agriculture in developing countries.
Whereas in the 1950s and 1960s development theory and
practice emphasized growth in productivity, by the 1970s
there was a renewed concern to implement programs that
conceived of development more broadly, to mean the possi-
bility of better lives for most people. This perspective
challenged a development field dominated by technical and
economic expertise. Efforts to develop more comprehensive
approaches that would bring together technical, economic,
and social considerations led to the two interdisciplinary
fields of WID and FSR/E.
In this brief introduction, justice cannot be done to
either field in its own terms. Rather, the historical and
practical considerations that favor their interaction and
the conceptual problems such a union can help to overcome
are reflected. The discussion will indicate how the
following chapters in this book contribute theoretical and
methodological insights that can help to make agricultural
development programs more efficient and equitable.


Most farms in developing countries are small scale,
with few resources other than family labor. Their
subsistence activities are multifaceted and their goals

complex, including both market and non-market considera-
tions. Minimizing risk is especially important when family
survival is at stake. Given the constraints they face,
small farmers actively seek ways to improve their produc-
tivity and to maximize the few resources at their disposal.
Agricultural technologies (including equipment, inputs such
as fertilizers and pesticides, and management practices)
are often designed for farmers with greater resources and
market orientation based on an essentially economic cal-
culation of costs and benefits. By the 1970s, development
practitioners became concerned that the benefits of agri-
cultural innovation accrued most easily to these wealthier
farmers. Yet small farmers constituted the majority of
producers and, ultimately, those most directly responsible
for the welfare of rural families and communities. What
were the social and economic costs of neglecting them? The
design of technologies appropriate to the majority of low-
resource farmers required an understanding of their parti-
cular constraints, goals, and practices that went beyond
strictly technical and economic criteria. The farming
systems concept emerged as a response to this challenge.
FSR/E is not a single approach, but an array of differ-
ent perspectives and methods. This diversity is a source
of debate and dialogue that continually enriches the field.
In this book there is likewise no orthodoxy, but rather a
collection of different points of view as to how to concep-
tualize and carry out farming systems work. The common
elements that underly most versions of farming systems
include: an explicit commitment to low-resource producers;
a systems approach that recognizes the complexity of small
farm enterprises; a focus on the farm family or household;
and a recognition of the importance of including farmers in
the research and extension process. The concept of
"domains" is used to denote the specific client group
(defined by environmental, ecological, and/or socioeconomic
criteria) to whom the project is oriented.

The WID field, similar to FSR/E, began with a concern
for the distribution of development benefits. Like farming
systems, women and development is far from a unified field
of knowledge. Not only does it include many strands of
research and practice, but the field has evolved rapidly
over the approximately 20 years of its existence, since
economist Ester Boserup published her ground-breaking work
Women's Role in Economic Development in 1970. Boserup's
work challenged the prevailing notion that economic

development, or modernization, would automatically improve
women's status by replacing traditional values and economic
backwardness with new opportunities and an egalitarian
ethos. She argued instead that economic innovations often
replaced women's traditional economic activities with more
efficient forms of production controlled by men. Examples
included the decline of women's cottage industries due to
competition from factories hiring predominantly men and, in
some parts of the world, the growth of modern service and
commerce sectors in which men predominated, in place of
women's traditional marketing practices. The recognition
that development, as practiced, might actually worsen
women's position relative to men's crystallized the new
field of women and development around a concern with
By the late 1970s, however, a growing research base on
women's economic activities showed that equity was inti-
mately related to more technical problems of efficiency and
productivity. If development undermined women's tradition-
al economic contributions, was this loss compensated by the
output of new forms of production? Were new economic
opportunities opening up for women? What was the impact of
these shifts on the welfare and productivity of the poor
populations of the world? The new emphasis on the poor
focused attention on women's importance as household pro-
ducers and providers in addition to their domestic roles.
No longer were they to be viewed simply as potential wel-
fare beneficiaries whose needs might be neglected by devel-
opment efforts. Instead, women were a mainstay of family
and community welfare, active producers whose potential
contributions were often overlooked or undermined. A
clearer understanding of changes in women's role in produc-
tion therefore was essential for the success of agricultur-
al development projects.
A decade of theoretical experimentation and empirical
research on women's role in development moved the field
from the stage of raising awareness and clarifying issues
to a search for practical applications. How could the WID
insights be applied to development work? One solution was
to create special projects or components devoted to women.
While sometimes successful, these all too often emphasized
women's domestic responsibilities rather than their pro-
ductive work. They also distracted from the more general
problem of improving the effectiveness of "mainstream"
development projects by making them more responsive to
gender differences among the client population. By what
practical means could such a formidable task be undertaken?

The first attempts to answer this question produced an
array of checklists of questions to be asked and data to be
gathered in each project setting. A series of case studies
were published as examples of how gender affected develop-
ment projects. Various institutions compiled handbooks
that specified how gender issues could be addressed at each
step of the project cycle. But there were not enough
experts trained in the analysis of such complex and varia-
ble matters as household division of labor, decision-
making, and income management. Some of the basic issues
could be specified in advance, but each setting required a
unique assessment of their relevance and of the interaction
with other important variables. While hiring more women as
project staff members appeared to be a good idea, the
gender of the researcher or practitioner turned out to be
no guarantee of the requisite analytical skills.
In response to this dilemma, WID efforts in the 1980s
sought to develop the tools of "gender analysis" and the
methods by which development practitioners could learn and
adopt them. USAID fostered a major effort to adapt the
Harvard Business School's case study teaching method to
training on gender issues in development projects. The
Office of Women in Development sponsored the writing of
several new analytical case studies that were compiled in a
handbook that also provided a framework and set of basic
concepts to be used in the case study analysis (Overholt et
al. 1985). The cases and the training method have been
widely used in training workshops that provide practice in
tackling a set of questions that might otherwise seem hope-
-lessly complicated. The strength of this approach is its
emphasis on the link between project or development goals
and gender differences in the client population. This
focus helps to clarify the relevant issues and to indicate
priorities for research and action.


The farming systems perspective is especially appro-
priate for such a process-oriented approach to gender
analysis. The FSR/E methodology consists of a series of
stages (diagnosis, planning and design of technology,
experimentation and evaluation, and dissemination) that
facilitate the specification of steps to be taken to
address particular aspects of research and extension and
how to make the best use of different kinds of data. But


FSR/E is also conceived to be an iterative, adaptive
process in which, once the project is well underway, the
various stages of research take place simultaneously. This
philosophy is intended to maximize the potential impact of
on-going farmer evaluations on the design and dissemination
of future technological changes. The research process
allows time to learn about the intricacies of farming sys-
tems and to incorporate new insights into more refined
measures and project adaptations. Other characteristics of
the farming systems approach especially important for gen-
der analysis include its focus on small farm households and
on the participation of farmers in the research and exten-
sion process.

Disaggregating Development Beneficiaries

The farming systems emphasis on reaching specific low-
income groups helped to illuminate women's roles in agri-
cultural development. Identifying small farmer constituen-
cies required the disaggregation of society into "target"
or "client" groups which brought women's activities into
greater focus. The interaction of socioeconomic standing
and gender was brought home by the growing recognition that
women in poor families played essential economic roles
that bore little resemblance to the activities of middle-
class and elite women in the same societies. These obser-
vations were confirmed by mounting evidence from research
that documented poor women's multiple economic activities,
low earnings and long work hours, and restricted access to
productive resources. Women played a central role in the
low-resource farm households that were the focus of farming
systems work.
The surprisingly high and growing proportions of
female-headed households dramatized women's economic impor-
tance in poor populations and revealed the extra con-
straints under which they often labored to achieve family
welfare (Buvinic and Youssef 1979). Rural out-migration of
men was rising in many parts of the world as a result of
development, leaving many women either temporarily or per-
manently in charge of their households (Palmer 1986).
Their efforts were often undermined by labor constraints or
by lack of access to productive resources, in part because
research and extension services were primarily oriented to
male farmers. The focus on female-headed households illus-
trated how disaggregation of beneficiary populations could
more precisely delineate appropriate interventions for
specific social groups. It also undermined the assumption
that development projects focused on male farmers always
would have the most effective impact on family welfare.

The Whole Farm System

The systems approach endorsed by the FSR/E constituency
lent itself well to illuminating women's economic impor-
tance. Small farm enterprises encompass multiple activi-
ties whose interaction is key to understanding management
decisions and practices. The configuration of a given
system changes readily over time in response to both inter-
nal and external factors. This holistic, dynamic perspec-
tive on small farming enterprises provided a framework
within which the family division of labor could be a key
focus. Social definitions of which tasks would be carried
out by men or women vary from one society, region, class,
or ethnic group to another. This variability indicates
that the division of labor is determined not by the phy-
sical difference between the sexes, but by the social
definitions of proper relations between women and men. The
concept of "gender" serves to distinguish the social char-
acter of these relationships, and the "sexual division of
labor" describes the allocation of tasks and responsibi-
lities to men and women in a particular situation. In
practice, farming systems practitioners may disaggregate
only so far, stopping short at the analysis of the division
of labor within the household or family. Agricultural
research has historically focused on specific commodities
whose production is market-oriented. FSR/E recognized that
small farm enterprises combine crops and animals. Yet the
perspective still overlooked other essential activities
carried out by farm families, including off-farm work,
home-based production for use or exchange, and the work
required to maintain the home and its inhabitants. WID
research revealed that women were often predominant in
these activities, especially those based in the home that
tended to be overlooked or viewed as merely "domestic"
work. While men often specialized in income-generating
activities, women typically combined household management,
child care, and work to generate earnings (both on and off
the farm). These competing demands on their time could
serve as a significant constraint to the adoption of new
forms of production that relied on women's labor.
In small farm households, decisions reflected priori-
ties and constraints related to a variety of activities and
goals, not just to those related to cash crop production.
The potential trade-offs between resources devoted to agri-
cultural production and investments in improved family
nutrition were of particular concern to farming systems
practitioners whose objective was to stabilize or enhance
rural welfare. The systems approach adopted by FSR/E
practitioners provided a starting point for integrating the

diversity of farm and non-farm activities within a more
complex model of the whole farm-household system. WID
practitioners collaborated by focusing attention on women's
importance in agricultural production, but also in focusing
on activities not generally defined as "production" that
are nevertheless essential to the well-being and economic
livelihood of rural households and communities.

Intra-Household Dynamics

The focus on farm families brought development work
much closer to the realities of poor families than was
possible using the country level statistics. The concept
of "household", sometimes used to denote a residential
unit, sometimes synonymous with the nuclear family, was
useful in the field of development and in the social
sciences in general (Schmink 1984). It provided an
intermediate level of analysis (between the individual and
the aggregate society) and a convenient unit for the col-
lection of empirical data. The existence of such primary
domestic units in virtually all societies implied an
attractive universality for the concept of household that
was familiar to researchers and practitioners from their
own personal existence. These perceived advantages, how-
ever, had hidden drawbacks. Development practitioners
often generalized from their own experience, presuming that
households elsewhere were similar to those in which they
lived, when in fact household structure and functioning is
highly variable. Whereas in advanced industrial society
productive work is largely separated from the home, the
same is not true for agrarian communities. Home-based food
processing, handicrafts, care of animals, kitchen gardens,
and manufacturing of such useful items as soap and clothing
for a peasant family are not analagous to the domestic
chores of a middle-class urban housewife. They are prod-
uctive tasks essential to household welfare. Whereas a
U.S. household typically depends on one or two monetary
wages for its sustenance, rural families in the developing
world rely on a diverse set of paid and non-paid activities
for survival.
In many societies women and men have quite separate
responsibilities, access to distinct resources, and /
differentiated control over returns from their own
activities. In fact, households are themselves systems of
resource allocation (Guyer 1981). The unitary neo-
classical view of household income inherited from the
advanced industrial nations is especially inappropriate for

such complex situations in which household members have
access to different resources and work opportunities, and
exercise differing degrees of control over separate income
streams that flow through the household. Household
decision-making is neither necessarily unitary nor
harmonious. Different members may decide about production
strategies, contribute labor to specific tasks, or bear
responsibility for the use of the commodities produced.
The complexity of intra-household dynamics implies that the
possibility of competing goals or priorities may require
negotiation among household members. Households are also
fluid; variability stems from responses to exogenous
changes (such as male out-migration), from internal
differentiation based on class, income, ethnicity, and
culture, and from demographic variables within the
household unit (that is, the pattern of family formation,
or the "life cycle" of the family).

Farmer Participation

The internal dynamics of small farm households affect
the process of client involvement in the research and
extension process. If work responsibilities, control over
resources, and decision-making are fragmented within the
family unit, who are the appropriate partners in the
research process and potential beneficiaries of the
proposed technologies? Since male household heads are
Typically the public representatives of family groups, it
is often assumed that information and resources conveyed to
them will "trickle across" to others in their household.
But indirect communications strategies are inefficient and
may omit the actual "user" from the process of FSR/E. This
omission represents a loss of valuable indigenous knowledge
and may lead to inadequate or incomplete application of
technological innovations. Since women and men may know
about different factors relevant to agricultural produc-
tion, the labor of one may not necessarily substitute for
the other. If farming systems projects are to succeed in
forging effective collaborative ties with their client
population, they must include both women and men farmers as
partners in the research process.


