Front Cover
 Title Page
 Map of project area
 Table of Contents
 List of Illustrations
 Background to the study
 An inquiry into factors underlying...
 A discussion of alternative implementation...
 Back Cover

Group Title: Women's roles & gender differences in development, cases for planners ; 3
Title: Agricultural policy implementation
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00086610/00001
 Material Information
Title: Agricultural policy implementation a case study from western Kenya
Series Title: Women's roles & gender differences in development, cases for planners
Physical Description: xvi, 68 p. : map ; 28 cm.
Language: English
Creator: Staudt, Kathleen A
Publisher: Kumarian Press
Place of Publication: West Hartford CT
Publication Date: c1985
Subject: Agriculture and state -- Case studies -- Kenya   ( lcsh )
Women farmers -- Case studies -- Kenya   ( lcsh )
Sex discrimination against women -- Case studies -- Kenya   ( lcsh )
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
Spatial Coverage: Kenya
Bibliography: Bibliography: p. 65-68.
Statement of Responsibility: Kathleen Staudt.
Funding: Women's roles and gender differences in development, cases for planners ;
 Record Information
Bibliographic ID: UF00086610
Volume ID: VID00001
Source Institution: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 11599878
lccn - 84029737
isbn - 0931816181 (pbk.)

Table of Contents
    Front Cover
        Front Cover
    Title Page
        Page i
    Map of project area
        Page ii
        Page iii
        Page iv
    Table of Contents
        Page v
        Page vi
    List of Illustrations
        Page vii
        Page viii
        Page ix
        Page x
        Page xi
        Page xii
        Page xiii
        Page xiv
        Page xv
        Page xvi
    Background to the study
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
    An inquiry into factors underlying gender differentials in agricultural policy
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
    A discussion of alternative implementation policies
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
    Back Cover
        Page 70
Full Text

4,V -






i 4


nv, m
., -z



Women's Roles & Gender
Differences In Development.

Kathleen Staudt

West Hartford

Copyright 1985 Kumarian Press
29 Bishop Road, West Hartford, Connecticut 06119
All rights reserved under International and Pan-American Copyright
Conventions. No part of this book 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 prior written permission
of the publishers.

Printed in the United States of America

Cover design by

Timothy J. Others

Library of Congress Cataloging in Publication Data
Staudt, Kathleen A.
Agricultural policy implementation.
Bibliography: p. 65
1. Agriculture and state-Kenya-Case studies.
2. Women farmers-Kenya-Case studies. 3. Sex
discrimination against women-Kenya-Case studies.
I. Title.
HD2126:5.Z8 73 1985 338.1'86762 84-29737
S ISBN:0-931816-18-1

29 Bishop Road
West Hartford, CT 06119

-~ -.KENYA-
AZ Kisumu

'"x Nairobi
i*RWANDA i *
SLam0 / /
I IvM.wanza ,Lm
BURUNDI Arusha: asa
Ujiji oTabora PEMBA
T A N Z A N I A Zanzibar
Morogoroo S
SMorogoro Dar es Sa
The Project Area


Preface ....................................................... ix

SUM M ARY .................................................... xi

BACKGROUND TO THE STUDY .................................... 1

Agricultural Policy Objectives................................. 1
Historical Perspectives on Women's Agricultural
R oles ................................................. 2
The Purpose and Design of the Study........................... 6
Research Methodology .................................. 6
Hypotheses ........................................... 7

The Land and the People..................................... 8

The District ........................................... 8
The Sublocations ....................................... 9

A Brief Description of Women's Roles in the
Study Area ............................................. 11

The Farm Unit ......................................... 11
Decision Making ....................................... 12
Control Over Income and income
Responsibilities ..................................... 13
Access to Land and Credit ................................ 13
Organizational Activity, Political Authority,
and Access to Agricultural Services ..................... 14

FINDINGS .................................................... 16

Implications for Kenya Agricultural Policy
Implementation ........................................ 16

Ordinary Implementation: Shikulu Sublocation .................. 16

Improving Agriculture Through a Field-oriented
Extension Service ...................................... 16

Providing Specialized Knowledge to Farmers
Through Formal Training Programs on an
Equitable Basis .................................... .... 19

Providing Production Loans to Farmers on an
Equitable Basis ......................................... 20

Supplying Information About Hybrid Maize and
Cash Crops ............................................ 21

Intensified Implementation: Shitoli Sublocation ................. 23

Improving Agriculture Through Extension
Services......... ................................
Providing Formal Training ...............................
Providing Product on Loans ..........................
Supplying Information About Hybrid Maize
and Cash Crops .................. ..................

Special Services to Saturated Areas:
Lirhembe Cooperative (Shitoli) ...........................

Encouraging Cooperative Membership .........................

Encouraging Cooperative Loans ..............................

Providing Tractor Services to Lirhembe
Cooperative Members ................. ................

Providing High-Grade Cows to Cooperative
Members Who Apply for Them ...........................

Performance Under Ordinary, Intensified,
and Saturated Services Compared .........................


Alternative Explanations of Findings .........................

Explanation 1: Prejudicial Attitudes and
Ideological Bias ....................................
Explanation 2: Women's Invisibility........................
Explanation 3: Administrative Convenience .................

A Discussion of Alternative
Implementation Policies .................................

Alternative A: Improved Management .....................
Alternative B: Intermediate
Structural Change ...................................
Alternative C Thorough Structural Change ..................

Conclusions ........................................

ENDNOTES..................... ............



; i















Table 1. Visits by Agricultural Instructor, by Farm
Management Type ....................................... 17

Table 2. Demonstration Plot Attendance by Farm
Management Type ....................................... 18

Table 3. Training by Farm Management Type ........................ 19

Table 4. Loan Knowledge and Acquisition by Farm
Management Type ....................................... 21

Table 5. Sources of Initial Information about Hybrid
Maize by Farm Management Type (Hybrid Maize Growers
Only) .................................................. 23

Table 6. Sources of Information about Hybrid Maize
Husbandry Practices by Farm Management Type
(Hybrid Maize Growers Only) .............................. 23

Table 7 The Impact of Intensification on Extension Visits,
by Farm Management Types ............................... 24

Table 8. The Impact of Intensification on Access to
Demonstration Plots ..................................... 24

Table 9. The Impact of Intensification on Training ................... 25

Table 10. The Impact of Intensification on Knowledge
About and Acquisition of Loans ........................... 26

Table 11. Initial Source of Information about Hybrid
Maize by Sublocation and Farm Management Type............. 26

Table 12. Source of Information about Hybrid Maize Practices
by Sublocation and Farm Management Type .................. 27

Table 13. Lirhembe Cooperative Membership by Farm
Management Type ....................................... 28

Table 14. Use of Maize and Fertilizer Loans in Lirhembe ............... 29

Table 15. Cooperative Tractor Use by Farm Management Type .......... 29

Table 16. Cooperative Grade Cow Applications by Farm
Management Type ....................................... 31

Table 17. Mean Number of Cash Crop and Food Crop Types
and Years Since Hybrid Maize Adopted by
Sublocation and Farm Management Type .................... 32

Table 18. Responsiveness to Two Crop Promotions by Farm
Management Typel Controlled for Different
Amounts of Services Extended
(Lirhembe Sample'Only) .................................. 34

Table 19. Mean Services by Administrative
and Farm Management Type............................... 36

Table 20. Cumulative Service Scale by Farm
Management Type Controlled for
Number of Cash Cops Grown.............................. 39

Box 1. Alternative Strategies to Reach Women Farm Managers ......... 48


Why should development planners and scholars of
development be concerned about women's roles and gender

No project that expresses its goals in terms of
production gains or increased benefits can afford to ignore
the economic potential and needs of one-half of the
population. Guidelines for the design and evaluation of
development projects sensitive to women's roles have often
been applied only to a narrow range of "women's projects."
Our view at the Population Council is that all development
efforts could be improved if the differential impact on both
class and gender groupings were considered.

The series of case studies on Women's Roles and Gender
Differences in Development was developed to demonstrate that
such analyses are not only essential, but also feasible
within existing structures.

These case studies make clear how inattention to
women's roles and gender differences is played out as
projects are implemented. Excluding gender as a variable,
or limiting women's roles to the welfare sector, results in
unintended effects, sometimes positive, but more frequently
negative. Many of the stated objectives of the development
schemes under study were not attained because project
designs were predicated on an incomplete picture of the
society to be served and drawn into participation.

The case studies draw largely from material that
existed originally in other forms (such as exceptional Ph.D
dissertations). From these materials has been extracted the
"case:" (1) salient aspects of the culture and society in
which the development project was placed, (2) the project
dynamics themselves, and finally, (3) an assessment of gains
and losses in different goal areas. To complement
individual case studies, the Series for Planners includes
monographs on broader development phenomena whose effects
are seen outside the confines of specific development
schemes. As of this writing, the Series includes two
monographs, one dealing with the effects of male out-
migration on rural women's roles and a second on the impact
of different styles of agrarian reform on women's roles and

These materials are intended to be used by students of
development and professionals in the field, including those
at the highest planning levels. By providing examples of
how individual development schemes have operated vis-a-vis
gender, we hope they stimulate in the reader an interest in
exploring what these effects might be in development
projects being designed, implemented, or evaluated. For
some time now, an understanding of class dynamics has been
seen as essential in designing projects for successful
outcomes. We I have the same conviction regarding the
importance of understanding gender differentials. We hope
that this study series positively advances that notion and
provides its readers with new skills and insights by raising
questions and suggesting alternatives.

We wish to thank each of our individual authors for the
exhaustive work they have put into forming their material
into case studies., ie commend Marilyn Kohn for her fine
editorial work in finalizing the material.