The chapters of this book are a selection of the papers
concerned with developing countries in Latin America, Asia,
Middle East, and Africa presented at the Conference in 1986
at the University of Florida (see Preface). The

Conference's primary objective was to bring together
scholars and practitioners with expertise and interest in
FSR/E to discuss state-of-the-art issues related to gender
in FSR/E. In order to maximize coherence, participants
were asked to prepare papers that would address specific
issues of theory, method, and policy related to FSR/E
across developing regions. The following questions were
posed under several topical areas:
The Whole Farming System. How can key components of a
farming system, including non-farm activities, livestock,
secondary crops, food storage, and food processing, be
identified? How does the division of labor by age and
gender constrain or facilitate specific economic or
productive goals? What is the potential impact of improved
agricultural technology on each of these goals and on
household members responsible for specific production
Intra-Household Dynamics. What are the key aspects of
internal heterogeneity of household units: differential
access and use of resources within households; multiple
enterprises and their interactions; substitutability and
specialization of labor in agricultural activities; market-
ing outlets and their relationship to differing or con-
flicting priorities and needs within farm units; and how
might proposed interventions alter the balance of power and
Institutional and Policy Concerns. How does the sur-
rounding environment beyond the farm gate at household,
community, and other social levels differ for men and
women? What specific constraints to production are posed
by these gender differences? How can FSR/E address
constraints such as legal status, restrictions on mobility,
domestic obligations, property rights, access to credit,
markets, and employment?
Definition of Research Domains. How can the key
components and actors within household and farming systems
be identified? What are key constraints to productive
activities and how does access to production inputs differ
by gender? What is the significance of, and interactions
between, multiple enterprises within the farm household
and how do they create different labor requirements, goals,
incentives, markets, and priorities for different family
On-Farm Research and Extension. Who are the specific
audiences for direct involvement in on-farm research and
extension? How do labor constraints affect proposed
solutions and how do labor patterns impact on household
members who differ in their access to resources? How can

extension strategies be devised that are responsive to the
productive activities of both women and men? Which
extension mechanisms are most effective in reaching both
male and female farmers?
Monitoring and Evaluation. How can strategies be
designed for monitoring the differential impact of FSR/E
interventions on different individuals and enterprises
within the farming system? What are the unanticipated
effects of technological change? In what ways does
misunderstanding of gender issues lead to inadequate
planning and design or diminished returns to FSR/E
projects? How can these effects be minimized?
The papers contained in this book do not exhaust the
answers to these questions, but they do provide a
beginning. Authors were asked to include a common "minimum
data set" in their case study material to facilitate
comparison. The following chapters contain many innovative
approaches to conceptualizing and carrying out farming
systems projects that effectively take gender into account.
They highlight several features of the farming systems
approach that could be improved by more attention to gen-
der, and they suggest practical ways that this could be
done. The book presents a comparative perspective on the
relevance of gender to farming systems work in the develop-
ing regions. Two dimensions run throughout the various
chapters: the presentation of site-specific data that will
permit in-depth analysis of specific cases and the search
for conceptual and methodological innovations.
Part I presents a set of articles that focus on key
theoretical and methodological issues relevant to the
farming systems approach. In Chapter 2, Rosalie Norem
summarizes the results of a survey of arming systems
projects that collected data on intra-household dynamics
and gender differences. Project staff expressed a need for
more intra-household data, especially on the factors
determining household variability (such as out-migration
and the family "life cycle"), on specific labor constraints
stemming from the gender division of labor, and on income
management within the household. Her findings also reveal
that different kinds of information are useful at each
phase of a project. In Chapter 3, Alison Evans discusses
some of the problems with FSR/E procedures that impede the
effective integration of gender considerations, including
the emphasis on market criteria and measure, and the
assumed homogeneity of the farm household. She presents a
framework of ideas to help in constructing a broader, more
dynamic model of farm-household systems. She also discusses
the relevance of gender at different points in the FSR/E

process, and institutional constraints that must be
overcome to improve attention to gender.
Janice Jiggins, in Chapter 4, continues the discussion
by focusing on the problems of communication between
researchers, farmers, and extension workers, using examples
from Zambia and Lesotho that show the rationality and
flexibility within the domestic domain. She explores the
difficulties of reconciling scientific knowledge systems
with those of indigenous people in the course of conducting
on-farm research. Lessons drawn from her examples point to
the need for establishing key field-household interactions
at an early stage of the diagnostic process and to develop-
ing methods for mutual communication of key concepts across
researchers and female producers' distinct knowledge
systems. She proposes the use of situation-analysis based
on critical incident technique and peer group workshops as
appropriate methodologies for improving communication and
Amalia Alberti focuses in Chapter 5 on the problem of
generating data sufficiently sensitive to gender differ-
ences to guide the definition of client groups during the
initial phases of a project. Echoing Evans' point that a
priority for wealthier, more market-oriented farmers will
tend to exclude women, she discusses the pros and cons of
different sources of techniques for data collection in
FSR/E projects. The following chapter, by Peter Wotowiec,
Jr., Susan Poats, and Peter Hildebrand, explores in more
detail how definitions of client populations need to be
modified in accordance with the problems posed at different
stages of the project cycle. They offer a refinement of
the conventional FSR/E concept of "domains" to distinguish
between "research domains" (that maintain variability),
"recommendation domains" (homogeneous for technology
testing), and "diffusion domains" (for disseminating new
technologies). In Chapter 7, Jonice Louden summarizes the
compelling reasons for incorporating gender issues into
FSR/E monitoring and evaluation systems, especially in a
country such as Jamaica where women play key roles in
agricultural production. The FSR/E process presents an
opportunity to collect systematically valuable information
that can help to inform project implementation and refine
gender-sensitive measures of key indicators of development.
Chapters 8, 9, and 10 take up conceptual and method-
ological issues beyond the farming systems universe. Dis-
satisfied with the market bias of standard economic models
of the household, authors Lila Engberg, Jean Sabry and
Susan Beckerson propose an alternative production activity
model based on measures of time allocated to

income-generation, subsistence, and home production. The
more integrated model suggests trade-offs between labor
allocated to cash and to subsistence activities in Malawi
that could have nutritional implications. Also concerned
with the concept of household Art Hansen presents data, in
Chapter 9, from Africa that suggests caution in concep-
tualizing and measuring the frequency of female-headed
households. His findings show that static surveys may
underestimate the probability that a woman will be a head
of household at some moment or moments in her life, thereby
reinforcing the importance of involving both men and women
in development efforts. Eva Wollenberg's Chapter 10
discusses the strengths and weaknesses of various time
allocation methodologies and their relevance to farming
systems work. She explores how four different approaches
to the collection of time use data were used in a Philip-
pine project. Her discussion emphasizes the dynamic nature
of the FSR/E process, a theme common to all of the chapters
in this section. Gender patterns and intra-household
relationships become relevant to different degrees and in
different ways at each point in the project cycle. These
chapters, and others following, provide concrete sugges-
tions as to how an unfolding farming systems project team
can collect and analyze the information that will enable it
to develop and adapt production technologies to the needs
of different users.
Diane Rocheleau, in Chapter 11, draws upon experience
from a broad range of countries; she details a land user
perspective as an appropriate method for incorporating
women as clients and active participants in agroforestry
projects. Her paper and that of Owusu-Bempah (Chapter 29)
pli- an important role in expanding the horizon of FSR/E to
consider the rural landscape as the context and focus for
projects in order to address the gap between natural
resource management and farming systems research. The role
and domain of women in the interface of these two areas is
clearly laid out as the next critical frontier for
expansion of household research and gender-based analysis.
Beginning with Part II, the Chapters explore gender
issues in FSR/E on a regional basis. Women's roles in
agricultural production are less visible in Latin America
than irAiother developing regions. The same is not true in
the Caribbean nations such as St. Lucia,-described in
Chapter 12 by Vasantha Chase. While(island womernplay a
S significant role in farm work and decT1ion=making, they
receive less income and fewer extension services than do
male farmers. Informal data collection methods reveal that
female-headed households face particular labor and input
constraints that limit their output, choice of crops, and

amount of land planted. A concern with integrating food
consumption into the farming systems approach led the
Caribbean Agricultural Research and Development Institute
(CARDI) to recommend labor-saving methods of backyard
garden production, oriented to improvement of family
nutrition. The link between nutrition and agricultural
change is also the focus of Eunice McCulloch and Mary
Futrell's Chapter 13. Their measures of the nutritional
output of cropping activities reveal the "low level steady
state" farming system that maintains Honduran families at
risk of persistent malnutrition.
In Chapter 14, Patricia Garrett and Patricio Espinosa
describe the steps taken by the Bean/Cowpea Collaborative
Research Project in Ecuador to adapt project activities to
gender and social class differentiation. Their rich
discussion of the FSR/E process demonstrates the importance
of women's participation in production and decision-making,
even in Andean Latin America where farming commonly is
assumed to be the domain of men. The same is true in the
Peruvian highland community described in Chapter 15 by
Maria Fernandez, were women are responsible for most tasks
associated with livestock production. Recognizing women as
a key source of knowledge on traditional livestock produc-
tion practices, the project team experimented with a
variety of strategies to draw them into active participa-
tion. Fernandez' argument reflects the emphasis in
Rocheleau's and Owusu-Bempah's papers on agroforestry of
the importance of women's knowledge about traditional
resource management practices.
Part III presents five case studies from Asia and the
Middle East. Chapter 16 by Rita Gallin and Anne Ferguson
and Chapter 17 by Jane Gleason present case studies from
Taiwan. Gallin and Ferguson use longitudinal data from one
village to show that a limited focus on farming activities
ignored off-farm work and failed to analyze the interac-
tions between the agricultural and industrial sectors of
society. The authors propose the term "household enter-
prise" as a way of dealing with interrelated farming and
off-farm work, and encourage researchers to view off-farm
activities as "central rather than tangential to FSR/E
analysis." Gallin and Ferguson discuss farm mechanization
and note that it did not displace women, but rather concen-
trated certain tasks among some male specialists and caused
some women to assume managerial positions previously
restricted to men. Older women also took over tasks of
younger women who then sought off-farm employment.
Gleason's detailed labor study in Southern Taiwan
complements the previous work, and together they present a
good example of why generalizations about gender and

agriculture should not be made for an entire country.
Gleason argues that in Southern Taiwan, as agricultural
mechanization increased, more women than men were displaced
and forced to other sectors of the economy. In her study,
availability of female labor increased the variety of crops
grown and the level of diversification, indicating that
women will be the users of modern vegetable technology and
will be most affected by changes in vegetable production.
Chapter 18 by Bahnisikha Ghosh and Sudhin Mukhopadhyay
studies time allocation by men and women in a rice-based
farming system in West Bengal, India. Though female labor
is subject to sociological confsants, the contribution of
women is often larger than men's. A change-to new rice
tec-nf6ogy-incre se-tfemale-labor, however the authors show
that the increased workload falls within the home produc-
tion sector and-was largely unaccounted for in the tradi-
tional economic literature.
Chapters 19 and 20 are based on work conducted at two
of the International Agricultural Research Centers. Thelma
Paris, in Chapter 19, describes how women were successfully
integrated into a crop-livestock project of the Interna-
tional Rice Research Institute (IRRI), in the Philippines.
Beginning as observers and slowly integrating themselves by
collecting and disaggregating data on household and produc-
tion activities, the section members of the Women in Rice
Farming Systems were able to produce useful information and
become fully participatory members of the project. As a
result of their efforts, the whole team began to recognize
that specific production activities are the responsibility
of women and that on-farm research needed to target them.
Women's livestock, particularly swine, rootcrops, and
vegetables that had not been previously addressed by the
project, were proposed as new areas of research as a result
of the incorporation of women's concerns. In addition,
subsequent training courses included a significant number
of women participants, and the course addressed women's
production problems.
Andree Rassam and Dennis Tully in Chapter 20, discuss
research at the International Centre for Agricultural
Research in Dry Areas (ICARDA) on gender and agricultural
labor in Syria. They find that though male and female time
contributions to crop production are similar, males are
more often involved in new technologies, especially mechan-
ization, while females are more involved in more tradition-
al ones such as hand labor. The authors expect continuing

mechanization to further reduce female agricultural activi-
ties and they propose additional research to determine the
impact of these changes.
Part IV contains a number of case studies from Africa
that includes the descriptions of cropping systems, labor
patterns, and work in on-farm, farmer-managed trials. Some
themes that emerge from the papers are: the separate
economies of men and women within households; the variabil-
ity of the sexual division of labor in farm tasks; and
increasing numbers of households headed by women and the
concomitant increase in work burden due to male migration
and divorce. Concerning the sexual division of labor,
tasks may be the same or different for both sexes; in
female-headed households the so-called "male tasks" are
performed by women out of necessity. Concomitantly, the
authors report that researchers and extensionists have
failed to recognize women's roles in farming, ignored
gender in the design of FSR/E projects, and not included
women much as trial cooperators. Women, especially female
heads of households, are often low-resource farmers who may
have special problems that research and extension need to
address. (Indeed, there are male low-resource farmers who
have many of the same constraints.) The question as to
whether or not gender accounts for separate recommendation
domains finds different answers in the papers due to
differential ecologies, social organization, and cropping
Chapter 21 by Margaret Norem, Sandra Russo, Marie
Sambou, and Melanie Marlett provides an example from The
Gambia of how a women's component was formulated and
operationalized as part of a larger, FSR/E project.
Existing women's societies were used as a basis for organ-
izing women and a maize-cowpea intercropping package was
developed. The women experienced difficulties with the
package related to pests, seed varieties, and labor pat-
terns. The project was able to use women's participation
in the trials and the problems they encountered to argue
for the need to include women in subsequent research and
extension efforts.
Jeane Henn in chapter 22, examines how government
policies and environmental constraints impact on intra-
household dynamics in Cameroon. Labor patterns, proximity
to roads and urban areas, and farm gate prices affected
incomes, differently in two villages. Food sales were very
important to women's incomes but marginal to men's incomes
that were mostly derived from cash crops. In one village,
men withdrew labor from food crops resulting in an increase
in women's work. However, the women close to roads and

urban markets were able to increase their labor output and
foodstuffs produced, and double their income while women in
the other village were not.
Jean Due provides data in Chapter 23 from Tanzania,
Zambia, and Malawi on how gender is important to FSR/E work
arguing that unless there are persons) on the FSR/E team
who are sensitive to the issues, important information will
be missed. In Tanzania, a diagnostic survey for bean/-
cowpea research revealed that women select seed and contri-
bute more labor on the crop than men. FSR/E work in Zambia
would be hindered by not knowing the extensive labor con-
tribution of women on the one hand, and the extent of off-
farm and non-crop income, some of which is generated by
women, on the other. Data on extension agents contacts
with farmers in Tanzania and farmers' farm income are
correlated and show that male contact farmers have seven
times the income of female heads of households, who are
rarely contacted by extensionists.
In Chapter 24, Timothy Mtoi discusses labor patterns in
one region of Tanzania and uses a model to test the signi-
ficance of female labor on expected risk and productivity
under two cropping systems connected to a FSR project. The
analysis shows that farm income would increase if the new
alternative farming system had female labor transferred to
it. However, policy decisions affect whether or not women
can participate in the new technology (i.e., be trial
cooperators) and obtain extension advice on the packages.
..Chapters 25, 26, and 27 focus on Zambia and provide a
more detailed set on the farming systems there and on the
results of FSR/E diagnostic surveys. Mary Tembo and
Elizabeth Chola Phiri discuss the traditional chitemene
system of shifting agriculture and its sexual division of
labor, the results of the colonial period that drained off
male labor, and the lack of extension credit services to
women. The result of these conditions has affected the
diet and nutritional status of the population because women
have taken to growing cassava (a crop that is less labor
intensive, but also less nutritious than millet or maize),
and farmers neglect food crops for household consumption
because of growing cash crops.
Chapters by Robert Hudgens and Alistar Sutherland
examine FSR/E diagnostic survey work of Adaptive Research
Planning Teams that became sensitized to the need to target
women farmers in research activities and in the determina-
tion of recommendation domains. Hudgens details the char-
actaristics of male and female-headed households in terms
of land holdings, draft power, source of inputs, and cash

sales. The comparisons show that there are both similari-
ties and differences between the two household types and
that the female-headed households experience labor short-
ages. Women tend to be isolated from government services
and their production is constrained by lack of exposure to
new ideas, inputs, and capital.
As part of the diagnostic phase of FSR work, Sutherland
compares women's and men's roles in three regions of Zambia
and argues that even within one country, gender roles are
influenced by cultural, economic, political, and ecological
factors. Labor, cash availability, and draft power tend to
divide households into recommendation domains. Gender is a
distinguishing factor in one region, but not in the others.
In Chapter 28, Anita Spring reports on two different
on-farm, farmer-managed trials in Malawi. In the first,
the inclusion of low-resource female farmers along with
high-resource male farmers generated two recommendation
domains. Improved maize cultivars and use of fertilizer
worked well in the better environments and with high
resource farmers, but would be disastrous in the low-
resource environments and with low-resource farmers. On
the other hand, the traditional cultivar was better in -
these situations. The second set of trials involved all
female cooperators to solve a technical problem of inocul-\
ating soybeans as well as the issue of whether or not male\
extensionists could work with female farmers. It was found
that the women could do trials with precision and that male
extension and research staff could work with women farmers
in terms of training and credit programs.
Kofi Owusu-Bempah, in the final paper, argues for
inclusion of farmers in the planning and design of
projects, and particularly calls for the involvement of
women in the selection of species to be included in
agroforestry projects in Ghana. His work represents a
largely private sector effort and, like Rocheleau, calls
for expanding the framework of analysis to include the
landscape perspective and the intersection of crops,
livestock and forest enterprises.
In conclusion, the papers in this volume contribute to
an understanding of how gender affects farming systems and
the way that FSR/E operates. The papers demonstrate that
by linking the two codes gender and FSR/E the agri-
cultural research and extension system can become more
efficient and effective in dealing with different groups of
farmers. The papers provide details of specific cases and
the methods used to incorporate gender perspectives and

There is no single recipe for action. Instead, these
studies from an array of ecological, social, and political
contexts demonstrate that it is both possible and practical
to use gender analysis as a tool in the work of
agricultural development.


Boserup, E.
1970 Women's Role in Economic Development. New York:
St. Martin's Press.
Guyer, J.
1981 Household and Community in African Studies.
African Studies Review 24: 2/3.
Overholt, C., M. Anderson, K. Cloud, and J. Austin
1985 Gender Roles in Development Projects: A
Casebook. W. Hartford, CN: Kumarian Press.
Palmer, I.
1985 The Impact of Male Out-Migration on Women in
Farming. W. Hartford, CN: Kumarian Press.
Schmink, M.
1984 Household Economic Strategies: Review and
Research Agenda. Latin American Research Review

Integration of Intra-Household
Dynamics into Farming Systems
Research and Extension:
A Survey of Existing Projects

Rosalie Huisinga Norem

This paper reports the results of a survey designed to
study farming systems projects that include an intra-
household focus in data collection, design, or implemen-
tation. Farming systems models (Shaner et al. 1982) have
recognized the importance of the household as a component
of the farming system, but until recently little has been
done to systematically "open the black box" of the house-
hold component in those systems models. Projects respon-
ding to the survey being reported here are among those
attempting to gain a more systematic understanding of the
inter and intra-household factors influencing farming
The primary purposes of the present survey are to
assess the types of information collected and used by
projects, to evaluate the methods used for obtaining the
information, and to gain some insight into how and why the
intra-household information is helpful. In addition,
project managers were asked to identity types of infor-
mation they would like to have, but are not available and
the constraints affecting various phases of their projects.


This study evolved from on-going work relating house-
hold concerns to farming systems work. When the Farming
Systems Support Project (University of Florida/USAID) was
first initiated, a family task force was organized to focus
on the integration of household and family concerns into
the farming systems perspective. One of the recommenda-
tions of this group was to develop case studies and
training materials that would promote such integration. In
a position paper on "Intra-household Dynamics in Farming
Systems Research: The Basis of Whole Farm Monitoring of

Farming Systems Research and Extension," Cornelia
Butler-Flora (1984) set the stage for an intra-household
dynamics and farming systems case studies project that was
subsequently funded and implemented.
Concurrent with the effort to develop training materi-
als that will sensitize people to the importance of intra-
household factors in farming systems work as well as in
other development efforts, there is a need for more know-
ledge about the kinds of data being collected and the
methods used by existing projects that are attempting to
focus on intra-household factors. As an attempt is made to
understand the complexity of household dynamics, ways to
obtain and analyze information within reasonable time and
other resource limits also must be found. Questions about
how much information on aspects of the household should be
obtained from whom have yet to be answered (Norem 1983).
This is not to suggest that one "right" way of focusing
on household dynamics and farming systems can be identi-
fied. Rather it is to suggest that by examining what is
being done and how effective researchers and practitioners
involved find current efforts, some guidelines can be iden-
tified that will be helpful in future planning.
It may be helpful to consider how information relating
to households can be broken into different units depending
on the purposes of the individual study. Overall, the unit
of interest in intra-household dynamics is the household.
The unit of data collection might be one or more household
members, other informants, or other existing information.
The unit of analysis can be an individual, the household or
subsystems thereof, a work group, or the farming system,
among other possibilities. Designing parsimonious data
collection and analysis procedures requires an under-
standing of how these units relate in various situations.
For example, it may be possible to obtain good data on
the unit of interest from only one person if what is
required is basic demographic information such as age,
gender, and education of household members. The household
is also the unit of analysis in this example. Information
about the tasks performed in the household is more likely
to require data collection from more than one person, or
extensive observation or record keeping to permit the
collection of enough information to focus on the household
as a unit of analysis. As a clearer picture of the state
of the art develops, it is hoped a clearer set of guide-
lines will evolve. The survey summary presented in this
paper is a first step.


The Farming Systems Support Project and Population
Council Intra-household Dynamics and Farming Systems Case
Studies Project was initiated with support from USAID and
the Ford Foundation in late 1984. In February of 1985, a
request for expressions of interest was sent to projects
and individuals on a variety of international mailing
lists. Over seventy-five expressions of interest were
received in response to the request. These expressions of
interest were used to develop initial lists of types of
data and data collection methods being used in projects.
The lists were used in conjuction with the case studies
project conceptual framework to draft a survey question-
naire. The questionnaire was reviewed by the case studies
project advisory committee and revised using the commit-
tee's suggestion.
The questionnaire was mailed to all projects that had
responded to the original request for expressions of
interest in the case study project, since those projects
had self-selected themselves in terms of interest in intra-
household concerns. A few other projects were also inclu-
ded in the survey. All questionnaires received (n=19) are
included in the summary, regardless of type of project.
Most are farming systems oriented, with one project being
specifically focused on women in farming systems. Seven
projects from Asia, seven from Africa, two from the Middle
East, and three from Latin America are included. The
titles and identifying information about the projects are
presented in Table 2.1.
Each project has a different specific target group, but
all projects have target groups of farms with multiple crop
systems. Fifteen projects report farms in their project
also have livestock, most for multiple use, including cash
income, food, traction, wealth, and prestige. The average
land holdings for farmers in the projects ranged from .89
hectares to 30 hectares, with an overall mean of 6.34


Types of Intra-household Data

Each project was asked to indicate whether or not it
has data about five general categories: (1) demographic
information, (2) household members' participation in
activities, (3) household members' access to production
resources, (4) household members' participation in decision
making, and (5) income and expenditure data, benefits from














University of Hawaii with University of
N. Carolina & Center for Soils Research (USAID)

Ministry of Agriculture and Food and
the Virginia State University (USAID)

Winrock International & Ministry of Agriculture

Bangladesh Agricultural University
(Bangladesh Agricultural Research Council)

Haryana Agricultural University (Haryana)

Silliman University Research Center (Ford

Cornell University, Ministry of Agriculture &
Food, & the Visayas State College of Agriculture

Soil Management CRSP

Farming Systems Development
Project Eastern Visayas
Non-Farm & Resource
Management Institute

Agricultural Research &
Production Project
Farming System
Development Division

Women in Farming Systems

Role of Farm Women in Decision
Making Related to Farm Business

Balinsasayao Agroforestry


Farming Systems Development
Project Eastern Visayas
Non Farm & Resource
Management Institute


Burkina Faso

Burkina Faso

Sierra Leone




Sierre Leone

Fulbe Agropastoral Production
in Southern Burkina Faso -
for USAID Agric. Sector Grant

Income & Agric. Investment in
a Bobo Village

Adaptive Crop Research
of Sierre Leone


Agricultural Technology
Improvement Project

Dryland Farming Research

Phase I: Land Resources
Survey Project
Phase II: Strengthening
Land & Water Division


Frederick Sowers, Univ. of California, Berkeley

University of Illinois, (Wenner Gren Foundation, NSF)

Southern Ill. University, Louisiana State University &
Ministry of Agriculture & Natural Resources (USAID,
and Government of Sierra Leon)

Bureau of Integrated Rural Development (BIRD, REDECASH)

Midwest Int'l Agricultural Consortium (MIAC) (USAID,
Kansas State University)

Ministry of Agric. National Dryland Farming Research
Station, (FAO/UNDP, Kenya Gov't. )

FAO/Ministry of Agriculture (UNDP)

TABLE 2.1 (continued)








Irrigation Innovation and Family
Farming Strategies in Israel

Syrian Households: Women's Labor
& Impact of Technologies

Livestock Production Systems
in Central State of Veracruz

Honduras Agricultural
Research Project

EMPARAD: Socio-economic
(Agricultural Development Bank)
Case Study


City Univ. of New York, (Faculty Research Grant)

Andree Rassam, (ICARDA & MEAWARDS)

Centro de Investigacion Ensenanza en Granaderia
Tropical, CIEEGT, Facultad de Medicina & Zootechnia
(Universidad Nacional Autonomo de Mexico, UNAM)

Consortium for International Development, New Mexico
State Univ. (USAID)

Miriam Reyes, Martinez Elba, Rosa Hernandes

farm production, food consumption and nutrition informa-
tion. Each of these categories include several specific
kinds of data. Table 2.2 shows the types of data collected
and the number of projects that collected each type.
Table 2.3 summarizes the ways each type of information was
or is used by the projects that responded to the survey.



Type of Information (N=19) Number of projects

Household Structure, Membership, and Size 19
Education 17
Ethnic Identity 16
Migration Patterns 9
Variation in Household Structure Over the Life Cycle 7
Household Members' Participation in Activities
Cash Crops by Crop 13
Subsistence Crops by Crop 12
Livestock Production 11
Other Primary Income Generating Activities 8
Major Tasks of Household Reproduction 11
Household Member's Access to Production Resources
In General 13
By Tenure Category 9
By Production Potential (Irrigated, Non-Irrigated) 8
Family 12
Hired 15
Exchanged 11
Seeds 14
Tools 15
Equipment 15
Animals 14


TABLE 2.2 (cont'd.)

Type of Information (N=19) Number of projects

Household Members' Access to Production Resources
Innovations or Improved Production Inputs
Information (Extension Contacts, Training, Etc.) 14
Technology Inputs Requiring Cash or Credit 10
Informal 12
Formal 12
Other 1
Household Member's Participation in Decision-Making
Land Use 13
Use of Family Labor 13
Use of Hired Labor 11
Use of Exchange Labor 9
Use of Technology Inputs 14
Use of Credit 12
Cropping Choices 13
Cultivation Practices 13
Uses of Harvested Crop and Residue 13
Marketing 12
Income and Expenditure Data
Each Household Member's Sources of Income 7
Each Household Member's Expenditures 7
Benefits from Farm Production
Use of End Products From Crop Production 11
Desirable Characteristics of Each Crop
or Crop Product 8
Each Household Member's Access to or
Control of End Products 6
Food Consumption and Nutrition Information
Diet Survey 4-
Nutritional Adequacy Analysis 4
Food Preparation Practices 6
Food Preferences 6
On-Farm Household Food Production 7



Number of Projects Reporting Use of
Information by Type of Information*

Use 1 2 3 4 5 6

Initial Project Design 11 5 5 3 3 -
Selection of a Target Group 11 4 8 3 4 -
Identification of
Recommendation Domains 4 5 5 1 4 -
Choice of Research Topic 7 6 8 3 6 -
Designing Trials 6 4 6 3 4 -
Selection of Participating
Farmers for Field Trials 6 5 6 2 4 3
Evaluation of Field Trials 2 3 6 2 4 2
Redesign of Trials 3 6 6 6 4 2
Technology Recommendations 8 5 5 4 6 2
Extension Efforts 7 4 5 6 4 -
Project Evaluation Design 5 2 4 1 3 1
Assessing Time and Labor
Constraints 13 11 7 8 1 1
Assessing Opportunity Costs
for Innovation 7 5 5 3 3 -

Note: *Type of Information:

1 Demographic Information
2 Household Members' Participation in Activities
3 Household Members' Access to Production Resources
4 Household Members' Participation in Decision-Making
5 Income and Expenditure Data, Benefits from Farm
Production, Food Consumption, and Nutrition
6 Other kinds of information collected include:
religious affiliation, inheritance data, and
information gathered from husband and wife

Demographic Information

The most frequently used methods for obtaining demo-
graphic information are pre-existing national surveys,
formal project surveys, participant observation, and son-
deos (rapid reconnaissance surveys). This information is
summarized in Table 2.4 for all types of data. Demographic
information is also available through pre-existing anthro-
pological studies and local village records for some
projects. Other projects collected information through
farmer records, community informants, time allocation
studies, team members' personal knowledge, and indepth case
Nine projects collected demographic data before the
project began, five during the diagnosis stage, and seven
parallel with on-farm testing. Ten projects collected
demographic data on an on-going basis.
All projects collected demographic data on household
structure, membership, and size. Most also have informa-
tion about education and ethnic identity. Nine projects
have data on migration patterns and seven have them on
variation in household structure over the life cycle.
Table 2.3 shows that demographic information was most
useful in the early planning stages of projects.
Respondents were asked to identify the most helpful
information for each type and to give an explanation of how
or why the information was helpful to their project. They
were also asked to indicate any information in each cate-
gory that they did not have but wished they did. These
open-ended questions provide more detail related to intra-
household concerns than the tabulated results shown in the
The specific demographic information identified as most
helpful to a project varies by project, however, some
generalizations can be made. Gender and age structure of
the household is mentioned by several respondents,
sometimes alone and sometimes in conjunction with other
information such as labor and income. This information was
useful in identifying target groups and in designing trials
that ascertain labor bottlenecks and total household
activity patterns. Household structure is also reported as
an important consideration in designing extension efforts.
Ethnic information is mentioned second most frequently as
the most helpful demographic information, because farming
practices and values about female participation vary
according to ethnicity.



Data Clo11~r~tinn Mpthrvi

Demographic Information

Household Members'
in Activities


Household Members' Access to
Production Resources

Household Members' Participation
in Decision-Making

Income and Expenditure Data,
Benefits from Farm Production,
Food Consumption, and Nutrition

National Surveys
Formal Survey
Participant Observation

Formal Survey
Participant Observation
Community Informants

National Surveys
Formal Surveys
Participant Observation

Formal Survey
Team Member's Personal
Knowledge; Participant

Formal Surveys
Participant Observation

The two kinds of demographic information least often
available, migration patterns and variation in household
structure over the life cycle, are the ones most frequently
named in response to the question, "Are there demographic
data you do not have that you wish you had collected?."

Household Members' Participation in Activities

Formal surveys, participant observation, and community
informants are the methods most frequently used to obtain
information about the participation of household members in
various activities (see Table 2.4). As with demographic
information, projects obtained this information in a vari-
ety of ways. Each of the methods listed above for demo-
graphic information is used by at least one project to

Tvye of Data

Tv~~~~~e_ ofDt at olcto e

obtain activity data, group meetings provide an additional
source of information about household members' activities.
Three projects collected activity data before the pro-
ject began, six during initial diagnosis, and six parallel
with on-farm trials. Nine projects collect activity data
on an on-going basis.
Specific questions were asked about type of activity
data collected and method of disaggregation. Twelve
projects collected task assignment data disaggregated by
gender and age. Nine projects disaggregate by position in
the household as well. Five projects have information
about time allocation.
Thirteen projects report collecting some information
about the participation of household members in various
activities. Most frequently collected information (n=13)
is about activities related to production of cash crops,
with subsistence crops and livestock production information
available for twelve out of nineteen projects. Other acti-
vities within the household receive less attention as indi-
cated in Table 2.2. Table 2.3 indicated that projects use
activity data less often than demographic data. Activity
data are used most often to assess time and labor
Information on household members' activities was most
helpful in designing research and targeting interventions.
Respondents would like more detailed information about non-
production activities and several express a desire for
activity data covering a period of time up to a year. The
complexity of activity data is pointed out and difficulties
with processing such data are mentioned.