Judith Bruce
The Population Council

Ingrid Palmer
Editor of the Series



Many Kenyans depend on agriculture for their
livelihood. In its efforts to increase productivity,
improve farmers' livelihood, and expand export crops, the
Kenyan government relies on an agricultural extension
service to provide information to farmers.

In a large part of western Kenya, men have
traditionally migrated elsewhere in search of wage labor,
while women stayed at home and managed agricultural
production. Even prior to the establishment of this
pattern, women performed a major portion of the agricultural
labor and management of the food supply. Virtually all
women belong to organizational networks through which they
share agricultural labor and provide mutual aid. Men
dominate administrative offices and political authority
networks which provide contacts and information about
valuable agricultural services.

The purpose of this study is to compare the access of
female-managed farms to agricultural services with that of
jointly managed farms (i) to determine whether there are
inequities and (ii) to evaluate the consequences of those
inequities for female managers' farm performance and, by
extension, overall agricultural performance in Kenya.

Seven agricultural policy objectives derived from
official documents and implementation realities in the
Kakamega District of western Kenya frame this analysis, as
summarized below:

1. To assist all farmers to improve
agriculture through a field-oriented extension

2. To provide specialized knowledge to
farmers through formal training programs on
an equitable basis;

3. To provide production loans to farmers on
an equitable basis;

a. To encourage cooperative membership and
receipt of cooperative loans;

4. To provide available technology to
cooperative members in an equitable manner;

5. To provide other cooperative benefits to
members who apply for them;

6. To supply information to all farmers about
hybrid maize and cash crops; and,

7. To decrease farmers' risk through
adaptiveness and diversification.

The study is based on field research carried out in
western Kenya from 1974-1975 !in three types of
administrative settings: ordinary, intensified, and
saturated. Survey data from 212 households in Shikulu the
ordinary sublocation representing a standard or typical
level of services, are first analyzed to identify typical
patterns of service for female and jointly managed farms.
Then the study examines access in Shitoli the "intensified"
sublocation where, as the term suggests, increased services
are provided, to assess whether intensifying staff and
resources affects i disparities between household types.
Finally, a subsample i' Shitoli of the Lirhembe multipurpose
cooperative members 's analyzed to determine the effects of
further increasing or "saturating" services. A quasi-
experimental design is used to assess the impact of
inequitable service delivery. Two administrative units,
similar in culture, economy, and geography, but different in
administrative intervention, allow qualified judgments to be
made about programmatic causes of differential performance.

Households, the units of analysis in this study, are
divided into female-and jointly managed farms. The women in
female-managed farms, two-fifths of the sample, were either
widowed or separated from their husbands, who had migrated
outside the area for wage employment. Jointly managed farms
had a man present, but they have not been termed "male-
managed" because of women's substantial involvement in farm
work and decision making, even with husbands present. These
jointly managed farms are husband-wife teams, or corporately
managed farms with intergenerational management since land
has not yet been parcelled to sons. Data show that female-
and jointly managed farms are similar in their economic
standing and in the amount of land they farm.


The analysis reveals that the objectives outlined above
were not met, due to the preferential delivery of services
to farms with a man present. In sum, female-managed farms
always received feweri services than jointly managed farms,
and gaps increased as the services become more valuable.
Comparisons of both sublocations show that female managers'
access to services in intensified areas was greater than in
ordinary areas, but that gaps between management types
remained the same. While gaps between the sexes were
smallest under conditions of saturated services in the
multipurpose cooperati e sample, except in the allocation of
valuable grade cows, women managers were less likely to be

enrolled as cooperative members.

Extension Visits

The most common services in Kenya are the home visits,
conducted by extension officers, which allow for personal
contact and specific recommendations. This costly
administrative strategy is, however, rife with
discrimination potential. Group extension demonstrations
made at local farmers' plots offset tendencies for officers
to skew visits to the privileged and are effective to the
extent that farmers are aware of the meeting and able to
attend. Yet few group demonstrations are given in western

Findings indicate that female-managed farms always had
less access to extension visits and demonstration plots than
jointly managed farms (See Tables 1, 2, 7 & 8. This can be
explained by the overwhelming predominance of male extension
officers (in a culture where men and women strangers avoid
contact), symbolic male authority over households despite
extensive male absence, and the tendency for announcements
about demonstration plots to be made at community meetings
largely attended by men. Presumably, the presence of more
staff should increase access to services for groups
heretofore underserved, without depriving those who already
have access. In the intensely served area, farmers received
more visits, though disparities remained.


Courses at the nearby farmer training center provide a
one to two-week intensive experience for farmers. It is
rare for farmers to complete a course, but jointly managed
households were four times more likely to have a member
trained than female-managed farms. (See Tables 3, 9.) The
minimal access for female managers is not surprising, given
the tendency demonstrated in the research sample to pass
information about opportunities through male communication
networks and to use training as a way of conferring status.
Moreover, female managers faced difficulties arranging for
household responsibilities and cultivation tasks to be
carried out in their absence, as they could not rely on
resident spouses. In the intensely served area, farmers in
general were only slightly more likely to receive training.


Few of Kenya's small farmers received Guaranteed
Minimal Return crop insurance loans, or loans from the
Agricultural Finance Corporation. In the case of the
former, minimum acreage requirements made virtually all

farmers ineligible, as for the latter,, a guarantee such as a
land title deed or reg lar salary is necessary. Land reform
was only completed in his part of Kenya in 1973, but since
then it isimainly men -- in whose names title deeds are
overwhelmingly held -- who have access to this valuable
service. Applying for a loan is a long and complex process,
intricately connected! to local political contacts with
authorities and institutions. While only 2 percent of those
in jointly managed farms acquired loans, an additional 12
percent applied for or could correctly describe the complex
application process, in contrast to the 1 percent of female
managers who knew the application process. (See Tables 4,
10.) Given women's greater responsibilities for the
production and storage of food for family substenance, their
need for loans is acute.

Special Service to Saturated Areas

The multipurpose cooperative society saturates services
in the provision of easy access to maize seed, fertilizer,
grade cow loans, and tractor services. Once enrolled in the
cooperative, farmers experience less differentiated access
to maize seed and fertilizer loans, as compared to the
larger sample; female managers had slightly more access
among the few who used the short-lived tractor service
(Table 14, 15). However, for the valuable grade cows, a
service tied closely to the cooperative political process,
jointly managed| farm| with a man present were twice as
likely to obtain a cow than female managers, even though
equal numbers applied ('Table 16). Overall, female managers
were less likely to be recruited into the cooperative and
thus to benefit from the service (Table 13).

Supplying Information

Information about hybrid maize, the most important
innovation in the area, !spread differently to households,
depending on whether a !man was present. Members of jointly
managed farms were more likely to hear about hybrid maize
and its husbandry practices directly from technical sources,
while women generally learned about them from their
neighbors, members of women's groups, iand traders (Tables 5,

Performance Under Different Settings Compared

What is the significance of these inequalities for
farmers' performance? A controlled comparison of the
minimally served Shikulu and the more intensely served,
historically iadvantage'd Shitoli area reveals that under
conditions of unequal, male-oriented agricultural services,
female managers experience a decline in their relative farm


performance compared to men and a decline in their absolute
performance compared to female managers in the ordinary area
(Table 17). However, land scarcity also plays its part in
this decline. With land being less scarce and with access
to minimal services in the ordinary area, female managers
are able to share labor with other women in their rich
organizational networks, diversify, and adopt innovations
faster than farms with a man present. In the intensely
served area, the adoption rates of particularly appropriate
crop mixes for female managers integrated into the extension
delivery were double the rate on jointly managed farms
(Table 18). When performance is not distorted by
inequities, female managers' notable performance has also
been documented in other studies. For example, research on
yield per acre just south of this site found that female
managers had higher productivity than men when access to
education and extension were held constant.


Extraordinary levels of services seem necessary to
compensate for the gender differentiation that occurs under
ordinary implementation, but this is achieved at the cost of
farmers' dependency on costly administration. Three
explanations for such differentials are set forth in this
study, based on observation and research. First,
prejudicial attitudes and ideological biases against women
persist -- even against innovative, wealthier women managers
farming on relatively large land sizes. Second, women are
conceptually and politically invisible to the
administration. As individuals they rarely initiate contact
with the administration because of the suspicion and
obligations they might incur, and in groups, they tend to
operate autonomously of local politics, except for
occasional linkages through wealthier women, whose
priorities differ somewhat from those of low-income women.
Third, administrators find it convenient to continue long-
standing practices of neglecting women managers.

The lesson which emerges from these observations is the
need to transform agricultural services if female-managed
farms are to be reached, rather than simply to augment
existing practices or to intensify services. Policymakers
might consider three sets of alternative recommendations
(see Box 1) designed primarily to affect the point of
contact between farmers and field staff. Virtually all
recommendations are cost-free, except for the energy needed
to alter administrative inertia. The three strategies build
on one another, leading from improved procedures to spread
responsibility among men and women professionals,
paraprofessionals, and community groups. Recommendations
are broken down into categories of staff and staff training,
extension and farmer training, credit, and cooperatives, and
are summarized below.


In the first, improved management strategy,
administrators might alter the incentive structure in which
field staff work to reward effective and equitable
administrative performance among the largely male staff who
now tend to avoid women managers. Administrators could
further alter the composition of agricultural staff to
recruit more females to serve as agricultural rather than
home economics staff. Moreover, greater emphasis on group
extension, to both men's and women's groups, would expand
coverage in equitable ways at lower costs than individual
home extension. I Staff should also assure that information
about training opportunities passes through women's networks
as well as men's. Administrators might also establish
minimum quotas to increase the percentage of loans to women
managers, and examine application requirements to remove the
discrimination that results from a land reform in which
primarily men's names were placed on title deeds.
Cooperatives should be encouraged to recruit women managers
specifically and to call for joint membership when partners
live together.