Household Members' Access to Production Resources

This study breaks production resources into sub-
categories of land, labor, capital, innovations, and
credit. The projects represented use a variety of methods
to obtain resource information; the most frequent are
pre-existing national surveys, project-conducted formal
surveys, participant observation, and team members'
personal knowledge.
Six projects collected data on access to resources
before the project began, seven during initial diagnosis,
and seven during on-farm trials. Five projects collect
these data on an on-going basis.
As indicated in Table 2.2, fourteen of the seventeen
projects have some resource information. But examination
of Table 2.3 suggests that fewer projects overall use this
information than use demographic data. More projects use
resource information than activity data for most purposes,

but more projects use activity data to assess time and
labor constraints.
The answers to questions about the most useful resource
data and why and how it is useful indicate that land
resource information is perceived as most helpful for more
projects than other kinds of resource data. The responses
also indicate that the usefulness of resource access data
is very project specific. Data on access to resources is
likely to be helpful in research design and in selection of
field trial locations. There is a pattern among responses
about the kind of resource information respondents would
like to have had but that was not available. More infor-
mation about monetary income, including gifts and remit-
tances, is mentioned in several contexts, including credit,
opportunity costs for innovations, and access to capital.

Household Members' Participation in Decision-Making

Fourteen projects in the survey have some data about
decision-making within households. These data are col-
lected most frequently through formal surveys, team mem-
bers' personal knowledge, and participant observation.
Other methods are used, but in a project-specific manner.
Only three projects report having decision-making data to
use in initial project design. One project collected data
during the initial diagnosis and four collected data
parallel with on-farm trials. Six projects are continuing
to collect decision-making data.
Table 2.2 indicates that projects which have decision-
making data also have information about most of the sub-
categories identified, including land use, labor use,
technology use, cropping and cultivation practices, and use
of production outputs. Table 2.3 suggests that projects
are not using decision-making data extensively. Eight
projects use decision-making data to assess time and labor
constraints, and this is the most frequent use reported.
Responses to open-ended questions about the usefulness
of decision-making data express the theme that a better
understanding of household dynamics permits more know-
ledgeable identification of target groups. Eight
respondents indicate their projects could use more detailed
decisionmaking data that would reveal more about the effect
of decision patterns.

Income and Expenditure Data, Benefits From Production, and
Food Consumption and Nutrition Information

Seven projects have information about these categories
of data. Formal surveys and participant observation are

the most common methods of obtaining the information.
There are some differences among the categories. Parti-
cipant observation is most likely to be the source of
information about food consumption and nutrition informa-
tion, and is not as likely to be a source of production
benefits data. Formal surveys are used by several projects
for all three categories.
Only one project had data in these categories before
the project began. Three projects collected the data
during the initial diagnosis, four collected the data
parallel with on-farm testing, and eight collect the
information on an on-going basis.
Table 2.2 details that eleven projects have information
in at least one of these three categories. Income and
expenditure data are least frequently available. Table 2.3
shows a fairly equal distribution of the use of specific
kinds of data available in these categories over the vari-
ous phases of the projects, especially in the design and
implementation of field trials.
Since this section of the survey combines three
categories, the answers to questions about which infor-
mation was most helpful, why so, and how, are somewhat
complex, but they also point out the need to integrate
information about overall production and consumption
patterns in the household. For example, respondents noted
the importance of looking at off-farm income and cash
income from food crops and of understanding the households'
reliance on local markets both for food and income. They
also mention the need to assess the opportunity costs of
innovations based on total inputs and total income
generating possibilities.
Respondents answering the question about information
they wish they had collected, mention primarily better
income data for individual household members monitored over
time. Several note the difficulty of obtaining reliable
income data and state the importance of finding better ways
of obtaining such information.

Most Effective Methodologies

Respondents were asked to select the study or activity
of their project that was most effective in collecting
information about intra-/inter-household variables relevant
to farm production and the ones that were most useful in
determining project decisions concerning research priori-
ties, cooperating farmers, technology acceptance, etc.
Nine respondents name the formal survey as most helpful.
This is usually done at the beginning of the project.
Eight respondents identify participant observation as the

most useful activity for obtaining household information.
This tends to be on-going. Four respondents state that the
sondeo as most useful. The sondeo took place anywhere from
the beginning to the third year of the project. Ten
respondents report the head of household as the primary
informant, whether male or female. Seven projects tried to
include at least one other adult household member. Three
relied on whoever was at home with a preference for the
head of household. One case study involved all members of
the household.

Constraints to Projects

Respondents were asked to identify constraints that
affected the study design, sample selection, conduct of the
study, data analysis, or applications of the data to their
project or activity. These responses are summarized in
Table 2.5. Ten respondents report physical, logistical or
resource contraints on sample selection for their projects.
Detailed answers to open-ended questions about these con-
straints show the most common one is transportation, either
because of the lack of means or because of the difficulties
of the terrain.



Number of Projects Reporting
Physical/ Cultural/
Logistical/ Social
Phase of Project (N=19) Resource Political

Study Design 9 5
Sample Selection 11 4
Conduct of Study/Activity 7 6
Data Analysis 7 1
Application of Data to
Project/Activity 3 1

In order of descending frequency, other constraints
mentioned are funds, language, personnel, political situa-
tions, and ethnic group considerations. In many instances
the constraints are named in conjunction with one another,
such as ethnic concerns and language difficulties.


Several summary points can be made from the information
provided in the survey. First, there is a wide variation
in the kind of data being collected about households with a
common focus on the household as a unit of interest. The
data are most often collected from heads of household, so
for some kinds of data there may be difficulty in using the
household as a unit of analysis. For example, data about
decision-making describe a dynamic intra-household process,
but data involving several household members require com-
plex data collection procedures. It is important to exam-
ine alternatives in determining which information is
important for which stage of a project and how to obtain it
as efficiently as possible. One respondent pointed out
that designing more standardized methods of data collection
and analysis is difficult because of the unique aspects of
each project, but the same respondent also emphasized that
efforts to move in this direction will save significant
resources and hopefully eliminate the need for each future
project to make the same mistakes.
There was general agreement among respondents that more
information about intra-household concerns was important.
One project is collecting additional information in most of
the categories in the final stages of the project with the
hope that it can be used in future planning.
This survey gives us one measure of what projects are
currently doing and assesses perceptions of the importance
of intra-household data. The next step is to suggest a
household minimum data set that projects could collect.
This would not eliminate the need for specific data for
specific projects, but would provide the potential for
comparative analysis and the beginning of theoretical



Flora, C. B.
1984 Intra-Household Dynamics in Farming Systems
Research: The Basis of Whole Farm Monitoring of
Farming Systems Research and Extension. A Position
Paper. Manhattan, KS: Department of Sociology,
Kansas State University.
Norem, R. H.
1983 The Integration of a Family Systems Perspective
into Farming Systems Projects. Conference
Proceedings, Family Systems and Farming Systems
Conference. Blacksburg, VA: Virginia Polytechnic
Institute and State University.
Shaner, W. W., P. F. Philipp and W.R. Schmehl
1982 Farming Systems Research and Development:
Guidelines for Developing Countries. Boulder, CO:
Westview Press.

Gender Relations and Technological Change:
The Need for an Integrative Framework
of Analysis
Alison Evans


Over the past decade, agricultural research has focused
on the rural poor and the need to strengthen the production
systems of small farms. In sub-Saharan Africa a substan-
tial proportion of agricultural activity and food produc-
tion continues to be organized at the household-farm level,
and with concern increasing over the region's food pro-
ducing capabilities, researchers and policy makers have
sought to identify ways of raising the productivity and
incomes of self-provisioning farming households (Henn 1983;
World Food Council 1982).
Farming systems in sub-Saharan Africa are extremely
diverse, but a number of shared conditions and character-
istics permit some generalizations. First, household mem-
bers typically comprise the main source of on-farm labor,
but households also require large amounts of labor for non-
farm and reproductive tasks. Second, rural women, espe-
cially poor rural women, make a substantial contribution to
all types of agricultural activity, and in most cases have
responsibility for the production, preparation, and distri-
bution of food for the household and for sale, and for the
reproduction of family labor (FAO 1984). Thus, if agricul-
tural research and policy is to have a positive impact on
food production and security in sub-Saharan Africa,
researchers and policymakers must link new technology and
extension advice more directly to the needs and priorities
of smallholder cers general and omen producesigeneraers in
The record of technological intervention in sub-Saharan
Africa has been very variable and the literature contains
many examples of the contradictory and unintended effects
of technology change for traditional farming households
(Dey 1984; Bryceson 1984; Agarwal 1981). It is always the
poorest and smallest farming households that are most

vulnerable to the "unplanned" effects of "modernization,"
but it is the effectiveness and well-being of women
producers in poor households that appears to be most at
risk (Agarwal 1981; Kisseka 1984; Whitehead 1981).
The fact that technological changes have complex, dif-
ferential, or even perverse effects on the productivity of
small farm systems indicates that agricultural research and
extension must be reoriented to identify and deliver solu-
tions to on-farm problems that are reliable and predictable
within the whole farm system and above all consistent with
the needs and interests of both women and men (CGIAR 1985;
Hahn 1985; Fresco 1985; Maxwell 1984).


The driving force behind recent research and develop-
ment in small farm agriculture has been Farming Systems
Research and Extension (FSR/E). Although FSR/E does not
follow one model it does tend to pursue a single aim: to
develop technology that helps small farmers improve output
(yield) and increase marketed food supplies while pro-
tecting the natural environment (FSSP 1984). A major
advantage distinguishing FSR/E from earlier research and
development approaches is its potential for integrating
comprehensive, interdisciplinary knowledge about small
farmers' technical needs with the wider systemic impli-
cations of technological change (Shaner et al. 1982).
The FSR/E approach tries to identify the distinctive
components and parameters of the "total" farm system in
which a small farm-household functions. In most FSR/E
models, the family labor unit or household is at the center
of the conceptual framework and the farming system is
defined as:

a unique and reasonably stable arrangement of farming
enterprises that the household manages according to
well defined practices in response to the physical,
biological and social environments and in accordance
with the household's goals, preferences and resources
(Shaner et al. 1982).

In their functionalist approach to farm systems analy-
sis, FSR/E practitioners try to understand how factors
-under the control of household members (endogenou-factors)
like family labor and the techniques fipr5ductioninter-
act withphysical,_biological,-and socioeconomic factors
beyond household-control (exogenous factors) like climate,
prices, and the availability of credit, affect the agricul-
tural welfare of the farm household. By combining

knowledge of the whole farm system with extensive on-farm
research and experimentation, systems researchers should be
able to recommend technological alternatives that meet the
priority needs and preferences of small farmers and evalu-
ate which technologies fit where and why, and monitor who
In practice, the FSR/E approach is used primarily to
reduce the degree of error in finding the appropriate tech-
nical "fix" for market-based problems of agricultural pro-
duction. Thus, FSR/E practitioners have restricted them-
selves to finding technological solutions to a limited
range of crop and livestock production problems, using
mainly market criteria to measure the costs and benefits to
farm-households. The methodology used to identify farming
households' needs has been implicitly restricted and the
choice of recommendations has been limited to suit a nar-
rowly defined, technocratic approach to on-farm develop-
ment. FSR/E practitioners have not yet built a broad
understanding of the diverse needs and preferences of small
farms or of the differential effects of technological
change for members of farm-households. The FSR/E approach
is particularly limiting when considering gender issues in
agricultural development. In particular, FSR/E is ill
equipped to:

(1) understand the distribution of costs and benefits of
alternative technological choices and extension pro-
jects between men and women within farm households and
under what conditions technological change benefits men
but not women, or in fact harms women, individually or
as a group or
(2) identify and meet the technical needs of women as
farmers, particularly as producers and consumers of

FSR/E has not accomplished what it set out to achieve
at the-socio-economic level. Researchers have failed to-
consider how farming systems are constituted along gender
lines-and to analyze the influence that gender differentia-
tion-has in shaping technological and economic -utc6;oime-sin
agricultural development. The methodology contains proce-
dures and assumptions that have tended to reinforce gender
First, FSR/E uses markereto monetary_criteria to ideD-
tify productive components in the farm_system and to
measure their contribution to household-incomeand welfare.
May-of the value-adding or service activities that are not
directly comparable in monetary terms are left out or are
considered costless in the economic equation. For example,

FSR/E does not value non-monetized, household-based activi-
ties such as food production (especially secondary foods),
processing and preparation for direct consumption, fuel and
water collection, and house maintenance and repair.
From micro-level research, it is shown that preferences
or market criteria have important gender implications
because while men, women, and children all perform some of
these non-market tasks, it is women who contribute most of
the labor, energy, and resources (as well as exchanging
labor and resources with other households) to get this work
done. By treating non-market, home-based work as
complementary to the farm unit rather than an essential
component of it (Fernandez 1985; Lele 1985), production
economists have neglected important links between market
and non-market enterprises. For example, the linkages
ignored include those between water and fuel collection,
that in sub-Saharan Africa is almost always non-monetized,
with crop and livestock production, food production, and
food processing. More importantly, economists have
neglected the significance of gender differentiation, the
division of labor, and resources by gender in shaping
relations between food and non-food production, market and
household-based enterprises, and consequent responses to
economic incentives and new technology.
Second, production-economists and technicians have
concentrated on ways of improving-yields and the volume of
marketed output to increase household income and welfare.
In this process they tend to focus on specific sub-systems
like the cropping system, and within that, particular crops
with obvious market potential. This selective approach has
often been pursued without carefulconsideration opfthe
possibility that production constraints stem from other
J( parts of the farming system or that farmer decisiTi-making
has to do with the 'operationhof the farminqsystem as a
whole ahd not with isolated-parts of it.
This implies that planned changes within the cropping
i|system, for example, that should increase aggregate output
pjand income, may have unanticipated and differential effects
jon individual household members' production. The welfare
effects of planned changes can be predicted only if the
full extent of household members' involvement in the farm-
ing system is known, and more importantly, if it is known
how the activities of men, women, and children fit together
within the prevailing technical and social relations of
Systems researchers have often overlooked appropriate
indirect or non-farm solutions to farm-based problems. For
instance, farm output may increase if the productivity of
household-based activities, many of which are women's

responsibility, were to improve. Higher productivity in
these tasks may have a positive impact on the amount or
quality of labor available for market-oriented production.
It has been demonstrated empirically that women's
household-based activities will feed directly into the
satisfaction of household basic needs whereas the link
between market-based activities and income and household
welfare is more spurious (Hart and Popkin 1975). If women
can collect water or-fuel more easily, they could expend
more time and energy in directly productive activities. Or
by raising the productivity of subsistence food crop pro-
duction and processing women could improve the health and
efficiency of labor, especially during the low energy or
"hungry" seasons.
Third, the FSR/E approach is based on a stereotyped
model of "the household" as a homogeneous unit of produc-
tion and consumption that is at variance with the reality
in many rural areas.
Furthermore, it is assumed that the distributional
effect of technological change on farm household members
can be read off from shifts in the demand curve for family
labor within a specific sub-system. Technical_change,
however, does not effect the division of labor and /
resources in any ystema ic way (White~ead ~81T _The
relations between women aid~men in farming households are
not solidified_within the technical division of labor nor
are their.interests and needs necessarily complementary.
Households are not.homogeneous units and the needs and
preferences of.women and men are not easily defined with
respectto a single household production function. Even
more importantly, the household decision-making process can
be only partially explained by economic factors. The
choices of men and women in the household and beyond are
the outcome of an intricate bargaining process mediated by
normative relations of power and control.
In many West African countries, men and women occupy
different positions within the farm economy, not only in
terms of task allocation, but also in terms of their dif-
ferent access to factors of production inputs, credit,
others' labor time, off-farm employment opportunities,
and markets. In polygynous households, senior wives also
have structurally different economic positions from junior
wives. In a number of decisions, men and women have a
joint economic interest; in many others, often those
associated with direct welfare activities, their interests
conflict or their economic and technical needs are
different. For example in Sierra Leone, women's interests
in improving swamp rice farming and intercropping to
produce a marketable surplus have been in conflict with

men's interests in upland farming and their use of work
gangs and household labor in the production of market
crops. Polygyny also compels wives to have independent
access to cash income that involves management of non-farm
income generating enterprises such as palm oil processing,
soap and pottery making, and dyeing cloth. These enter-
prises are kept separate from men's and require women to
maintain separate budgets and resources even when living
with men in the same extended household (Richards et al.