Using the second strategy, which builds on the
recommendations above but also alters structures slightly,
administrators might consider extending extension coverage
with low-cost men and women paraprofessionals who live in
areas they serve. hSimultaneously, staff from Farmer
Training Centers should be brought to the community to
expand coverage to those who cannot easily leave farms for a
week or two. At the same time, it must be recognized that
home economics training is no substitute for agricultural
training. Furthermore, alternative means to deliver credit
could be made available in group loans to existing or newly
formed women's groups,! secured through group liability and
consequent peer pressure. Lending facilities, mobile or
located closer to farmers, would expand coverage to the less
mobile, such as women managers.

The last strategy alters structures more significantly.
Staff would work with community groups to share
responsibility for decisions about agricultural change.
Locally adapted research would be conducted according to a
"farming-systems" approach. Agricultural and home economics
extension staff and training would be integrated into a
single service. And lastly, land reform would assure that
producers' names, including women's are on title deeds.

Beyond these specific recommendations, other general
policy changes should be considered such as fairer prices to
farmers along with reduced or eliminated restrictions on
trade across district boundaries, both of which would
improve incentives for farmers. Women must also be
integrated politically, as individuals and in groups, to
learn about and receive the benefits or rural development.




Most Kenyans depend on agriculture and cattle
production for their livelihood. Through policy, the
government aims to supply information, inputs, and
incentives to farmers to increase their productivity. To
reach farmers, the agricultural ministry relies on a far-
reaching agricultural extension service to deliver
information and services to farmers. If extension fails to
reach its clientele in terms which enchance productivity,
then rural development does not proceed as effectively as it

The overall purpose of Kenyan agricultural policy is to
improve yields of cash and food crops and thereby secure
national self-sufficiency, more exchange earnings from
export crops, and improved livelihoods for rural people.
Kenya relies largely on the delivery of agricultural
services to serve these purposes. The various policy
objectives that flow from these purposes and services are
listed below, including both national and locally specific
concerns encompassed in the area of this case study.1

1. To assist all farmers to improve
agriculture through a field-oriented extension

2. To provide specialized knowledge to
farmers through formal training programs on
an equitable basis.

3. To provide production loans to farmers on
an equitable basis and, specifically, to
encourage cooperative membership and receipt
of cooperative loans.

4. To provide tractor services to Lirhembe
cooperative members in an equitable manner.

5. To provide grade cows to cooperative
members who apply for them.

6. To supply information to all farmers about
hybrid maize and cash crops.

7. To decrease farmers' risk through
adaptiveness and diversification.

In Kenya, as in much of Africa, women are the primary
cultivators, managers, or co-managers of farm operations.

Women are responsible for later stages of land preparation;
they also plant, weed,! and harvest crops. Men, on the other
hand, are responsible for early stages of land preparation
such as clearing or burning bush, for plowing, and for
cattle care. Yet despite women's extensive involvement in
agriculture,' Kenya shows the same tendency evident elsewhere
in Africa: I as has been noted in reference to Tanzania,
"governmental agricultural policy has been designed as if
women were not the majority of agricultural producers."2 On
the surface, however,, official agricultural policy appears
to be gender-neutral.1 Neither male preference, nor any
other type of preference on the basis of region, economic
status, or ethnicity, is officially encouraged.
I i f I
This study examines the impact of agricultural policy
on women in the Kakamega District in western Kenya in the
1970s. Before turning to look at the specific findings, it
is useful Ito briefly review the process of agricultural
implementation from the 1930s to the 1970s and thus obtain a
historical perspective on women's agricultural roles and the
origins of gender differentiation in agricultural policy.


The German anthropologist Gunter Wagner conducted an
ethnography in the Kakamega District among the Maragoli and
Bukusu subgroups of the Baluyia during the 1930s, providing
a historical benchmark against which to assess change.
Wagner related the descrpition of a Logoli elder (of the
Maragoli group) on the sexual division of labor, and adds
his comments:

It is the wife's work to sweep, to
grind, to cook, l to build the fires, and to
clean out the cattle partition. She carries
water from the spring, buys the cooking-pots,
and gathers firewood. She brings the salt
(i.e., she burns the salt reeds in special
pits and filters the ashes), cleans the walls
of the house and the surface of the yard with
cow dung, beats! the floor of the house (so
that it becomes hard and level), and 'knows'
about the food on, the food shelf.3

According to Wagner, "In addition to this list of
household duties, the wife performs the greater and more
strenous part of thQi garden labour, as hoeing is almost
entirely woman's wor ." It is of interest that field work
is missing from thed elder's description, although his
comment that women "know" about food may suggest some female
role in managing the fruits of her' labor. However, her
obligation is to maximize the family's interest through sale

or barter. Theoretically, a wife's profit belongs to her
husband, but in practice, says Wagner, for a husband to
inspect or question a woman's disposal of food was rare and
considered meddling. A husband's main duties are, according
to the elder,

to care for his stool and beer pipe, to
sweep his yard, to look after cattle, goats,
and sheep, to know the place of digging, to
cut the grass in the garden, to know when the
roof of the house gets deteriorated and to
pull the grass for thatching, to plait the
string (for tying the grass) and to hand the
grass (to the thatcher who is always a

Early government documents recognized women's
predominance in agriculture, as did ethnographies, but
reflected frustration in attempts to integrate men into
farming. According to district agricultural files from the
1930s, "man has not the agricultural lore of the women."4
Although token efforts to promote men's integration in
agriculture with a sparse field staff were made,
conscription, tax, and labor policy (the earliest of which
involved forced labor) pulled what were termed "able-bodied
men" from rural areas. Early preferential education for
boys, which later provided men with the skills needed for
occupational mobility, increased the likelihood that they
would leave rural areas.

To stimulate production for the war effort during World
War II, guaranteed high prices were offered for maize,
resulting in substantial production increases and the
emergence of the area as a major maize exporter.5 With the
absence of so many men during that time, it must be
recognized that it was women who responded to price

Until the 1950s Emergency, which signalled an end to
white settler farmer privileges, little stock was put in
creating commercial farming among Africans, excluded even
from growing certain cash crops such as coffee. The
Swynnerton Plan, published in 1954, aimed to intensify
agriculture through promoting cash cropping, mixed farming,
and food crops, in that order of priority, through land
reform, training, and extension. Secure individual titles,
it was thought, would provide farmers with incentives to
improve farms and with collateral for loans. In a sole
paragraph, the plan called for women instructors at Farmer
Training Centers (FTCs) to secure women's cooperation in
agricultural development.6 Land was consolidated and
registered largely in men's favor; officials neglected women
farmers' need for incentives to improve productivity,

4 I

security, and obtain loans.

The plan also laid the foundation for a "progressive
farmer" approach in which "able, rich, or energetic Africans
would be able to acquire more land and bad or poor farmers
less, creating a landed and landless class."7 Such an
approach meshes with the diffusion-of-innovation approach,
which calls for attention to the risk-taking, open-minded
minority that adopt' innovations before others.8 Both
Swynnerton and diffusion-on-innovation perspectives ignore
how political influence, wealth, and sex-separate
communication networks distort this seemingly neutral
process. Policy analysts affiliated with the University of
Nairobi Institute for Development Studies later criticized
the focus on the wealthy (some of whom were innovative,
while some were not) rith the least need.9

District archival documents reveal that agricultural
officials were frustrated that they received few requests
for agricultural assistance.10 Given only a limited number
of extension staff, who attended to those most likely to
complain if unserved (the politically influential and/or
wealthy), service distribution was probably quite skewed to
the so-called progressive farmer. Yet during the 1950s, in
conjunction with an e panded community development approach,
group extension, including women 's groups, reached its
height in Kakamega, thus diluting the service delivery skew.

Descriptions of agricultural assistants visiting
women's demonstration plots, women's groups, and women
attending District Agricultural Officer barazas (local
community groups) continued throughout documents of the
1950s. In Kakamega, documentation as early as 1951
described outreach to women's barazas using agricultural
assistants as i "propaganda agents"' to spread the use of
manure. 11 ; Both national and divisional documents reported
employment of female agricultural staff. Reinforcing this,
community development officers sponsored "homecraft
institutes" which included the distribution of vegetable
seeds, fruit trees, and talks on compost heaps. An
extensive treatise on the acuteness of agricultural
problems, attached o the 1955 District Annual Report,
openly discussed men s alienation from agriculture and the
need to teach women cash agriculture for their livelihood.
Annual and division reports in 1959, when land consolidation
efforts were very visible, noted that women's institutes
"switched to practical matters, including homecrafts, making
of dolls, land consolidation and enclosure, improved farm
methods, and the sale of bean seeds."

The most significant change after independence (1963)
was the development of hybrid maize at the Kitale
Agricultural Research Station and its extensive promotion

among farmers. Hybrid maize has the potential of raising
output per acre from the approximately 6 bags (540 lbs.)
achieved with local maize, to 35 bags, though in practice
the best of farmers achieve not more than 20 bags per acre.
Purchased seed, chemical fertilizer, and most importantly,
labor-intensive husbandry practices, such as more thinning,
weeding, and proper spacing, improved output. By the time
of this research, officials estimated that 80 percent of
Kakamega farmers used hybrid maize. However, despite the
new demands on women's labor and management skills imposed
by hybrid maize, women once again became invisible in
agricultural policy documents, and the absence of
agricultural outreach strategies to women is notable.

Beginning in the 1960s, passion fruit and tea were also
promoted. "European" dairy cows, which double milk
production with labor-intensive care, were introduced to
farmers with loan assistance through the Agricultural
Finance Corporation (AFC). Crops proved risky for farmers
to adopt. For example, passion fruit seedlings and special
loans for fertilizer and equipment were provided to farmers,
along with a promise to open a juice factory in the Kakamega
district once 500 acres were planted. By 1975 only half
that acreage was planted, and fruit was rotting on the trees
for lack of a market.