Most FSR/E models take a selective and comparatively
static view of the farming system. There is a need for
more searching concepts and methods to look at gender in
farm-household systems, to understand the gender
distribution of costs and benefits of technological
choices, and to identify women's technical needs.
Unfortunately, due to a lack of empirical evidence this
paper can only offer some analytical suggestions on how
such a framework might be built.

Conceptual Issues

First, a conceptual model that treats the small farming
unit as an interlocking system of market production,
subsistence, and reproductive activities performed at farm
and household levels should be developed. Reproductive
activities include primarily the domestic services
necessary for family survival. Central to this
inter-related model are the multi-activity household and
the different sets of economic and social conditions that
shape the economic participation and welfare of women, men,
and children. These sets of conditions are not immutable.
Depending on the total matrix of productive and reproduc-
tive activities that comprise any one farm-household
system, they are likely to alter with seasonal shifts, the
life cycle development of the household, and emerging
employment and market opportunities in the local economy.
Such a framework should be able to generate data on the
key sets of conditions that shape the differential partici-
pation of household members and give explicit attention to
the most important interactions between enterprises that
vary by gender.
The anthropological literature for sub-Saharan Africa
suggests that data needed center on intrahousehold resource
allocation and on the relative weights of market, non-
market, social, and economic forces in shaping the

differential distribution of women's and men's labor,
resources, management, and control within the total farm-
household system. Such information could be used to
investigate the possibility for productivity gains in non-
monetarized enterprises that have either a direct or an
indirect productive impact on household members. Agricul-
tural researchers and extension personnel also need such
information to understand why planned changes in the tech-
nological and economic balance of farming systems can have
unanticipated and sometimes detrimental effects on other
enterprises outside of the production arena.
In Africa, a commonly cited problem for women farmers,
particularly in the poorest households (Dey 1984; Guyer
1980; Whitehead 1981), is that technological changes have
intensified women's workloads without adequate compensation
or have eroded their access to land and non-farm income.
Without understanding exactly how women's multiple activi-
ties interlock with each other and with other enterprises
in the system, neither the aggregate net effect of techno-
logical change upon the total workload or returns to labor
of individual household members, nor the differential
productivity and welfare effects on women and men can be
established clearly.

Methodological Issues

A broader conceptual model raises a number of important
methodological considerations for farming systems research.
Recommendation domains. While agronomic, topographic,
and general socioeconomic data will continue to be crucial
in selecting recommendation domains, different household
forms, family structures and composition must be increas-
ingly incorporated. In Africa, it is essential to consider
polygynous households, consensual family structures,
female-headed households, and the complex variation in
authority and power structures that determine the location
of decision-making and access to resources in household
units (Allison 1985). At this stage it is important that
researchers be aware of the different economic and social
positions that women and men occupy in African households
and how these are manifested in different decision-making
processes, separate budget and production functions, and
different access to inputs and markets, incentives and off-
-farm employment opportunities.
Problem identification. Information about the differ-
ent positions of women and men in the farming household
must be linked to precise data about patterns of resource
allocation, labor utilization, and the multiple fit of
women's and men's responsibilities in order to identify

major constraints and bottlenecks as they vary by gender.
The degree of flexibility and substitutability of the
organization of farm-household activity must be taken into
account as well as the potential trade-offs or conflicts
that might emerge between the genders when a constraint
within one sub-system creates another elsewhere in the
system. Further research should focus on:

(1) the degree of flexibility and substitutability between
the role of women's labor in farm production and
household-based production relative to men's;
(2) the substitutability between women's labor and capital
(productivity enhancing technologies) relative to other
household members; and
(3) the flexibility and substitutability between women's
labor and that of other family or household members,
including husbands, children, neighbors, or hired labor
(Lele 1985).

Much of the information that can be gathered about
these points will be qualitative rather than quantitative,
but is nonetheless valuable. Researchers and extension
personnel should resist the temptation to devalue the
importance of qualitative data in planning on-farm research
and in evaluating responses by farm-households.
On-Farm Research, Extension, and Evaluation. Gender
differences must be an integral criterion woven into the
analysis of technological constraints and the recommenda-
tion of solutions and must also be considered when choosing
target farmers with whom to conduct on-farm trials. It is
particularly important that researchers recognize the mixed
strategies that women employ to meet basic household needs
and consider the potential impact that trade-offs or cur-
tailments within these strategies have for individual and
household welfare.
It is also important to trace the distribution of male
and female labor time and resources within total production
cycles. Thus, crop production from cultivation to proces-
sing, consumption, or sale should be disaggregated to
understand the possible welfare implications of changes
within the cycle. For example, changes in cropping patterns
may intensify women's unpaid labor during weeding,
harvesting, and processing stages and conflict with their
labor needs elsewhere in the farm-household system.
Despite creating a greater workload, a new crop variety may
offer women potential income gains through the processing
and sale of crop by-products over which they have marketing
control. The dominant effect cannot be established a

Finally, it is necessary to acknowledge that household
services play an economic role in maintaining the
productive function of the farming system and to identify
the non-pecuniary costs borne by individual household
members (usually women) in providing such low productivity
services under time-consuming and arduous conditions.

If gender issues are to be given explicit attention
within agricultural research and technology development,
then an alternative analyticalframework is required. Such
a framework should_identify. the different sets..of condi-
tions .that characterize the economic .participation.of-women
and menin small_ farm systems. and view- the ,integrated-
farming systems as a gendered system of production and
reproucion. The purpose of such a framework is not only
Eo examine existing evidence with greater rigor, but also
to generate specific hypotheses, data, and methods that
break down gender biases within agricultural research and
extension. A number of problems that must be addressed to
build an alternative framework include:

(1) Overcoming the methodological division between tech-
nocratic and social science research that challenges
the integration of quantitative and qualitative methods
and data;
(2) Resolving time conflicts, such as the time scale that
agricultural research institutes or donors demand for
FSR/E but that cannot accommodate the amount of time
that sociological or socioeconomic research usually
requires; and
(3) Channelling research funds and personnel into more ex-
tensive data collection and methodological approaches.
(Funding for scientific research into minor food crops
and technologies for non-farm activities is extremely

Despite these problems, to have a clear understanding
of the relationship between gender relations and the econo-
mics of farming systems and technological choice it is
important to question the conceptual basis of existing
methodological approaches in agricultural research and
development (see Figure 3.1) and to build alternative
frameworks for gathering empirical data and testing
hypotheses in which gender relations are an explicit

Extra H.H. Intra H.H.
Resources Resources

H.H Farm



Raw materials IonFoo 1
Management INPUTS I
eL nt water [ I--LIVESC cash ii

Male labor Fodder
jFemale labor _
Child labor (lab

repair wag
-Childcare non-
-Domestic chores ino
-Social obligations

Consumption goods Payments
(food, durables) (interest, tax)
and services. t
SEXPENDITURES (wages, cash, kind)

Human capital
investment & Credit
security (formal, informal)

Producer goods,
factors, investment
& savings.

KEY: ---- flows of family labor in/outside the F-H.H. system.
flows of non-labor inputs a outputs (& payments).

PLEASE NOTE: For clarity, the elements of the system have not been disag-
gregated along gender and age lines. These variations are captured when the
framework is applied to the situation of each family member in turn and then

Acknowlegement: Many of the ideas for this diagram come from C.D. Deere and
A. de Janvry "A Conceptual Framework for the Empirical Analysis of Peasants"
American Journal of Agricultural Economics. 1979.




Agarwal, B.
1981 Agricultural Modernization and Third World
Women: Pointers From the Literature: An Empirical
Analysis. WEP Research Working Paper WEP/10/WP21.
Geneva: International Labor Organization.
Allison, C.
1985 Women, Land, Labour and Survival: Getting Some
Basic Facts Right. IDS Bulletin 16:3:24-29.
Bryceson, D.
1984 Women and Technology in Developing Countries:
Women's Capabilities and Bargaining Positions.
Prepared for INSTRAW, United Nations.
1985 Women and Agricultural Technology. Bellagio,
Italy: Rockefeller Foundation and International
Service for National Agricultural Research.
Dey, J.
1984 Women in Food Production and Food Security in
Africa. Women in Agriculture 3. FAO: Rome.
1984 Women in Agricultural Production. Women in
Agriculture 1.
Fernandez, M.
1985 Social Components in Peasant Farming Systems: a
Research Proposal. Unpublished.
Flora, C. B.
1984 Intra-Household Dynamics in Farming Systems
Research: The Basis of Whole Farm Monitoring of
Farming Systems Research and Extension. A Position
Paper. Manhattan, KS: Department of Sociology,
Kansas State University.
Fresco, L.
1985 Food Security and Women: Implications for Agri-
cultural Research. Presented at the International
Workshop on Women's Role in Food Self-Sufficiency
and Food Security Strategies. Paris: ORSTOM/CIE,
1984 Newsletter 2:3.
Guyer, J.
1980 Female Farming and the Evolution of Food
Production Patterns amongst the Beti-of
South-Central Cameroon. Africa 50:4:341-356.

Hart, G. and B. Popkin
1985 A Note on the Interdependence between Economic
and Welfare Factors in Rural Filipino Households.
University of the Philippines, Institute of Economic
Development and Research. Pp. 75-5.
Henn, J.
1983 Feeding the Cities and Feeding the Peasants:
What Role for Africa's Women Farmers? World
Development 11:12:1043-1055.
Kisseka, N.
1984 Implications of Structural Changes for African
Women's Economic Participation. Unpublished.
Lele, Uma
1985 Women and Structural Transformation. Economic
Development and Cultural Change 34:2:192-221.
Maxwell, S.
1984 I. Farming Systems Research: Hitting a Moving
Target. II. The Social Scientist in Farming Systems
Research. Institute of Development Studies
Discussion Paper 199.
Richards, P., J. Karimu, et al.
1981 Upland and Swamp Land Rice Farming in Sierra
Leone: The Social Context of Technological Change.
Africa 51:2
Shaner, W.W., P.F. Philipp and W.R. Schmehl
1982 Farming Systems Research and Development
Consortium for International Development. Boulder,
CO: Westview Press.
Whitehead, A.
1981 A Conceptual Framework for the Analysis of the
Effects of Technological Change on Rural Women.
Technology and Employment Program Working Papers WEP
2-22/WP79. Geneva: International Labor
World Food Council
1982 National Strategies to Eradicate Hunger. New
York, NY: World Food Council, United Nations.

Problems of Understanding and
Communication at the Interface
of Knowledge Systems

Janice Jiggins

One of the fundamental justifications for the practice
of FSR/E is that its methodologies promote a desirably
close relationship between farmers, researchers, and exten-
sionists in the determination of research criteria and the
design and choice of interventions. Considerable effort is
going into persuading researchers and extensionists to
understand that:

The goals and motivations of farmers, which will
affect the degree and type of effort they will be
willing to devote to improving the productivity of
their farming systems, are essential inputs to the(
process of identifying or designing potentially
appropriate improved technologies (Norman et al.

Much less effort is being devoted to helping farmers to
understand the values, rationales, and objectives that lie
behind research and extension behaviors. As FSR/E becomes
more involved with on-farm trials, particularly farmer-
designed and managed ones, the more important it is for
FSR/E practitioners to find ways of articulating their own
rationality and making it accessible to farmers.
Some of the FSR/E practitioner rationality is conveyed
implicitly in the course of working closely with farmers.
A 1985 circular on the Agricultural Technology Improvement
Project (ATIP) program, distributed to local agricultural
staff in Mahalapaye, Botswana, notes:

In explaining these meetings, it is important the
farmers understand this is a new approach to research
in which we want to work together with them to discuss
and evaluate alternatives, rather than just rely on
collecting information from them (ATIP 1985).

But conversations with farmers in areas where FSR/E
teams have been working, reveal that farmers remain greatly
puzzled by such things as to why researchers insist that
fields or plots be measured in certain ways and what those
measurements mean. Or, what it is in the logic of the
researchers' world that makes them value, for example, cer-
tain livelihood activities such as field cropping above
other activities that seem to the farmers themselves
,equally necessary components of their livelihood. The
(breakdown of communication and understanding seems greater
Between women -- as farmers, food processors, traders and
consumers and male researchers. This is due not only to
sociocultural distances between them. Male researchers may
understand little of the rationality of the domestic domain
in their own worlds. The researchers' lack of an explicit
frame of reference in this sphere or use of a partial or
biased one -- influences their own set of mental constructs
by which they perceive and interpret the world of women
within farming systems. Communication difficulties thus
-are compounded.
There are a number of threads that might be disen-
tangled here. This paper will pull out and untwist only
one: the logic of flexibility within the domestic domain,
illustrated by examples from Lesotho and northern Zambia
(1). The data are not complete from the FSR/E point of
view, having been collected for other purposes, but they do
highlight a number of points that FSR/E theory and practice
needs to take into account. The paper concludes with
suggestions about how this might be done.
The data are drawn from two areas of acute seasonal
stress. In the Lesotho site, a longitudinal study of
energy flows suggests a bimodal pattern of stress (Huss-
Ashmore 1982) while the second area in northern Zambia has
a short period of moderately erratic, within-season rain-
fall and a long, dry, cool period of six to seven or even
seven and a half months, with the time of acute hunger
falling in the January-February period after weeks of heavy
labor and declining food stocks (author's unpublished field
notes, 1979-80).
Both are areas of high male out-migration and income
insecurity. Risk and loss minimization figure highly in
farming system strategies. Women, as household heads and
as farmers within male-headed enterprises, respond to
climatic and income uncertainties by trying to maintain
flexibility, typically in four spheres of the domestic

(1) in production, by maintaining reserve crops and
varieties in household gardens and in wild habitats;

(2) in the timing of operations, volume of product handled,
and technique used, in the spheres of food processing,
storage and food preparation;
(3) by altering the mix, timing, and quality of performance
of their multiple roles; and
(4) by manipulating whatever room there might be for
substitutabilty of labor and obligation between men and

For reasons of space, the following illustrations are taken
from only the first two spheres.


In the peculiarly distorted economic situation of
Lesotho, the day to day survival of rural households is
largely a matter of how women maintain themselves and their
children. The rationality of the farming system is not
determined by physical and climatic features these only
set limits to what is possible. It is determined by the
rationality of women's lives, that is centered in (though
by no means confined to) the domestic domain. Within that
domain, there is one resource which is critical: fuel
supply. It could be said to be the key both to cropping
choices and household food availability. Huss-Ashmore

Because fuel is essential for processing almost all
foods, it can be considered a critical resource for the
maintenance of health and nutritional status. In
Mokhotlong the type of fuel used and the time spent to
procure it, vary according to the seasonal availability
of dung (1982:156).