An extension-oriented approach continued after
independence, but services were specialized in a manner that
increasingly divided men and women into agriculture and home
economics. With help from the U.S. Agency for International
Development, a Home Economics Division was created in the
Ministry of Agriculture, and training in the Unites States
was provided for the top two tier officers along with a
special three-mouth course at Egerton College for third-tier
home economists. Normative conceptions about the desirable
place of farm women are evident in the U.S. advisor's

As land ownership and the desire to
utilize land for cash cropping becomes more
important in the African areas, it is
believed men will assume an increasingly more
responsible role in the actual day-to-day
operation of the farm. Thus farm women will
have an opportunity to spend more time with
the care of their children and home. 12

It appears then that Kenya's agricultural extension
approach in its initial and recent phases (at least up
through the seventies) was founded upon some implicit
assumptions about the extent and desirability of women's
roles as farmers. A number of studies, including the
present one, are finding these assumptions invalid. 13


The purpose of this study is to compare the access of
female-managed farms ito agricultural services with that of
jointly managed farms (which have a man present) to
determine whether there are inequities in the delivery of
agricultural services. The study also explores reasons for
bias in service delivery, focusing on historical
agricultural policy priorities, the interaction of class and
gender, the administration of! delivery, farmers'
performance, and women's demands -- both individually and
collectively -- for agricultural services. Finally, the
study examines the consequences of inequitable delivery for
women managers' agricultural performance.

Research Methodology

As noted above extension services are the major
vehicle through which rural farmers are encouraged to adopt
new crops landl more productive farming methods. These
services are offered t three levels of intensity: ordinary,
intensified, or saturated. In ordinary implementation,
agricultural field staff -- approximately one for every one
or two sublocations -- provide extension advice to
individuals in home visits and to groups in "demonstration
plots." The agricultural instructor, assigned to the
sublocationj level (the smallest administrative unit), has
the most frequent contact with farmers. He or she reports
to a technical assistant at the location level, who in turn
reports to a divisional officer. Though both men and women
are employed, 98; percent of the field staff are men. 14
Agricultural instructors also provide information about
credit and short courses at Farmer Training Centers.

Where ordinary services are intensified, line staff
share extension tasks with specialists, increasing the
likelihood that farmers will have contact with varied staff.
The extension plan also called for one home economics
officer per location, though in practice the ratio is often
far less (for example, in Kakamega District only 4 of the 17
locations had a home economics officer). In saturated
settings, not only is the ratio of staff to farm households
higher, but greater institutionalized support for innovation
and production is available. This includes services such as
credit, training in efficient crop techniques, assistance in
the introduction of inew crops (with the provision of free
maize seedlings, for example), labor-saving technologies and
technologies to assist in the intensification of farming,
and the establishment of:cooperatives.

The current study, based on field research in western
Kenya from 1974 to 11975, examines all three levels of
implementation. The intervening \years have seen some

changes which are not reflected in this study. However,
many of the basic policy, cultural, and environmental
conditions described above here still prevail not only in
western Kenya, but also in many parts of east and southern
Africa. This study seeks to offer new empirical information
and an analysis that might usefully inform agricultural
extension policy in general.

Survey data is drawn from 212 households in two
sublocations of Idakho, Kakamega District: Shikulu and
Shitoli (see Appendix and Note 24 for sample selection and
data analysis procedures). After a brief description of the
sublocations and of women's roles in the area, the study
looks at services implemented on an "ordinary" basis in
Shikulu to understand typical patterns of service delivery.
Then access in Shikulu is compared with access in Shitoli,
where services are intensified, to assess whether the
intensification of staff and resources affect disparities.
Finally, a subsample of Shitoli sublocation, consisting of
members of the Lirhembe multipurpose cooperative society, is
analyzed to determine the effects of saturating services on

Data on farm diversification also are compared for both
sublocations to assess the impact of inequitable service
delivery. This aspect of the study is based on a quasi-
experimental design. Two administrative units, similar in
culture, economy, and geography, but different in
administrative intervention, allow qualified judgments to be
made about programmatic causes of differential behavior.

Participant observation stems from residence in
Kakamega District for seven months, and more specifically in
Shikulu sublocation, where the Andala family graciously
welcomed me as an additional member. Finally, documents in
the Kenya government archives and secondary sources were
analyzed over the course of four months in Nairobi, where I
was an affiliate of the Institute for Development Studies at
the University of Nairobi.


The specific hypotheses tested in this research are
listed below:

1. The access of female-managed farms to
agricultural services is less than that of
jointly managed farms. (Subhypotheses to be
tested involve specific services such as
individual visits from extension officers,
attendance at demonstration plots, training,
and credit.)

2. The more valuable the service, the
greater the gaps |in service provision between
the two types of farms.

3. Intensification of services improves
female managers' access to services relative
to female managers elsewhere, but gaps
between them and jointly managed farms remain
the same.

4. Female managers' farm behavior has no
effect on the delivery of services.

5. Male preference in agricultural service
delivery reduces women managers' agricultural

6. Women's labor groups which share
information and labor offset systematic
biases in the delivery of services under non-
landscarce conditions.

7. Female managed farms make fewer demands
for agricultural services than jointly
managed farms, both individually and

8. Women's exclusion from the political
mainstream reduces government accountability
to women and consequently lessens commitments
to deliver services equitably to women.

9. Gender of manager surpasses economic
status or land size as a cause for service
delivery differentiation.


The District

The Baluyia people, a densely settled agricultural
people who form the third largest ethnic group in Kenya, pop-
ulate Kakamega District, western Kenya. They are comprised
of relatively autonomous subgroups, each containing a loose
collection of patrillineal clans, but they share both a
common Bantu language, Luluyia (with some dialectical
variations), and similar customs. The Idakho people,
organized into 30 clais, are one such subgroup, located in
the south central portion of the district. Idakho location
contains nine sublocations, including Shikulu in the west
and Shitoli in the east. The largest clan, the Bashimuli,
resides in Shitoli; Ithey have provided leadership for the

location for nearly a half century, from Chief Sivachi to
his son, Chief William.

With a population numbering three-quarters of a
million, Kakamega is the second most densely settled
district in Kenya, with 220 persons per square
kilometer. 15 The density is partially responsible for the
many men who migrate to large farms in the White Highlands
and to urban areas in pursuit of wage employment. Male out-
migration was 30 percent in 1937 and reportedly doubled by
the 1950s. Since the post-World War II era, Kakamega
district has been the largest labor-exporting area in Kenya,
with the most uneven sex ratio. 16

The government labels Kakamega District a "high
potential" agricultural area, given its annual rainfall of
70 inches per year, which permits two growing seasons.
Maize and beans, the most common staple food crops, are
grown along with millet, sorghum, root crops, and cowpeas.
Virtually all farmers keep some chickens, used for feeding
visitors. Only in recent decades have women started eating
eggs, thought earlier to cause barrenness.


Ten miles from one another, the two sublocations of
Shikulu and Shitoli are similar in most agricultural
respects. Both areas are populated by the Idakho subgroup
of the Luyia people. Climate, rain, and soil features are
similar, and the same proportion of men work away from home,
leaving women to manage the farms. Both areas had the same
mean economic standing, based on a five-point scale of
material consumption items. 17 The average farm size was
2-1/2 acres, though it was smaller in Shitoli (1 1/2) and
larger in Shikulu (3). Approximately half of the farm
households owned a cow, but low cash levels and limited
grazing land limit cattle holding. Although coffee has been
a cash crop since Kenyan independence in 1963, it was grown
by only 15 percent of the farmers in the study sample.

Finally, both sublocations were subject to similar
mission influences, and schools were introduced in similar
time periods. Approximately half the population in each
sublocation identifies itself as Catholic, with the
remainder being Protestant.

The key differences, therefore, relate to the amount of
agricultural services available.

Shikulu: Ordinary Services

Shikulu is the less well served area, receiving
ordinary services. One agricultural instructor served
approximately 2,000 farm households in a large territorial
unit of 39 Isquare kilometers. Recently transferred into
that sublocation (Eransfer rates are high), he was
nevertheless from Iddkho location and thus spoke the local
dialect. He had nd bicycle, but relied on occasional
vehicle traffic or the two buses which traveled each morning
and afternoon to and from the tarmac road. Like the author,
he spent much time traveling on foot. Shikulu is isolated
from the administrative mainstream, being the furthest
sublocation from the location center, division headquarters,
and tarmac road from Kisumu. The nearest FTC, at Bukura,
was located in the neighboring location, about ten miles

Shitoli: Intensified and Saturated Services

In great contrast to Shikulu and to most sublocations
in the district, Shitoli had the equivalent of three
agricultural instructors, who served approximately 1,500
farm households. Its' smaller territorial unit of 9 square
kilometers is more densely settled than Shikulu (714 per
square kilometer, compared with Shijulu's 225).
Administrative loffic s are located on Shitoli's boundary;
this has led historically to closer administrative contact
between Shitoli and the government.

In addition, with assistance from the local Member of
Parliament (M.P.), who is from the area, the Danish-
supported Lirhembe multipurpose cooperative society was
established in Shitoli. Cooperatives in Kenya as in other
settings are established as an instrument for integrated
small farmer development. They serve other secondary
functions such as political mobilization repayment of
political debt, and they also serve to establish a
government presence inl remote area. Whether these additional
functions are viewed as appropriate or problematic, the
aspect of primary importance to this study is the provision
of a structure through which services can be delivered. For
example, members had access to easy maize seed and
fertilizer loans, to al tractor, and to high grade cows. (A
second cooperative, for coffee, also existed in the Shitoli
area, but was not examined in this study.)

Thus the cooperative structure in Lehrembe resulted in
a saturation of services.


The data in this section are drawn from historical
documents, other published research, and the current
findings. It is intended to provide a brief view of how
women operate in the research area.

The Farm Unit

As a result of the changes in the economy outlined
earlier, men now migrate as a normal part of their adult
life cycle. Thus households, for the purpose of this study,
have been divided into female- and jointly managed farms.
The women in female managed farms were either widowed or
separated from their husbands who had migrated outside the
area for wage employment. Jointly managed farms consisted
of husband-wife teams, or corporately managed farms with
intergenerational management because land had not yet been
parcelled out to sons. Forty percent of the 212 sample
households consisted of female managers, a figure that
corresponds very closely to the 36 percent for the
district. 18

Women with absent husbands faced several types of
circumstances. Some had been all but deserted by their
husbands, who permitted them to use the land and who
retained patrilineal claims over children. Though financial
assistance is difficult to calculate reliably, nearly one-
third of these married women with absent husbands appeared
to be getting no financial aid from husbands. 19 Still
another group of women were widows who were self-sufficient,
but were often assisted by children. These women had an
added, nontangible benefit of the prestige that comes with

In the sample farms with a man present (60 percent),
four distinct patterns were evident in nearly equal
proportions. Some husbands were locally employed in
government or owned a shop. These men may have aided in
farm decision making, or transmitted information from male
communication networks and contacts with political and
agricultural influentials, but they did little actual farm
work. Others had retired from wage employment, but they
arbitrated local conflicts and did not do farm work. A
third group consisted of men who had developed an interest
in farming; some had even purchased plows. These men may
have been genuinely interested in farming or simply
reconciled to farm work, given the poor prospects for wage
labor. A fourth group of men showed no interest in farming.
Often on small farms, there simply is not enough work to
keep two persons occupied, except during peak labor periods.
On these farms, men appeared to be infected with a sense of

malaise and discontent about chances for cash-related upward
mobility. A signifi ant number of small farms with a man
present are among the poorest.

Many female managers absorbed men's responsibilities
and coped with peak labor periods without the assistance of
another resident adult. In the sample, half the female
managers were responsible for cattle care. Also, many hired
laborers and negotiated to have their land plowed prior to
planting season. FoUr women were even managing a plowing
business their husbands had begun. Men, by and large,
"worked with money" if they could; that is, they sent money
for oxen plowing and commercial seeds on which farmers now
depend. Yet most husbands worked in low-paying jobs, as
night watchmen and domestics for example, and their small
salaries were quickly spent.

The patterns out ined suggest that female-managed and
jointly managed farms stagger alli levels of the local
socioeconomic hierarchy, measured by income and land.
Fifty-eight percent of !women farm managers and 54 percent of
farms with a man present had a low economic standing;
similar proportions hold small acreage. Women farm managers
inflate the low acreage category slightly because widows are
part of the group.

Decision Making

Women frequently make many of the agricultural
decisions about husbandry and storage practices, crops
grown, and Itime of planting; most of these agricultural
activities are within the women's realm, even in farms with
a man present. For example, in response to questions about
when hybrid maize was grown or what practices are used, some
husbands openly deferred to their wives. In a subsample of
107 farmers questioned about initial decision making
connected with hybrid maize, 34 percent of the decisions
were made by women, l percent by men, and 28 percent by
both spouses (in 7 percent of the cases, a son made the

Women without independent decision-making
responsibilities were generally young and insecure.
Contacts and influence in this society are solidified by
blood and clan ties. Women, as in-marrying strangers, do
not have those kinds of ties. After years of child-bearing
and residence in her husband's clan, a woman is perceived to
be somewhat! linked to that clan, but for all practical
purposes (and as demonstrated in the exogamy rules for her
children's marriages) lshe is not a member of her husband's
clan and never will be.

Control Over Income and Income Responsibility

Women have the traditional, though informal, right to
dispose of their produce. Male absence enhances this
control over produce. Thus, women have some access to
money, but in a context of ever-constricting land size, as
land is divided among sons. That, combined with large
families that must be fed, means that the wisest strategy
often is not to sell the family's primary source of food.
Women obtain money from trading, brewing beer, selling
produce, and from their working husbands. Only 2 percent of
the women in the sample earned a salary from government or
wage employment, such as teaching. Fourteen percent of the
women regularly brewed beer (busaa). Regular trading
(buying and reselling) was an activity of only 6 percent of
the women in the sample. Female traders were aware of the
disparities in access to cash between them and male
shopowners, who can acquire government loans. As one said,
"Men have thousands (referring to shillings); women have
hundreds." 20

Women, thus, have considerably less access to and
control over income than men. Almost all men have worked
for wage employment at one time or another. Nor can it be
assumed that family income is shared. Men rarely contribute
more than occasional meat or milk for day-to-day household
maintenance, and are not expected to do more. Household
maintenance is women's financial responsibility. This
includes providing for the basic needs of the family,
including food, water, and clothing. Men are expected,
however, to pay children's school fees and finance the plow

Access to Land and Credit

Patrilineal rules of inheritance still confine women's
access to wealth. Land consolidation and registration,
begun in the late colonial era, were completed for most of
Idakho in the early 1970s. Although in earlier eras land
had been managed and passed patrilineally, the traditional
process had nonetheless guaranteed secure access to women
cultivators. Land titles, now held largely in men's names,
are alienable; a full fifth of the sample claim land
changes. Women now work in a tenant-like relationship to
land, as no concept of joint tenancy exists. Elsewhere in
Nyanza Province among the Luo, a study found women owning
only 6 percent of land parcels after registration. 21

With no control over land, women lack collateral for
loans. Therefore, if a woman wants a loan, her husband must
put up a guarantee in the form of a title deed or salary,
which can be difficult to arrange if husbands are not
regularly at home, or are not interested in the farm. The

only genuinely independent women for loan purposes are
widows, who can arrange legally, through a complicated and
costly process, to have the title deed transferred to their
names. Putting; title deeds in male names has solidified
male control over powerful resources,i and this has obvious
implications for; acquiring other resources as well.

A man can obtain a loan without his wife's awareness,
yet loan default and subsequent land auction mean women have
no access to land tolproduce for their children or for a
livelihood. In the area, an elite woman widow was
threatened with foreclosure when she became liable for an
agricultural! loan her late husband acquired (without her
knowledge) and squandered for nonagricultural purposes.
This kind of incident occurs with enough frequency to be
cited by women as a major source of anxiety. 22

Organizational Activity, Political Authority, and Access to Agricultural Services

Organizational activity among women has flourished.
Over 90 percent of the women in the sample belong to one or
more groups. All t ese groups, complete with names and
officers, are recognized by local residents. Women continue
the tradition of collective labor in agriculture, in
buhasio groups, where members work for a fee, divided at
year's end, and for food and tea with lots of sugar. Women
also belong to bulala groups for mutual aid in times of
crisis; the aid is collected in kind or in cash. Women, who
dominate church;membership in rural areas, form groups for
church fund raising or for praying. !While men are part of
other networks,I sit on government Icommittees, and hear
announcements, few! are sustained members of groups like
women. !

On the whole, though, women's organizational activity
is not integrated into the political mainstream. During
elections, women's groups or their leaders support
candidates with promises of votes in exchange for short-term
benefits, such as matches or packets of sugar. But women
are neither strategically located for influencing local
administrators nor included in information dissemination
networks. Formal political authority and communication
networks emanating therefrom are largely men's affairs.
Women are rarely in positions of authority, except to
represent women. Furthermore, in communities like these,
which have undergone substantial economic differentiation, a
woman in formal authority is usually of elite status and
often has different int rests, perceptions, and priorities
from other women.i
\ ', ;


Because of agriculture's importance in the local
economy, agricultural services, credit, and staff contacts
are central to power in Idakho. The agricultural
instructors and technical assistants frequently meet and
advise local administrators as well as meet weekly with the
subchief and chief at barazas (local meetings). Thus, the
agricultural field staff belong to a communication network
that dispenses information and material resources,
subjecting them to political contact and pressure. In these
networks, information about loans and training is passed,
and valuable contacts are made.



This study, like others in the series, presents
findings in relation to policy objectives. The following
section analyzes the implementation of Kenya's agricultural
policy objectives as outlined in Part I, but presented here
in a different order. Objectives relevant to ordinary
implementation in Shikulu sublocation will be examined
first. The effects of intensification and saturation in
Shitoli sublocation will be examined next. To obtain a
broader picture of the effects of saturation, the impact of
objectives which wer relevant only, to a saturated setting
will be considered as well. Finally, a comparative analysis
of women's performance in all three settings will be
undertaken, and considered in relation to the broad
objective of increasing adaptiveness and diversification.



As indicated ear ier, in Kenya extension agents deliver
and adapt information about agricultural research to farmers
in their homes and at demonstrations on local farmers'
plots. Also knowledgeable about training and credit, the
agent can link these beneficial services to farmers. The
government aims to have one agent per sublocation, although
in practice one may serve two or three sublocations.
Extension agents foous on hybrid maize and its associated
husbandry practices, vand on cash crops, such as sugar cane,
coffee, European vegetables, and passion fruit, and
encourage crop diversification so that farmers spread their
risks, should one or several crops fail.

Home extension visits allow individual, personal, and
sustained contact on he farm itself, where recommendations
can be made to suit farm conditions. The disadvantage of
home visits is that they are costly and rife with
discrimination potential. It is also difficult to monitor
the quantity of indi idual home visits, as field officers
operate with great freedom. Studies in Kenya have found
that extension officers visit an average of only 4-5
households iper week. 23 These studies also show that a
quarter to half of the extension officers were misinformed
on crops about which they extend information, and that
officers work in an administrative structure with few
incentives for improving or making their performance
equitable. Some farmers in the current sample spoke

disparagingly of the minimal advice agents had to offer or
of contradictory instructions they received.