The preferred fuel is compacted dung, readily available
during the winter from the kraal close to the homestead,
that, when dried in uniform slabs, burn with the slow, even
heat necessary for the long cooking of dried grains and
pulses. In households without a kraal, women have to
purchase the dung or manipulate kTinrelations to obtain it.
When the cattle are moved in the summer to the high pas-
tures, women must use horse and cattle dung, picked up from
the fields and trails, that is less dense and takes more
time to gather. Both sorts are kindled with resinous,
woody shrubs that become scarcer as the summer passes but
may be the only source of summer fuel if insufficient dung
can be gathered. For a short period, kraal dung may be
kindled with maize cobs as they are threshed in the winter.

It is fuel availability rather than food availability that
determines which foods are eaten at different seasons:

The supply of slow-cooking protein sources is not used
equally throughout the year but is depleted during the
cold season when appropriate fuels are available.
During the summer the population relies heavily on wild
vegetable protein sources, which require more time to
locate and gather but which can be rapidly cooked
(Huss-Ashmore 1982:157).

One might think that these interactions and their
further entwining with the water/fuel/grain seasonalities
of sorghum beer-brewing, a source of income and wage work
that is important for being one of the few available to
women are not so terribly difficult for a researcher to
discover. However, single visits to households during an
exploratory survey may fail to discover seasonalities that
are both interdependent across disciplinary boundaries
(fuel-forestry; cropping; post-harvest domestic food tech-
nology) as well as across gender boundaries (cattle are
men's business). The fact that a critical key to the func-
tioning of the farming system operates within the domestic
domain may continue to conceal its significance during a
verification phase of FSR/E work as well.
A further difficulty arises when researchers try to
measure the quantities involved such as the cooking time of
various foodstuffs using different fuels. An anthropolo-
gical study of Sesotho measurement concepts points out:

A woman knows how long to cook vegetables because she
knows when they are ready. One woman, preparing
bread, was asked how she would cook it: Until it is
ready (bo butson). Pressed for precision, she thought
carefully and then said: Five or six hours (li-hora -
again the English word). In fact, she cooked the bread
for an hour and a quarter and saw that it was perfect
when she took it from the steam oven. It was ready
both in English terms and her own. At no time did she
refer to any kind of time, not even in the sun. It was
not the time that made it ready. It was the cooking
(Wallman 1965:240).

The researcher is concerned with the measurement of
time but the woman is concerned with the measurement of
"readiness" and there is no reason why the measurement pro-
cess should not begin with readiness (What does it look
like? Is it hard or stiff or does it run? etc.). Instru-
ments such as time allocation studies, useful as they are

as indicators of the range of activities and claims on
labor, make invisible whatever it is that women themselves
see themselves as allocating or conserving. In Huss-
Ashmore's case, women collect wild vegetable proteins not
because they are a preference food nor because there is
nothing else available, but because women primarily wish to
conserve fuel.
The difficulty does not lie in using the measurement
units that make sense within the rationality of the user.
Researchers and extensionists alike are trained in the con-
cepts of scientific agriculture and these concepts may have
no equivalents in the knowledge system of the woman while
the woman is trained in the concepts of her indigenous
knowledge system and may have no way of knowing the signi-
ficance of the concepts used by the researcher and exten-
sionist. This has little to do with any differences in
ethnic background of the actors and a great deal to do with
the difficulty of articulating the rationality of one sys-
tem in the terms of the rationality of another.


In the northern Zambia location, the fact that local
vegetables and fruits form an important part of the diet is
well-established and there are even a few research programs
investigating the more important species (MAWD 1983). What
is not so readily accepted is that these may have charac-
teristics that yield benefits not provided by modern vari-
eties of the main food crops, however abundant they might
be. Interventions that make their production more diffi-
cult by switching labor or land use, for example, may also
make the seasonal management of diets more difficult,
unless the market provides substitutes at affordable
The question of the timing of agricultural innovations
with respect to the role of market provision of those goods
and services previously supplied within domestic and local
economies merits a short digression here. In agriculture
in industrial countries and albeit to a lesser degree in
irrigated agriculture in developing countries, research
organizations work within and for production and knowledge
systems that are well-defined, well-organised, and highly
interactive. There are at least four main components:

(1) farmers, who are organised and able to contribute to
research programming through a variety of channels;
(2) powerful industrial organizations engaged in the
business of transforming primary production into a
range of consumer and industrial goods, who are well

able to signal to researchers their own technical
requirements or even, by paying for research, to
determine what crop characteristics meet the needs of
their own technical processes;
(3) powerful commercial organizations engaged in the busi-
ness of wholesaling and retailing produce and processed
foodstuffs, that are able to insist on high quality
standards in defence of existing and new sales; and
(4) consumers, who, either through their purchasing power
or through consumer organizations and lobbies, also
signal their preferences to researchers.

The case is quite different in most dryland areas in
developing countries. Except perhaps for the richest
farmers, producers are not organized nor politically
powerful and have few if any links with researchers. The
range of transformation processes occurs largely within the
domestic domain using local technologies. Wholesalers and
retailers operate in fragmented and often non-competitive
arenas in which the overall level of sales is depressed and
quality carries no premium. Finally, consumers have weak
purchasing power and few, if any, organized channels for
expressing their preferences.
If the inherent yield potential of many dryland areas
is judged to be low, with scant chance that the value of
the marketed output will ever pay for or induce the kind of
infrastructural developments witnessed in irrigated envi-
ronments, then presumably, it will be necessary to preserve
a continuing capacity to derive benefits from the goods and
services presently obtained from the biomass through trans-
formation within the local community and the household
economy. The challenge becomes that of raising capacity
without displacing too many of the benefits presently
obtained from within the micro-economy rather than in
raising capacity by concentrating on only a few benefits
(higher yield) and externalizing the provision of the rest.
Varietal characteristics must continue to an unknown extent
to meet the demands of domestic transformation processes,
technologies, and end uses. The following example des-
cribes just such a situation.
The local vegetables (fruits not included here for the
sake of simplicity) being produced on one farm during
February 1980 at Sambwa in Mpika District of Northern
Zambia include: four cassava varieties (masanga uko,
matutumushi, muntulunga, ucongo); three finger millet
varieties (mwaangwe, mutubia, mwambe); two varieties of
groundnuts; and local maize. Each as a very specific
place in seasonal production and food management. For
example, mwaangwe is a sweet, very early maturing finger

millet that provides one of the first new food crops in the
year and a sweet beer for working parties as the main
harvesting period approaches. Two of the cassava varieties
have palatable leaves (masanga uko and matutumushi the
latter much sought after by the wild pig) but these fall in
the cold season (June, July), so some are dried early in
the season for later consumption.
In addition to these main food crops, there were five
distinct production sites around the compound (and one fur-
ther away in the dambo or wide valley bottom) tended by the
two adult resident women, on which were grown a mix of wild
and semi-wild plants and "crops" promoted and officially
marketed by the government. A number of the wild plants,
such as busoshi (Sesamum alatum) and chimamba (Sphenostylis
erecta), also occur naturally (respectively on disturbed
soil and around anthills) but on this farm could be con-
sidered as true crops, for they were deliberately planted
on chosen sites, protected from chickens and weeds, fertil-
ized with household rubbish, and the product traded in
Mpika market. Their utility is partly a reflection of the
low and erratic yield of groundnuts but, they have a
utility as snack foods at a time of the year when women may
cook only once a day or even once in two days. The
perennial chimamba ensures that some kind of snack is
always going to be available. It would only lose its
utility if alternative snack foods were to become available
at the critical time of the year when women are busiest or
the preparation and cooking time of cassva and millet were
to become less or women's cultivation labor were to be
Another example is provided by the great care the women
took to maintain the balance between the availability of
the staples (millet, cassava) and the availability of oily
or slippery foods for the relish that are added to the
nsima or thick, coarse porridge. The nsima is almost
inedible in sufficiently large quantitleshwithout such a
relish to ease it down. In conditions of scarce and expen-
sive commercial cooking oils, unreliable groundnut harvests
in the face of erratic rains, and the time required for
shelling and pounding groundnuts, the softness and slipper-
iness of some local vegetables were highly desired charac-
teristics. Pupe (Fagara chalydea) is an important dry
season resource in this respect. Slippery local vegetables
have the additional advantage of needing no blanching or
treatment with potash when they are dried for preservation.
Both the men and the women had been experimenting with
vegetable production, as the following examples illustrate.

The male head of the compound had been trying white cab-
bage, tomatoes, and cucumbers from seeds supplied through
the Horticultural Marketing Board in Mpika, in a dambo
garden at the end of the rainy season. He found that the
cucumbers grew best but were the least needed for domestic
use as they already grew a satisfactory range of cucurbits.
The tomatoes were well-liked for their flavor and the soft-
ness of the flesh but were tiresome to eat because of their
tough skins, the very characteristic that made them suit-
able for the rough marketing conditions. The women had
been experimenting for many years with lubanga (Cleome
gynandra) selecting for larger leaf size without sacri-
ficing any of the tenderness. They reported, too, that
they could get a higher price in the local market for the
larger-leaf variety.
The men in the compound scored consistently lower than
the women on the following tests using vegetable sources:
identification by sight; recall of the main physical
descriptors and husbandry, processing techniques and length
of storage; and preparation for eating. Zambian and expa-
triate members of the nearby agricultural college who were
engaged in conducting and supervising trainee extension
workers in farm surveys, were asked to share their views of
the role of local vegetables in the farming system. They
all referred the question to the home economic staff, who
were trained to work with "western" vegetables, and with
few resources to work in the field, knew only those local
vegetables that had been used in the household in their own
home areas.
There are further problems of communicating knowledge
between distinct knowledge systems. The production sites
where the local vegetables were grown changed shape and
size as women took advantage of rainfall patterns in the
advancing season to make additional sowings. At the same
time, neither the market value of the product traded nor
the opportunity cost of female labor (based on market wage
rate) would seem adequate measures of the value of either
women's labor time or of the local vegetables to the far-
ming systems. The women themselves used a notion of con-
venience that appeared to be a combination of character-
istics such as: ease of growing near the house; availabil-
ity (fresh or dried) at moments critical from the point of
view of diet management; ease of processing and prepara-
tion; timing of labor inputs; and substitutability for
other crops. The notion encompassed the principle of
flexibility. In this respect, they were reluctant to
choose paramount characteristics either for any one crop or
between the range of crops. Instead, local vegetables were

viewed as a bundle of biomass that enabled them to manage
their resources and responsibilities to the best advantage.


There are a number of important "lessons" that can be
drawn from this brief review. To summarize, these are
reduced to two:

(1) the need to develop methodologies for establishing
the key field-household interactions at an early stag
of the diagnostic process, and
(2) the need to develop methodologies for mutual communi-
cation of key concepts across the boundaries of
researchers' and female producers' distinct knowledge,

Two techniques that might prove to be useful diagnostic
instruments for researchers of any background are situation
analysis based on the critical incident technique and peer
group workshops.
The former is widely used in diagnostic sessions
between researchers and carefully drawn panels of users in
industrial and commercial practice. It involves informal
but structured interviewing which, as users identify prob-
lem areas and describe the boundary conditions, focuses on
a "critical incident" that exemplifies one of the problems.
The incident is then analyzed in depth, leading into
discussion of desirable ways to deal with it. Each of the
problems is similar treated in turn.
Peer group workshops are widely used throughout the
ESCAP region in the development of local, self-managing
groups and income-generating projects and by the Food and
Agriculture Organization's (FAO) Marketing and Credit
division in the promotion of female entrepreneurship. They
are based on the understanding that knowledge and expertise
exists among local communities, together with a diagnostic
capacity attuned to local realities. They draw on the
expertise of those who are locally recognized as knowledge-
able within the subject problem area by facilitating the
preparation of case studies of their successes. These
cases are exchanged and analyzed at workshop sessions,
leading to identification of interventions that would allow
these successes to be replicated. A great deal of exper-
ience now exists to guide the preparation and implemen-
tation of workshops with those who have little formal
education and to facilitate the participation of service
officers (agricultural researchers or extension workers).


Both these techniques have the added advantage that
J ; they eliminate some of the stages of "translation" of know-
S ledge concepts and, with careful preparation, it is not too
Difficult to identify those items which, though denoted
S differently, refer to a standard unit. Returning for
S example, to the case of the cake that is "ready", the
Researcher can measure the hours it takes to cook and the
baker the cooking that is needed to make it "ready". Both
are referring to a standard referent, although the baker
S might be interested in the number of mouths it feeds and
the researcher in its unit weight and composition. The
difficulty comes when one is using a knowledge concept that
Shas no referent in the knowledge system of the other. The
difficulty is, in a sense, one-sided. Researchers are
often keen to learn about and understand the concepts of
; producers but have little awareness of the constructs and
-< values inherent in their own knowledge system. Where the
knowledge system of male agricultural researchers and
extension officers does not encompass either an experi-
ential nor trained understanding of the domestic economy,
the problem seriously undermines FSR/E practitioners'
claims to be conducting systems based technology
development. .-

>1) /


(1) The research was carried out between January 1979 and
September 1980 in the Central, Northern and Luapula
Provinces of Zambia by members of the Rural Development
Studies Bureau, University of Zambia, Lusaka.


Huss-Ashmore, R.
1982 Seasonality in Rural Highland Lesotho: Method
and Policy. In A Report on the Regional Workshop on
Seasonal Variations in the Provisioning, Nutrition
and Health of Rural Families, pp. 147-161. Nairobi:
Ministry of Agriculture and Water Development (MAWD)
1983 Handbooks for Agricultural Field Workers,
Zambian Local Vegetables and Fruits. Lusaka:
Department of Agriculture, Ministry of Agriculture
and Water Development.

1-- y


Norman, D., E. Simmons, H. Hays
1982 Farming Systems in the Nigerian Savanna,
Research and Strategies for Development. Boulder,
CO: Westview Press.
Wallman, S.
1965 The Communication of Measurement in Basutoland.
Human Organization 3:4:236-243.

From Recommendation Domains to
Intra-Household Dynamics and Back:
Attempts at Bridging the Gender Gap
Amalia M. Alberti

One of the concepts to emerge from the Farming Systems
approach to agricultural research and extension is that of
the recommendation domain. Defined as "a group of roughly
homogeneous farmers with similar circumstances for whom we
can make more or less the same recommendation" (Byerlee, et
al. 1980), the underlying assumption is that farmers of
households within the same recommendation domain will have
similar responses to proposed technology (Shaner, et al.
1982: 44). Recommendation domains are used to focus the
research process and expedite dissemination of the
technology thereby facilitating the extension phase.
The debate continues in the farming systems literature
and among farming systems practitioners about both the more
relevant criteria and the preferred timetable to identify
and to elaborate recommendation domains. The position of
those who maintain that the early delineation of recommen-
dation domains precludes considerations that are not read-
ily evident or initially salient (Cornick and Alberti
1985), is countered by others who maintain that the early
identification of recommendation domains permits their
progressive refinement (Franzel 1984). Still others
(Norman and Baker 1984) point out that in the last analysis
both the target groups identified and the nature of the
technology recommended tend to reflect the expertise of the
team members in a particular Farming Systems Research/
Extension (FSR/E) project.
This paper argues that, first, recommendation domains
sensitive to gender issues are difficult to develop due to
scant documentation of women's participation in agricul-
tural and farm-related activities in local areas, and
second, if developed, they are difficult to implement due
to several features common to many FSR/E projects. Indeed,
it seems that the greater the pressure for prompt

elaboration of recommendation domains, the greater is the
likelihood that women's roles, as well as their concerns
within the FSR/E context, will be overlooked because of
insufficient time to draw them out. The long term solution
to satisfactorily addressing gender issues in farming
systems, however, lies less in attempts to develop appro-
priate recommendation domains and more in efforts to revise
the FSR/E framework so that gender issues are deliberately
and self-consciously entertained. Until these changes
occur, several key questions are proposed to assist FSR/E
practitioners in assessing what gender related issues are
potentially relevant in a particular FSR/E site and whether
or not they can be addressed feasibly within the existing
project framework.