Group extension, through lectures at demonstration
plots on local farmers' plots (including those of some elite
women farm managers), offsets the tendency to skew visits to
few farmers by opening up services to all. Demonstration
plots can be educational in practical and experiential ways,
because farmers visualize new tasks, such as how to apply
fertilizer, and can then later view progress on the local
plot. Agricultural instructors are more likely to prepare
for the presentation they make, and such events are easily
monitored (perhaps accounting for the better preparation).
Yet demonstration plots in Western Kenya are few and far
between, perhaps once a growing season in a sublocation,
partly due to the preparation time association with
acquiring a plot to use, finding inputs, and publicizing the
event through announcements in barazas. Attendance is based
on the initiative of the farmer. In the past farmers were
often compelled to attend.

Historically, compulsion was used in other ways as
well: seedlings not planted in lines would be uprooted,
farmers would be fined for improper coffee care, or could
face compulsory labor. As part of efforts to eradicate soil
erosion, farmers were often forced to work on bench
terracing. An ongoing resentment and suspicion of agents
was sometimes reflected in farmers' remarks.

Home Visits

Farmers were asked whether their farms had ever been
visited by an agricultural instructor. Their responses,
presented in Table 1, show that female-managed farms were
significantly less well served than jointly managed farms.
About half the female-managed farms had never been visited,
in contrast to only a quarter of the jointly managed farms.

Table 1. Visits by Agricultural Instructor, by Farm Management Type 24

Female-Managed Jointly Managed

Farm never visited 49%/42 28%/36

Farm visited once 51%/43 72%/91
or more

(Yules Q=.42; Significance=.01) N=212


How did women! fare in jointly managed farms? Because
many husbands were not involved in farm work and not at home
during the day, often women were in fact alone at the farm
when agricultural instructors visited, and agricultural
instructors, generally spoke to "whomever was there." When
the husband was present during a visit communication
patterns were diverse; depending, as one might expect, on
the personality styles of the couple and the agricultural
instructor. While te general norm was for the husband to
speak and to represent the household, this did not
necessarily mean that Ithe woman left or stayed silent; thus
women in jointly managed farms may have more frequent and
direct contact withi agricultural instructors than women
managing farms alone.

It is difficult to determine, however, whether or not
the quality, intensity, and duration of the communication
between an instructor and lone woman, one whose husband is
away for the day, compared favorably with that between an
instructor and husband. This probably also varied with
personal style. It is also important to bear in mind that
many middle-aged and older women have a strong sense of
personal efficacy and considerable prestige derived from
their reputation as hardworking farmers and mothers of many
children. Thus, with increasingi age, the potentially
problematic nature of communication between women and
instructors becomes l ss important.

Demonstration Plots

As Table 2 indicates, fewer farmers have attended plots
than have received 'visits, but the disparities between
management types were less marked than in visits.

Table 2. Demonstration Plot Attendance by Farm Management Type

Female-managed Jointly managed
No. household member 62%/52 46%/59
1+ household members 38%/32 54%/68
attended i

(Yules Q=.31; Significance=.05) N=211

Lower attendance rates for woman managers could mean
that they were uninterested in, unaware of, or unable to
utilize this service. In this context, it becomes relevant
that demonstrations were usually announced at the weekly

barazas, posing a problem for women who rarely attend those
because of time constraints and custom (it has traditionally
been the man's responsibility to attend). Barazas last
several hours as announcements are made and local disputes
adjudicated. The percentage of women who can spare such
time rarely exceeds 10 percent.

Among jointly managed farms, equal numbers of men and
women attended demonstrations, the explanation for which is
twofold. First, male presence brings a farm directly into a
valued male communication network. Second, the presence of
two or more adult laborers on a farm permits the release of
one or the other from farm and child-care activities to
attend the meeting.


Another service, training, was provided at farmer
training centers, which hold short courses that usually last
one to two weeks. Though highly subsidized by the
government, a fee of shs. 10/50 (approximately $1.50) is
required, and this is a sizeable sum for farmers without a
regular cash income. 25 Training represents a direct and
intensive service for farmers, as it lasts longer than other
services and makes available highly qualified teachers.
National and District Reports since independence indicate
that a third of all FTC participants are women, but they are
virtually always tracked into home economics courses
covering a wide scope of topics, with at best a third of
course attention relevant to agriculture.

Bukura Farmer Training Center is a 5-15 mile trip from
farms in both sub-locations sampled, and has been open since
1923. As the results in Table 3 reveal, completing a course
at the farmer training center is rare, a surprising finding
given Bukura's proximity. But a full third of the farmers
were unaware that the training center existed. Aside from
that, great disparities exist, with jointly managed
households being four times more likely to have a member
trained than female-managed households.

Table 3. Training by Farm Management Type

Female-managed Jointly managed
None in household 95%/80 80%/102

Someone in household 5%/4 20%/25
(Yules Q=.66; Significance=.01) N=211

For women managing farms alone, a one- to two-week
training period presents special problems. They must
arrange for day-tolday household responsibilities and
cultivation to be provided for while they are away. In the
84 female managed farms in the sample, only four women had
ever had any trainingL

Chiefs and assistant chiefs help agricultural
instructors! recruit and select trainees through
announcements at baiazas or through word of mouth. Many
farmers perceive that attendance at Bukura is an honor,
achieved by invitation only. The awareness of and
opportunity to attend training is thus more closely linked
to the local power structure than the first two services

In addition, jointly managed farms, husbands were often
wary about wives being away for extended periods of time,
and, in some cases, the chief or assistant chief had to
persuade husbands to allow their wives to attend. Of the 25
jointly managed farms where some member had attended a
training course, six of those trained were women, and the
remainder were husbands or sons.

But it! is noteworthy that in ainumber of cases where
sons were trained, they had since moved away from the farm
and found \non-agricultural employment. In several cases
where husbands were trained, the men were old and, for all
practical purposes, retired from active farms work; their
invitation to be trailed simply conferred status.


The earliest type of loan available to respondents was
the Guaranteed'Minim m Return (GMR) Iadvance for maize seed
and fertilizer, but the minimum acreage required for maize
was beyond the holdings of most ordinary farmers in the
location. Agricultural Finance Corporation (AFC) loans
became available after land reform was completed in 1973,
when farmers could acquire individual land title deeds.
Most lending benefits the large farm sector. Kenya's
official plan indicates that 1 percent of smallholders
receive these types of loans. 26

Loan application and acquisition procedures integrate
farmers more fully into the extension service. In addition,
the district-level personnel involved in administering loans
were of higher caliber than the ordinary agricultural

Table 4 shows whether or not respondents knew about
loans and loan procedures, or had actually received a loan,
by farm management type. In light of the above information,
it is not surprising that ordinary farmers acquired only
three loans, and these went to jointly managed farms. An
additional 7 percent (16 households) could correctly relate
the application process. Only one of the 19 households that
knew anything about loans was female-managed, and that
particular woman was not only wealthy, but was linked to the
local power structure by participation on the development
committee, marriage to another committee participant (though
he was absent), and an in-law relationship to the local

Table 4. Loan Knowledge and Acquisition by Farm Management Type

Female-managed Jointly managed

Knew nothing about 99%/83 86%/109

Knew application 1%/1 12%/15
process or had
applied for GMR/AFC

Acquired either GMR -- 2%/3
or AFC loan

(Q=.86; Significance=.01) N=211


Because of the special husbandry practices required to
achieve high output from hybrid maize, the widespread
acceptance of hybrid maize implies an immense communication
task for agricultural instructors. These practices are not
something farmers would adopt on the basis of experience
alone: they involve detailed expertise on spacing,
thinning, planting time, weeding size, and techniques for
applying fertilizer.

When farmers who grew hybrid maize were questioned about
their initial source of information about the crop, the
results showed that communication channels for innovation
are gender-specific. Almost half of the respondents from
jointly managed farms heard about hybrid maize directly from
technical sources, which was true for only a quarter of


female-managed farms Female farm managers heard about
maize from neighb rs' in both formal and informal
communication network s, the latter consisting mainly of
communal agricultural labor groups.

Table 5. Sources of Initial Information about Hybrid Maize by Farm
Management Type (Hybrid Maize Growers Only)

I I Female-managed Jointly managed

Agricultural staff 25%/21 44%/51

Neighbors 43%/36 33%/38

Market/stockists 25%/21 9%/10

Other (employed outside;
radio) 7%/6 15%/17

(Gamma= -.23;
significance = .001) N=200

Women managers also heard about new crops from the
market area, which they visit several times a week to sell
produce, to grind maize for cooking, or to purchase
household goods, land in particular from stockistss," the
shop-keepers who self agricultural inputs such as seed and
fertilizer. Also ai the market are further extensions of
informal women's networks and groups, such as those that
exist among women traders. (Traders often buy packets of
seed and sell them in small portions to buyers lacking
enough cash for a fdll bag of seed.) Such networks are
useful in Ithe diffuse on of agricultural innovations, but
shopkeepers located far in the interior purchase innovative
inputs well |after thedr initial introduction in an area.
Hybrid maize farmers were also asked about the husbandry
practices they used and, when they responded with some
proper technical practices (as did most farmers), the source
of information was requested.

In this case, too, female managers received considerably
less information from agricultural staff than did men. Less
than half of the fe ale managers had direct access to
technical information rom those trained in agriculture, in
contrast to the almost three-fourths of persons from jointly
managed farms with suc access. Most female managers found
out about hybrid maize practices from second-hand sources.

Table 6. Sources of Information About Hybrid Maize Husbandry Practices by
Farm Management Type (Hybrid Maize Growers Only)

Female-managed Jointly managed

Agricultural Personnel 43%/35 72%/83

Non-Agricultural 57%/46 28%/32

(Yules Q= -.55; N=196
Significance = .0001)


What are the effects of intensified services on
equitable administrative performance? While increasing
resources is not a sufficient condition to guarantee more
effective services, it has been argued that in theory, at
least, administrative performance could improve in both
quality and quantity with increased resources for staff and
support services. Many problems in administrative
performance are said to result from financial constraints
under which development administrators operate. Low
budgets, staff shortages, insufficient transport vehicles,
and the like are given as explanations for the failure of
administrators to service clientele adequately. 27
Moreover, bureaucratic innovation is predicted with
increased resources such as time, finance, material, or
skills. 28 Increased resources can supposedly enhance
innovation by expanding an administrative clientele (to
include, for example, women) without depriving those
clientele already served.

The section below compares access in Shikulu and Shitoli
in relation to the four objectives outlined above and
illustrates some of the limitations of this approach.

As Table 7 indicates, both female-managed farms and
jointly managed farms in Shitoli generally received more
extension visits than all farms in Shikulu. However,
disparities still remained between farm management types in
the intensified Shitoli area, although the gaps were
narrowed somewhat. Furthermore, it is noteworthy that
intensification seems to merely raise the access level of
female managed farms to that of jointly managed farms in

typically served areas.

Table 7. The Impact of Intensification on Extension Visits by Farm
Management Type

Female- Jointly Female- Jointly
managed managed managed managed
Farm never visited ,61%/28 37%/23 36%/14 20%/13

Farm visited 39%/18 63%/39 64%/25 80%/52
1+ times
N=108 N=104
Q=/45 Q=.38
Tau B=.24 Tau B=.18
at the .05

In the case of access to demonstration plots available
data show that int nsification of services did not
substantially affect the gap between the farm types, as
Table 8 demonstrates.

Table 8. The Impact of Intensification on Access to Demonstration Plots

Female- Jointly Female- Jointly
managed managed managed managed
No household member
attended 72%/33 56%/35 50%/19 37%/24

1+ household member
attended 28%/13 44%/27 50%/19 63%/41
Q=.32 Q=.26
Tau B=.16 Tau B=.13



The existence of more staff in Shitoli should increase
the likelihood that farmers hear about training. However,
as Table 9 shows, overall only a slightly larger number of
farmers (both in female-managed and jointly managed farms)
were exposed to training with intensified services. Gaps
between the sexes narrowed slightly, from 16 to 14
percentage points. But here the Shitoli female managers did
not do much better than the Shikulu joint managers,
suggesting the importance of communications and women's
mobility constraints for this service.

Table 9. The Impact of Intensification on Training


Female- Jointly Female- Jointly
managed managed managed managed

None in household
trained 98%/45 82%/51 92%/35 78%/51

Someone in household
trained 2%/1 18%/11 8%/3 22%/14

Q=.81 Q=.52
Tau B=.24 Tau B=.18
at the .05 level)


The convenience of more and nearby specialist staff in
Shitoli should increase farmers' access to loans under the
conditions of intensified services, but population density
and consequent small farm sizes in Shitoli offset these
prospects. The minimum floor of GMR crop insurance loans is
far too high for even relatively large landowners in Shitoli
to qualify, and AFC loans were commonly promoted for "grade
cows" which require grazing areas of at least an acre
planted for fodder grass. Thus, as Table 10 indicates,
neither region seems to have been very informed about loans,
and increased services had less of an impact than might be

Table 10. The Impact of Intersification on Knowledge About and Acquisition of


Female- Jointly Female- Jointly
managed managed managed managed

Knew nothing 100 /46 81%/50 97%/37 91%/59
about loans

Knew loan 16%/10 3% 7%/5
process or had
applied for
GMR/AFC loans

Acquired loan 3%/2 -- 2%/1

S Q=1.0 Q=.58
i iau B=.30 Tau B=.13
(Significant at
the .05 level)

As Table 11 shows, in both sublocations same-sex
communication is still a common vehicle for information
transmission'. Unexpectedly, a larger proportion of Shitoli
jointly managed farms, a proportion nearly comparable to
that among female managers, cited neighbors as an /initial/
source of information.

Table 11. Initial Source of Information about Hybrid Maize by Sublocation and
Farm Management pe

Female- Jointly
managed managed

Female- Jointly
managed managed

Agricultural staff





42%/ 19







31%/12 45%/26

44%/17 41%/24



5%/2 10%/6

*Significant at the .01 level.
**Significant at the .10 level. N=103



As shown in Table 12, the source of information for
hybrid maize practices however, continues to vary according
to management type. Though far more women in highly
serviced sublocations learn practices from agricultural
sources, the proportion was just close to the proportion of
jointly managed farm representatives in an ordinarily served
sublocation. Even in Shitoli, many women heard about
practices from their neighbors, members of women's groups,
and traders.

Table 12. Source of Information about Hybrid Maize Practices by Sublocation
and Farm Management Type


Female- Jointly Female- Jointly
managed managed managed managed

Agricultural staff 30%/13 61%/35 58%/22 83%/48

Non-agricultural 70%/30 39%/22 42%/16 17%/10

N=100 N=96

Yules Q=-.57 -.55
Tau B=-.31 -.27
Significance=.01 .01


At this point, data from the Lirhembe Cooperative, an
even more intensely served subsample, are analyzed to
determine whether the saturation of services offsets usual
biases against women managers and addresses their special
difficulties in densely settled Shitoli with its food
shortages, dependency on cash, and burdensome work setting.
High land density in this area makes it difficult for women
to grow the quantity of food necessary for feeding their
families without cash support.


Lirhembe's provision of easy loans, grade cows, and
tractor services to its members should address farmers'
needs. First, however, women managers must enroll as
cooperative members or convince generally absent husbands to
do so. Thus, before examining female manager's access to
loans, it is necessary to explore their propensity to join

the cooperative.

Though female membership in the cooperative is not
prohibited, managers assumed that male "heads" would make
such decisions and sig* for the family, and no recruitment
effort was directed at women managing farms alone.

Table 13. Lirhembe Cooperative Membership by Farm Management Type

Female-managed Jointly managed

Member 46%/18 72%/47

Non-member 54%/21 28%/18

(Yules Q= -.51; Significance=.01) N= 104

The figures for cooperative membership, displayed in
Table 13, are striking. While almost three-fourths of the
jointly managed farms n the Shitoli sample were registered
at the society, less than half of female-managed farms were
registered. In addition, in the subsample of 65 cooperative
members, only 28 percent (or 18 farm households) are female
managed. All the known females registered who comprise only
7 percent of members, were widows.


Cooperative maize seed and fertilizer loans were easy to
acquire, compared to1 the government AFC loans. All one
needed to do was join the cooperative, sign for the
agricultural inputs, and agree to repay either in maize or
in cash, presumably at the end of the growing season. No
title deeds were required. In fact,; people avoided payment
at the end of various seasons, and declining numbers had
repaid each year since the loan was offered. In 1975, no
capital was available to purchase iinputs to make loans
available. The populace was massively indebted, but the
indebtedness lacked any penalties. Terms of repayment were
widely understood at the time of interviews (the fourth year
of the program), but members perceived that the society was
"not strict" about repayment. Little or no selectivity was
involved in the provision of this loan.

Table 14. Use of Maize Seed and Fertilizer Loans in Lirhembe Cooperative by
Farm Management Type

Female-managed Jointly managed

Used loan 78%/14 87%/41

Did not use loan 22%/4 13%/6

(Yules Q= -.32) N=65

As Table 14 shows, jointly managed farms were still more
likely to use loans than female-managed farms. Perhaps the
long-standing custom for husbands to represent households
and negotiate transactions outside the home explains the
difference. Nonetheless, it was possible for women to sign
for the agricultural inputs, even when membership was in
their husband's name. In examining Society ledgers,
approximately 20 percent of the signatures for inputs were
women's names, while the number of women members, as
indicated above, was only 7 percent of the total. Still,
fewer women's signatures were found than expected, perhaps
partly because of the long-standing custom among women, by
choice or otherwise, of operating autonomously from
government assistance.


Another concrete benefit available to farmers was a
tractor. Tractors could alleviate burdensome hand digging
of the soil for cultivation, which is always women's work.
Given widespread perceptions that this service was limited
to persons with political influence--since committees rather
than the co-op clerk made decisions--people were pessimistic
about its accessibility. Table 15 shows the distribution of
tractor services by farm-management type.

Table 15. Cooperative Tractor use by Farm Management Type

Female-managed Jointly managed

Tractor used 29%/5 23%/11

Tractor not used 71%/12 77%/36


Somewhat surprisingly, similar proportions of female-
managed farms and jointly managed farms received tractor


services. Yet all in ll, only a quarter or so of the farms
in the subsample were, serviced. Contrary to findings on
maize and fertilizer usage, data on tractor service
indicated that female-managed farms. In fact, the slightly
higher proportion could be accounted for by this service
being particularly in tune with women's needs. The tractor
was available for two ears, until its mechanical failure,
at which time no capital was available for repair.

One ofithe most valuable services offered to farms has
been high-grade cows, which produce twice as much milk as
local cows.! At Lirhembe, grade cows were "borrowed" and
repaid with young heif rs which were Ithen borrowed by other
cooperative members on the same basis'. Plans had been made
to acquire !a milk c oler for the central collection and
marketing of milk. Milk sales were also to be used to repay
the cost of barbed wire fences loaned to farmers who
received grade cows, a requirement for this service. Funds
for this cooler were not available, however, in 1975.

Program impact had been genuinely redistributive. Many
poor farmers obtained h cow; these farmers would never have
accumulated the cash to purchase such a costly animal. 29
And as the previous section demonstrated, for loan purposes
the regularly extension service had virtually written off
farmers who did have sufficient, albeit small-sized, land.