Among the more common techniques suggested for the ini-
tial stages of problem diagnosis leading to the formation
of recommendation domains are reviews of secondary data,
informal interviews with persons such as local officials,
residents, and extension workers, and an exploratory survey
of farmers sometimes combined with or followed by a formal
survey (Harrington and Tripp 1984; Shaner, et al. 1982).
The obstacles to uncovering the extent of women's involve-
ment in the total or select phases of a farming system
embedded in each of these techniques are discussed briefly.

Secondary Data Reviews

Much of the literature on women in agriculture pub-
lished within the last ten years underscores the extent to
which the involvement and contributions of women in this
area have been underrepresented (Deere and Leon 1981; Lewis
1981). Nevertheless, secondary sources such as census data
and local agricultural reports that continue to ignore or
underestimate female contributions abound. When FSR/E
staff consult these materials they are likely to accept the
data as factual unless they are aware of the possibility
that female participation in agriculture may be masked or
otherwise distorted. Only when sensitized to this bias may
they be persuaded to seek additional corroboration before
dismissing gender as a potentially relevant variable.

Informal Interviews and Exploratory Surveys

Local officials and extension agents can often provide
site-specific information that a FSR/E project staff member
would be hard-pressed to obtain so efficiently otherwise.
Information on female involvement in agriculture, however,
is less likely to be obtained for several reasons. First,
cultural values may intervene. When female agricultural
activity is associated with poverty, not only are male
officials unlikely to discuss such activity on the part of
female members in their own household, but they may well be
reluctant to discuss such activity on the part of female
residents in general presuming that it would reflect nega-
tively on the socioeconomic status of the community.
Assuming that these local officials and extension
agents are almost always exclusively male, attempts to
adjust for these gender-related "blind-spots" by speaking
with their wives or other female household members may not
yield substantially different results (Alberti 1980). To
the extent that these women partake of the elevated social
status of their households, they are unlikely to make pub-
lic their own involvement in farm-related tasks or to imply
difficult socioeconomic conditions within the community by
referring to such activity on the part of other women
unless it is to demean them.
Second, it has been found that male farmers routinely
underestimate the degree and undervalue the importance of
female involvement in farm-related activities in which they
too participate (Bourque and Warren 1981; Deere and Leon
1981; Alberti 1986) and ignore or are unaware of the extent
of female involvement in farming activities in which they
do not share. Hence asking male farmers about the
participation of females in agriculture will not necessa-
rily elicit accurate information.
Finally, the reluctance of the national male FSR/E
staff members, especially if they are from the local area,
to ask questions that are deemed inappropriate by local
standards must be considered Moreover, cultural norms
may restrict male field staff members' access to women for
interviews. As yet another possibility, male FSR/E staff
members may resist interviewing women because of their own
attitudes about female participation in agriculture.

Formal Surveys

The advantages and limitations of formal surveys have
been widely discussed. Within the farming systems litera-
ture, Chambers (1981, 1983) Chambers and Ghildyal (1985) is

perhaps the most outspoken and graphic critic as he
conjures up visions of "30 pages of questionnaire ... which
if asked are never coded, or if coded never punched, or if
punched never processed ... examined ... or analyzed..."
that a number of us have also seen (1980: 4).
Two points about formal surveys on women's involvement
in agriculture must be raised. The first is that preparing
a questionnaire assumes knowing what is necessary and how
to design the queries. While this ought eventually to be
true for initial surveys conducted during project develop-
ment, this is not always the case for women's issues pre-
cisely as a consequence of some of the limitations just
discussed. Secondly, many formal surveys are designed to
be administered to either the male or female head of house-
hold, but not to both. Generally, the household member
available when the interviewer arrives responds. However,
the survey form frequently lacks an item to indicate who
was actually interviewed and whether or not that person was
male or female. Hence, even if relevant questions about
women's involvement in agriculture and farm-related
activities are included, it is impossible to disaggregate
male and female responses and to analyze them for
consistency and comparability.


Given these constraints what site-specific information
might be readily available that would expedite developing
recommendation domains sensitive to gender differences?
Readily available refers to information that could be
elicited over a few days through informal conversations
with local residents, teachers, and other persons working
in the area. Rapid Rural Appraisal procedures recommend
doing this in conjunction with a field trip around the
project area (Chambers 1980; Beebe 1985). The field trip
is essential to provide visual information to accompany
verbal accounts. Lines of inquiry otherwise not considered
may be opened when the information from these two sources
does not concur.
The information obtained from responses to the fol-
lowing questions ought to enable FSR/E practitioners to
contextualize the situation of women in the FSR/E setting
in broad strokes. At the same time, it would facilitate a
quick assessment of whether or not the FSR/E project, as it
exists or could feasibly be modified, can viably address
the gender issues relevant to that site. Where addressing
those issues is possible,

collecting the kind of information needed should follow
ideally to inform analyses of intra-household dynamics in
FRS/E (See for example, Flora 1984; Feldstein 1987).

What Are the Local Cultural Norms Regarding Female
Agricultural Activity? Is More Than One Culture
Represented in the Project Area?

In many parts of Latin America, particularly indigenous
regions of the Andes, women work side by side with men in
the fields. In other areas, such as Honduras, women are
seldom seen working in the fields beneath the direct rays
of the sun and may well be embarrassed if they are seen.
Asian women such as those from Bangladesh are rarely field
workers while many of their Indian counterparts assume the
major role in most if not all phases of rice production.
The differences are largely the result of cultural
variations whose dominant mode of expression may be
religious or ethnic or some combination of the two. What
is important is that when a certain portion or subportion
of the population of an area shares a particular cultural
orientation, it is possible to make certain assumptions
about the kinds of roles women are likely to assume within
an agricultural setting and to be forthcoming with
information about those roles. For example, if visible
agriculturally productive activity on the part of women is
highly circumscribed, it can be expected that even when
women do engage in such endeavors, they will be extremely
difficult to document.
When more than one cultural group is represented in an
area, additional factors may come to bear on the situation.
Is one group dominant and the other subordinate? Is the
participation of women in agricultural and farm-related
tasks the same for both groups? Are the norms regarding
such involvement the same? If the norms vary, which norms
do agricultural extensionists and field workers represent?
In culturally complex settings, it is important to spe-
cify the cultural group or groups to which a recommendation
domain applies. This should help clarify and explain what
would otherwise be unanticipated responses to a recommended
technology. Factors that might be involved include differ-
ential access to extra-household labor by ethnic group or
different production objectives despite use of the same
traditional technologies.

Does Women's Participation in Agriculture Vary by Social
Class? If so, in What Ways?

There is an ever-growing consensus that participation
by households, and by women within those households, in the
farming system is highly contingent on social class. Women
from land-poor households who engage in farming tasks tend
to work longer hours at those tasks and generate propor-
tionately lower returns than other women. Often they are
the women who have been left behind while their male part-
ners migrate in search of wage employment. Women from
landless households are clearly the most vulnerable as they
are increasingly dependent on an ever more tenuous agri-
cultural wage labor market that relegates them to more
restricted and marginal employment opportunities even as it
expands commercially (Hart 1978; Stoler 1977; Sen 1985;
Stolcke n.d.; Young 1985; Horn and Nkambule-Kanyima, 1984;
Chaney and Lewis 1980).
While these trends may be widespread, they are not uni-
versal. Knowing whether or not they are valid for a
particular setting should provide some clue as to how
candid men or women are likely to be about female
involvement in agricultural and farm-related activities.

Do Women Specialize in Food Production and Subsistence

Despite broad variations in patterns, the preeminent
role of women in the production of food for home consump-
tion appears to cross continental bounds (Chaney and Lewis
In Latin America the evidence is widespread that the
majority of women who directly engage in agricultural pro-
duction at the household level primarily raise basic crops
intended for home consumption though they may also market
small portions of those crops. If the household also raises
a cash crop, it is likely to be under the control of the
male head of household, even when women contribute labor to
its cultivation. The more the household's agricultural
activities are commercially oriented, the less likely it is
that women of the household will be directly involved in
agricultural production. However, when hired laborers are
present, women of the household are usually expected to
provide support services such as food preparation and are
occasionally called upon for managerial activities (Deere
and Leon 1981; Bourque and Warren 1981; Alberti 1986).

In Asia, the scenario is distinctly different. Despite
broad variations in the extent of women's direct involve-
ment in rice-based agricultural economies due to ethnic and
religious differences, women are always involved in the
processing of rice and frequently bear major responsibility
for its transplanting, weeding, and harvesting. When the
household's access to rice fields is insufficient to meet
its own consumption needs, women as well as men are likely
to seek work as agricultural laborers with rice as the pre-
ferred medium of payment. Participation in the harvest of
kin and neighbors, if not in the planting as well, is
another strategy geared to insure a ration of rice (Hart
1978; Sen 1985; Dey 1985). In each instance, the over-
arching objective is to obtain food that can be immediately
used by the household.
In contrast with rice cultivating areas, in the Philip-
pines for example, the cultivation of cash crops such as
coconut, tobacco, and commercial varieties of root crops
such as cassava and sweet potato, is dominated by men.
Root crops grown for home use, however, are often under the
immediate control of women (Cornick and Alberti 1985).
Until recently, the situation in Africa presented what
had probably been the most consistent association between
crops and gender. Even now, food crops are grown almost
exclusively by women, though some women, particularly those
near urban areas, have begun to cultivate cash crops as
well (Ferguson and Horn 1985). In contrast, men continue
to concentrate their efforts in cash crop production.
Getting a sense of the pattern that predominates for a
given FSR/E project should help us identify the crops and
animals that women tend to work with as well as to assess
the FSR/E project's capabilities in those areas.


After a discussion of how this information may be
fruitfully used, the following areas must be considered.
First, the knowledge gained should enable researchers to
better identify the variables that are particularly rele-
vant in relation to women in farming systems in the project
area. Second, it can provide guidelines to estimate the
validity of the information and data that does exist.
Third, it highlights the kind of information that is
available while giving some indication of what is lacking.
This should help to assess what additional information is
needed and to appraise how sensitive its collection may be.

For example, the knowledge that there are two ethnic
groups within the FSR/E project area should immediately
prompt the question as to whether or not their attitudes
toward agriculture and women's involvement in agriculture
are the same. If they differ, ways of systematically
distinguishing responses by ethnic group becomes important.
The shortcoming of these illustrations is that real
life situations rarely fall into compartments that vary so
neatly along a single dimension. Rather, multiple varia-
bles combine and fuse, whether systematically or errati-
cally, resulting in ever more complex relationships. Their
salience is heightened as they interact with some of the
more common features of farming systems projects. Let us
examine some of these characteristics and the way they
interact with gender concerns.


Site selection for farming systems projects often
results from political and economic decisions that occur
outside project bounds (Shaner et al. 1982; Harrington and
Tripp 1984). Marginal areas are less likely to be
selected. Not only do they tend to lack political lever-
age, but projects in such areas are more prone to failure
due to their residents' restricted access to resources.
Women who engage in agricultural and farm-related activi-
ties, however, are frequently concentrated among the
resource poor who are commonly located in more marginal
Despite occasional efforts to the contrary, farming
systems projects are frequently commodity-oriented either
as the result of project mandate, team member expertise, or
a combination of these factors (Norman and Baker 1984).
Furthermore, a commodity orientation is frequently aligned
with a commercial orientation. As has been discussed,
women are more likely to cultivate food crops with a view
to household consumption. Hence, when a FSR/E project has
a commodity orientation it may implicitly ignore women by
excluding crops of most concern to them.
Another constraint is that FSR/E projects tend to adapt
already existing technology, or "shelf technology," to a
particular situation, rather than to develop new technology
for a specific situation. They justify their approach on
the basis of insufficient resources and a time frame inade-
quate to allow for additional research. However, existing
technologies have tended to be capital intensive, and until
recently, to give demonstrated results only when adopted as

an entire package, rather than in steps over time. To the
extent that the women who engage in agricultural and farm-
related activities are concentrated among the resource
poor, they may be unable to adopt the new technology
because of insufficient cash resources, or if they have the
resources, they may be unwilling to adopt the new techno-
logy because it is inappropriate to their goals when they
are subsistence rather than commercially oriented.
Finally,to paraphrase Chambers, these factors interlock
(1980:3). As Harrington and Tripp note: "domains are
formed so that researchers can effectively deal with the
majority of farmers in a particular area" (1984: 14).
However, the only majority that women tend to constitute as
household level agriculturalists is that of the rural poor.
Nevertheless, even among these, some women have partners,
others are single, some are the only farmer in the
household, and still others are the only sources of labor.
Though women who directly engage in farming and farm-
related activities are unlikely to be wealthy, it is likely
that there is considerable variation in their access to
resources, even among those broadly labelled as "poor."
Women in agriculture tend to share a disadvantaged
position in male-oriented agricultural research and devel-
opment programs. The way they experience that disadvan-
tage, however, is mediated by their culture, resources, and
civil status. It is difficult for recommendation domains
that depend on homogeneous circumstances in key variables
to locate issues that relate to "women" equally despite
their diversity. What a true incorporation of gender
issues in farming systems implies is a revision of the
farming systems unit of analysis from the household to the
male and female members within the farming systems
household for the stages of problem diagnosis and design.
The information thus provided would enable farming systems
practitioners to make conscious though difficult choices
about where the FSR/E resources will be channelled, knowing
full well and in advance how those choices are likely to
affect men and women differentially.


Alberti, A. M.
1979 Metodologia Apropriada para el Estudio de la
Mujer Rural en los Andes del Ecuador. Presented at
the workshop Metodologia Apropriada para Estudiar a
la Mujer Rural. Quito: CEPLAES/Ford Foundation.
1982 Observations on the Productive Role of Women and
Development Efforts in the Andes. Proceedings of
the First Women in Development Workshop sponsored by
HIID/MIT. Cambridge, MA: Harvard University Press.
1986 Gender, Ethnicity, and Resource Control in the
Andean Highlands of Ecuador. Ph.D. Dissertation.
Stanford, CA: Stanford University.
Beebe, J.
1985 Rapid Rural Appraisal: The Critical First Step
in a Farming Systems Approach to Research. FSSP
Networking Paper No. 5.
n.d. Factoring New Information into Decisions About
On-Farm Trials. USAID Mission, The Philippines.
Bourque, S. C. and K. B. Warren
1981 Women of the Andes: Patriarchy and Social
Change in Two Peruvian Towns. Ann Arbor, MI:
University of Michigan Press.
Chambers, R.
1981 Rapid Rural Appraisal: Rational and Repertoire.
Public Administration and Development 1:95-106.
1983 Rural Development: Putting the Last First.
London: Longman.
Chambers, R. and B.P. Ghildyal
1985 Agricultural Research for Resource-Poor Farmers:
The "Farmer-First-And-Last" Model. Agricultural
Administration 20:1-30.
Chaney, E. M. and M. W. Lewis
1980 Women, Migration, and the Decline of Smallholder
Agriculture. Washington, D.C.: USAID Office of
Women in Development.
Cornick, T. R. and A. M. Alberti
1985 Recommendation Domains Reconsidered. Presented
at the Farming Systems Symposium: Farming Systems
Research and Extension: Management and Methodology,
Kansas State University, Kansas.

Deere, C. D. and M. Leon de Leal.
1981 Women in Agriculture: Peasant Production and
Proletarianization in Three Andean Regions. ILO
Rural Employment Policy Research Program:
Unpublished Working Paper.
Dey, J. M.
1985 The Concept and Organization of a Research
Network on Women in Rice Farming Systems. In the
Report of the Project Design Workshop on Women in
Rice Farming Systems. Los Banos, Philippines:
Feldstein, H., S. Poats, K. Cloud, R. Norem
1987 Intra-Household Dynamics and Farming Systems
Research and Extension Conceptual Framework.
Population Council/FSSP-University of Florida
Working Document.
Ferguson, A. and N. Horn
1985 Situating Agricultural Research in a Class and
Gender Context: The Bean/Cowpea Collaborative
Research Support Program. Culture and Agriculture
Flora, C. B.
1984 Intra-Household Dynamics in Farming systems
Research: The Basis of Whole Farm Monitoring of
Farming Systems Research and Extension: A Position
Paper. Mimeograph.
Frankenberger, T. R.
1985 Adding a Food Consumption Perspective to Farming
Systems Research. Prepared for the Nutrition
Economics Group, Technical Assistance Division,
Office of International Cooperation and Development,
United States Department of Agriculture.
Franzel, S. C.
1984 Modeling Farmers' Decisions in a Farming Systems
Research Exercise: The Adoption of an Improved
Maize Variety in Kirinyaga District, Kenya. Human
Organization 43:3:199-207.
Harrington, L. W. and R. Tripp
1984 Recommendation Domains: A Framework for On-Farm
Research. CIMMYT Economics Program Working Paper.
CIMMYT: Mexico.
1985 Formulating Recommendations for Farmers,
Researchers and Policy-Makers: Issues in the Design
and Use of Socio-Economics Research. Presented at
the ESCAP CGRPT Center Conference "Towards
Recommendations for Research, Policy and Extension:
Methodological Issues in Socioeconomics Analysis of
Coarse Grains and Food Legumes," Bandung, 18-23

Hart, G.
1978 Labor Allocation Strategies in Rural Javanese
Households. Ph.D. Dissertation, NY: Cornell
Horn, N. and B. Nkambule-Kanyima
1984 Resource Guide: Women in Agriculture in
Botswana. Michigan: Bean/Cowpea Collaborative
Research Support Program.
Lewis, B. C.
1981 Invisible Farmers: Women and The Crisis in
Agriculture. Washington, D.C.: USAID Office of
Women in Development.
Norman, D. W. and D. C. Baker-
1984 Components of Farming Systems Research FSR
Credibility and Experience in Botswana.
Agricultural Technology Improvement Project,
Department of Agricultural Research. Botswana,
Africa: Ministry of Agriculture.
Sen, G.
1985 Women Workers and the Green Revolution. In
Lourdes Beneria, ed., Women and Development: The
Sexual Division of Labor in Rural Societies. NY:
Praeger Publishers.
Shaner, W. W., P. F. Philipp, and W. R. Schmehl
1982 Farming Systems Research and Development.
Boulder, CO: Westview Press.
Stolcke, V.
n.d. Social Inequality and Gender Hierarchy in Latin
America. Position paper presented at the SSRC
Stoler, A.
1977 Class Structure and Female Autonomy in Rural
Java. Signs 3:74-89.
Young, K.
1985 The Creation of a Relative Surplus Population:
A Case Study from Mexico. In Lourdes Beneria, ed.,
Women and Development: The Sexual Division of Labor
in Rural Societies. NY, NY: Praeger Publishers.

Research, Recommendation and
Diffusion Domains: A Farming Systems
Approach to Targeting

Peter Wotowiec, Jr., Susan V. Poats, and Peter E. Hildebrand


Inherent in the farming systems approach is the recog-
nition of the variability of the complex circumstances
farmers face while managing farms that are comprised of
inter-related crop, animal, household, and off-farm
enterprises. Diversity in farming systems must be
recognized in developing appropriate technologies for the
farmers that manage those systems. However, it is not
practical to conduct research tailored specifically to a
few individual farmers. Targeting entails the grouping
together of similar clientele so efforts can be suffi-
ciently focused. Although the concept of targeting might
seem contrary to the recognition of heterogeneity among
farms, it is an essential component of the farming systems
approach. When Farming Systems Research and Extension
(FSR/E) practitioners target a group of farming systems as
relatively homogeneous based on a few simple factors, the
existing variability among farms is often not sufficiently
considered. How can FSR/E teams define and target homo-
geneous groups of farming systems without losing sight of
the heterogeneity among them? Farming systems practi-
tioners take different positions on this issue (Cornick and
Alberti 1985).
One perspective stresses the early definition of homo-
geneous groups of farmers using the recommendation domain
concept to guide subsequent research activities. Collinson
(1979, 1980), Gilbert et al. (1980), and Franzel (1985)
advocate ex ante delineation of recommendation domains
based on secondary data and preliminary surveys, followed
by a formal survey to refine the domain boundaries. Both
Collinson and Franzel describe a technique of defining
recommendation domains through interviews with extension
agents and local authorities before actually initiating
activities with farmers. Early definition of

recommendation domains is usually based upon a few
relatively easily identifiable factors such as soil type,
agroecological zones, crop type, and management (Harrington
and Tripp 1985). These authors note the importance of
continuing the refinement of domain boundaries throughout
the sequence of on-farm adaptive research, but the subse-
quent reassessment of recommendation domains is often not
vigorously pursued.
A more recent view states that grouping farming systems
should not take place until the researchers have an ade-
quate understanding of the variability inherent in local
farming systems, usually not accomplished early in the work
in an area. Cornick and Alberti argue that recommendation
domains established early are frequently poorly conceived
and lead to a premature assumption of homogeneity. The
failure to consider potential variability from factors such
as long-term climate induced trends in cropping patterns,
household decision-making and labor allocation, or rela-
tionships between on- and off-farm activities, may bias
subsequent technology development. For example, Cornick
and Alberti (1985:1) note:

...the roles of women and children that can be critical
factors in the development and subsequent adoption of
technologies are often explicitly excluded from consi-
deration in recommendation domains. This occurs because
the usual time frame for development of recommendation
domains is inadequate to the task of understanding intra-
household dynamics and the importance they hold in the

In particular, socioeconomic factors are often not
fully integrated into domains defined early, either because
of the longer period of time necessary to gather this
information, or because of the absence of trained social
scientists as part of farming systems teams. One area
often poorly covered in early definitions of domains is the
different agricultural roles of men and women. Proceeding
with on-farm research and other activities on the basis of
a hastily achieved assumption of homogeneity could result
in inefficient subsequent research and the promotion of
solutions that are not appropriate to farmers (Cornick and
Alberti 1985:25) or technologies that may favor some
farmers (male) while causing disadvantages for others
In this paper the issue of variability versus
homogeneity in the targeting of farming systems research
and extension activities is explored. After a brief review

of targeting in FSR/E using recommendation domains, prob-
lems in the conventional use domains in FSR/E are described
in an attempt to bring together the two differing view-
points and to begin to resolve the question. The refined
concept of targeting allows for better inclusion of gender
variables in the definition of domains.


Targeting for Efficiency and Social Equity

FSR/E must differentiate between various potential
farmer-client groups and determine the particular needs of
each, if technologies are to be developed that clearly meet
those needs (Byerlee and Hesse de Polanco 1982). Most
literature on the subject of targeting in farming systems
has stressed the increase in efficiency of FSR/E activities
made possible through focusing upon specific, relatively
homogeneous farmer groups.
Efficiency in allocation of research resources is
essential if a program is to reach and benefit a maximum
number of farmers. By focusing scarce resources upon
roughly similar groups of farmers, research programs are
able to carry out investigations on a selected number of
representative farms and later can transfer the findings to
the comparable situations faced by other farmers.
Targeting is also important in justifying the farming
systems approach to institutional policy makers who are
concerned about social equity in the distribution of
resulting benefits. Farming systems practitioners use tar-
geting concepts to assist them in making decisions which
increase the likelihood of an optimal distribution of
research results among the members of a community.

Conventional Concept of Recommendation Domains

The concept of "recommendation domains" has been widely
used in targeting farming systems research since Perrin et
al. (1976) first introduced the idea. It is described and
defined by Byerlee et al. (1980:899) in the following

... a recommendation domain (RD) is a group of farmers
with roughly similar practices and circumstances for
whom a given recommendation will be broadly appropri-
ate. It is a stratification of farmers, not area:
farmers, not fields, make decisions on technology.

Socioeconomic criteria may be just as important as
agroclimatic variables in delineating domains. Thus
resulting domains are often not amenable to geographi-
cal mapping because farmers of different domains may be
interspersed in a given area.

Using this definition, neighboring farm households
might be placed in different recommendation domains because
of differences in availability of family labor. In socie-
ties where women cultivate different crops than those of
the men, female farmers could comprise a recommendation
domain separate from male farmers even if they are from the
same household.

Expanding Upon the Definition of Recommendation Domain

Perrin et al. (1976) originally conceived of the notion
of recommendation domains as an aid to researchers for
targeting the development of technologies to specific
audiences. The concept has been expanded since then to
include a number of additional situations and purposes.
Some of the most common applications of recommendation
domains include the following gleaned from current
literature on the topic:

(1) making policy decisions;
(2) identifying priority issues for research;
(3) specifying clientele for developing recommenda-
(4) selecting representative sites and farmer-
(5) focusing analysis of surveys and on-farm trials;
(6) orienting extensionists to groups of similar
(7) transferring adapted technology to appropriate
farmers; and,
(8) enhancing equitable distribution of FSR/E benefits.

As Harrington and Tripp (1985) point out, the domain
concept is vital to every stage of the on-farm research
process. However, it is apparent from reviewing the
literature on the subject that the definition of "recommen-
dation domain" not only changes at each stage, but also
varies according to the individual who applies it as well
as to the end result. The wide variability among farmers
and farms, and the dynamic nature of the farming systems
development sequence, contribute to the confusion that

exists among FSR/E practitioners as to the general meaning
and use of the term recommendation domain.

On-Farm Variability and Conventional Recommendation Domains

The emphasis by Byerlee et al. (1980) upon "farmers,
not fields" as the sole basis for the delineation of recom-
mendation domains is not always warranted because of the
variability found in some field situations. Cornick and
Alberti (1985) cite the case of farmers in the community of
Quimiaq in the mountains of Ecuador who manage different
cropping patterns in different agro-ecological zones, a
product of altitude, temperature, and rainfall variation on
the mountain slopes. Not only does each farm cross agro-
ecological zones, but the cropping patterns found in each
field vary greatly from year to year. For example, depend-
ing upon a farmer's perception of trends and yearly changes
in climatic conditions, bean or fava bean intercrops will
be assigned to maize fields located at varying elevations
along the slope.
Gender and intra-household variables are often
neglected in the process of defining a recommendation
domain because of the relatively more difficult and time
consuming task of collecting and analyzing data on these
variables. Existing information on gender and household
variables often offers few useful insights for defining
recommendation domains when compared to the secondary data
available on agroecological characteristics. In addition,
the gender and household data that may exist may be unob-
tainable locally. Nevertheless, superficial understanding
of these variables or the transfer of erroneous assumptions
without continued investigation can hamper design and deli-
very of appropriate technology.

Refining the Concept of Domains

The argument here is that the issue of targeting in
FSR/E has become confusing because the definition of the
term "recommendation domain" has been stretched to cover
too many situations and too many different purposes. Farm-
ing systems practitioners must develop a common understand-
ing of how the use and definition of "domains" change as
the farming systems sequence progresses from initial char-
acterization through problem diagnosis, testing, adapta-
tion, evaluation, and finally, to the delivery of the new
technology to farmers.
It is essential to account for the heterogeneity in
farming systems, even while delineating relatively

homogeneous groups. Refinement and expansion of the use of
domains in targeting will enable researchers to distinguish
applications of the domain concept, while still recognizing
the diversity among farm households and farming systems.
This can be accomplished by recognizing a problem focus
in the definition of the domains, by tying the changing
concept of domain more closely to the farming systems
sequence, and by stressing a greater inclusion of socio-
economic considerations into the targeting process. The
refinements outlined below are a sharpening of focus not a
changing of terminology, that will lead to increased
utility of this method of targeting in the field.
Any of the three types of domains described below may
be defined within specific geographic boundaries for ease
in conceptualization, but it is imperative to realize that
domains do not necessarily include all the area within the
boundaries prescribed. Because domains are based upon a
specified problem focus and upon socioeconomic considera-
tions in addition to the more geographically mappable fac-
tors of climate, altitude, and soil, they are actually
interspersed intermittently in a discontinuous pattern
throughout a geographic area.
The examples here will emphasize gender as a key factor
in delineating domains; other factors, such as class,
education,-language use, or food preferences, could also be

Research Domains: Targeting for Variability

A "research domain" is a problem-focused environmental
(agro-ecological and socioeconomic) range throughout which
it is expected that hypothesized solutions to a defined
problem could have potential applicability and therefore
should be tested. Research domains are determined during
the initiation of research activities, largely by consider-
ation of biophysical (agro-ecological) factors, with some
attention to socioeconomic and gender issues.

Recommendation Domains: Targeting for Homogeneity

Research domains are comprised of one or more agro-
socioeconomic "recommendation domains", that are tenta-
tively defined based upon the response of a specific tech-
nology to the actual agro-socioeconomic conditions found on
farms. A "recommendation domain" is a group of farmers (or
farmers and their fields) with a common problem for whom a
tested solution meets their biophysical and socioeconomic
requirements for adoption.

In the Ecuadorian case cited by Cornick and Alberti,
recommendation domains would be based not only upon farm
households, but also upon their separate fields that are
not contiguous but widely dispersed in location and alti-
tude. Each household might fall into several recommenda-
tion domains depending upon: (1) where their fields are
located along the agroecological gradient of the mountain-
side; (2) the climate-related crop management decisions
made for each of those fields; and, (3) the particular
problem solutions to be tested.
Other examples from West Africa demonstrate how gender
can be used to differentiate recommendation domains. In
many areas, men and women have separate fields, often
inherited from their same sex parent, that are not managed
communally by the household. Women traditionally grow rice
on their lands while men produce upland crops such as
groundnuts or sorghum on their own fields. In this system,
fields managed by a household pertain to different recom-
mendation domains depending upon both the cropping system
and the gender of the farmer manager. In one area of the
Ivory Coast, men plant yams in a cleared field. Women will
often care for the yam plants by weeding them while they
plant their vegetable crops in the space between the yam
plants. In this system, fields are neither men's nor
women's, nor would entire fields fall into a single
problem-focused recommendation domain. Rather, domains
would be determined by crops and their managers, male or
female, and contain pieces of many fields.
Recommendation domains are seen as tentative in nature
throughout the on-farm adaptive research process. Recom-
mendation domains are initially hypothesized by the FSR/E
team on the basis of on-farm exploratory and refinement
trials, information collected through directed surveys, and
subsequent on-farm verification trials. Over time, as more
information is gathered, the recommendation domains are
refined and redefined to closer approach reality.

Diffusion Domains: Targeting for Communication

"Diffusion domains" are interpersonal communication
networks through which newly acquired knowledge of agri-
cultural technologies naturally flows (Hildebrand 1985).
Informal flow of information through a community grapevine
is substantial (Rogers 1983). From farmer to farmer,
neighbor to neighbor, store operator to patron, information
about new ideas moves through a farming community. Aware-
ness of a new technology being verified in on-farm trials