Yet several requirements of the application process
have skewed receipt in favor of farmers with more land. To
receive a cow, farmers had to make available 1 1/4 acres for
grazing and planting f dder grass. Besides that, farms had
to have at least one acre for maize,1 in addition to space
for their house, meaning that only farmers with 2 1/2 acres
or more could obtain the grade cow loan, a size well above
the 1 1/2 average acreage.

Furthermore, given the high value of a grade cow and
the committee selection process used to determine
eligibility for th6 cow loan, the potential for
discrimination was high, be it political, economic, or
gender-based;. Not eno gh cows were available for the entire
membership. Influence and contacts were thus essential for
pressing or "reminding committee members of an application.
This is generally considered men's activity, particularly
because it is largely men who sit on the committee.

; ^

Table 16. Cooperative Grade Cow Applications by Farm Management Type

Female-managed Jointly managed

Applied; received 22%/4 47%/21

Applied; did not 39%/7 17%/8
receive cow

Did not apply 39%/7 38%/18
N= 65

As Table 16 indicates, although three-fifths of both
female-managed farms and jointly managed farms applied for
cows, a manager of the joint type was twice as likely as a
female manager to actually receive a cow. Female managers
were not disproportionately concentrated on smaller land
parcels, thus eliminating economic factors as the cause of


The documentation of thoroughgoing inequities in the
delivery of agricultural services is made more meaningful
when linked to farmers' performance. If discrimination had
no impact, then reforming or transforming agricultural
administration would be beside the point. This section
reveals not only that discrimination does have detrimental
impacts, but that relevant yet untapped women's groups are
vehicles for encouraging information dissemination, risk-
taking, and labor sharing.

Policy impact analysis attempts to determine whether
policies produce intended (or unintended) effects on a
populace, but it is difficult to sort out policy from other
effects. Ideally, impact studies are longitudinal,
measuring baseline characteristics of a sample prior to the
intrustion of a policy change and then after a designated
time period, using the same instrument to measure change.
In the absence of longitudinal data, controlled comparisons
between two virtually identical administrative units -- one
affected by the policy and the other not--permit qualified
generalizations about policy impact.

Using a controlled comparison of Shikulu and Shitoli,
this section tests the hypothesis that the performance of
female managers relative to joint managers is declining
under conditions of intensified, male-oriented agricultural

services. The similar ty of both sublocations, alike in all
ways except for the intensification of services and land
scarcity, compensates for the lack of longitudinal data.

To further compensate for the non-longitudinal design
of the study, farmers were asked factually oriented
questions about their past and present agricultural
practices. Unless farmers sell maize in larger quantities
all at once, they have little sense of output per acre in
unit measures; much food, in fact, lis stored after harvest
in granaries. Therefore, performance measures used included
crop diversification and timely adaptiveness to hybrid maize
innovations.! The more diversified the crops, the greater
the spread of farmers' risks should one crop fail. The
earlier hybrid maize was adopted, the greater farmers'

The table below reveals high performance scores for
Shikulu female managers -- indeed higher than that of
jointly managed farm's -- and low scores for their Shitoli
counterparts. jFemale'managed farms in Shitoli appear to be
experiencing a relative decline compared to jointly managed
farms, but also an absolute decline compared to female-
managed farms in Shikulu. This reverses the pattern
discovered iin relation to other objectives; Shitoli female
managers experience no improvement compared to their female
counterparts in Shikulu except in diversifying cash crops,
and their gaps with jointly managed farms are quite marked.

Table 17. Mean Number of Cash Crop and Food Cro Types and Years Since
Hybrid Maize Adopted by Sublocation and Farm Management Type 30


Female- Jointly Female- Jointly
managed managed managed managed

Mean No. 1.4 1.4 1.7 1.8
cash crops
Mean No. 2.2 2.0 1.6 2.0
food crops
Years since 3.6 3.5 3.3 4.2
hybrid maize

In Shikulu, with relatively few services both now and
historically, female-managed farms either surpass or equal
jointly managed farms in their crop diversification and in
the earliness with which they adopted hybrid maize. In
Shitoli, on the other hand, women score lower on all
i i '

measures, and particularly on food crops grown for family
consumption. This is noteworthy because food crops are
customarily thought to be within the work domain of women.
Jointly managed farms grow more diverse food crops, which
may be partially due to their greater access to services,
particularly the more recent ones which stress food.

Over the short term, women's commitment to and
experience in farming results in rates of new crop adoption
similar to those observed in jointly managed farms, which
have better access to extension services. Indeed, the mean
may actually underestimate female managers' abilities,
because female managers operate under greater labor
constraints than those faced by jointly managed farms. In
addition, this comparative performance of Shikulu women is
viewed against the low level of (male-oriented) service
which appears to have minimal impact even on farms with a
man present. In Shitoli, the male-oriented but generally
intensified services over a period of several decades have
incorporated both female and jointly managed farms into an
agricultural administrative framework, making improved
performance more dependent on service delivery.

Administrative services and land scarcity also
contributed to explaining differences in farm performance
under different levels of services. In Shikulu, an
overwhelming majority of women participated in communal
groups which share agricultural information and labor.
Women's organizational activities enhanced female managers'
performance. Furthermore, additional land enabled women to
put ideas diffused through this network into practice,
without risk to the basic food supply. Finally, there were
also market outlets for income-earning activities. In
Shitoli, extra land was simply not available and access to
assistance from women's labor groups was also less
available. Still, food crop diversification was maintained
in jointly managed farms through continued administrative
encouragement, which suggests that land scarcity does not
spell doom for food diversification.

The second more important explanation for gaps is the
systematic discrimination against women managers in a
service delivery system characterized by male preference.
The innovativeness of Shitoli jointly managed farms suggests
that governments do play some role in farmers' performance
which can offset the effects of land scarcity. Without the
support of the agricultural staff, Shikulu women managers
innovate independently by calling on group support in ways
that men cannot. Once a farming population becomes more
dependent on services, however skewed, the situation
changes. After a period of systematic neglect in support
services or valuable aid such as loans, performance


But how does one sort out the administrative versus
land scarcity explanations with any confidence? If land
scarcity were more important, one might very well "write
off" Shitoli, yet the!performance of jointly managed farms
suggests such a judgment would be premature. It would also
be wrong to under-estimate women managers under land
scarcity conditions simply because their organizational
resources and independent innovativeness are no longer as
functional as they' were. Indeed, with sufficient
administrative support, women once again surpass men, as
shown in Table 18.

Examining the saturated service subsample at Lirhembe,
cooperative members a opted special agricultural promotions
differently,' depending on the number of services they were
exposed to.i Initiallly, a cooperative network facilitated
the production and marketing of European vegetables, such as
cabbages, tomatoes, and onions; a Lirhembe-owned vehicle
(destroyed in an accident a year after the program started)
transported vegetables to nearby schools, markets, and
Kakamega township at minimal costs to farmers. Mexican 142
hybrid beans were another promotion which required purchased
seeds and special labor.

Table 18. Responsiveness to Two Crop Promotions by Farm Management Type,
Controlled for Different Amounts of Services Extended
(Lirhembe Sample Only)

Low servicing High servicing
(0-2 services) (3 or more services)

Female Joint Female Joint

Grew Mexican -- 30%/6 50%/5 31%/8
142 beans
N=28 N=36

Grew European 25%/2 50%/10 78%/7 48%/13
N=28 N=36

Unlike Shikulu female farm managers, who operated
effectively without administrative aid, Lirhembe female
managers did not perform as well with few services compared
to jointly managed farms, as evidenced in adoption rates.
In the high service category, however, female managers
respond dramatically to service relative to the negligible
response of jointly managed farms. !Besides suggesting the
improved performance which occurs with administrative
investments in female-managed farms, the data also lend


support to administrative rather than land scarcity
explanations for declining productivity. It appears that
female managers, because of their organizational resources,
have a performance edge in context with minimal services,
but that in more intensively administered contexts, the
balance is tipped in favor of farms where a man is present
until saturation services reverse the situation again for
certain crops. A key question, then, is whether policy
should aim solely at linking farmers' performance to costly
administrative support, or whether efforts could be directed
toward restoring once-innovative women's organizations with
their labor and information sharing with appropriate crop



Clear differentiation in the delivery of agricultural
services to jointly managed and female-managed farms has
been documented. In each service examined, discrimination
against female, managers was found, except in the brief
provision of tractor services to Lirhembe cooperative
members. The gaps increased as the value of the service
increased, from the rare demonstration plots and home visits
to the very aggravated gaps in specialized training and in
credit. Saturating services in Jan area only reduced
inequity when services were divorced from political
influence, but politics pervaded! the provisioning of
services. These inequitable patterns cast doubt on the
government's ability to realize policy objectives.

By way of summary, the information below contrasts mean
numbers of services for a cumulative service scale in the
ordinary, intensified, and saturated areas. Service items
were simply added. 31

Table 19. Mean Services by Administrative and Farm Management Type

Female-managed Jointly managed
S farms farms
Shikulu (N=108) .8 1.6

Shitoli (N=104) 1.6 2.2

Lirhembe cooperative 3.6 4.0
(N= 65)

Predictably, the more intensified the services, the
higher the mean number of services received. Yet increases
for female-managed fa ms lag behind jointly managed farms in
all settings. Quite strikingly, residence in intensified
settings merely raised female managers' access to the level
of jointly managed farms in ordinary areas. Yet declining
service differentials -- 0.8 in the ordinary area, 0.6 in
the intensified setting, and 0.4 in the saturated setting --
represent narrowing gaps with more administrative resources.
With the extraordinary staff and support services, however,
one would have expected the gap to be considerably narrowed

or eliminated. Extraordinary levels of service seem
necessary to compensate for the more typical sex
discrimination found in ordinarily served sublocations.
This slight narrowing of the differential is achieved at the
cost of farmers' dependency on a costly and ultimately
unreliable administration, particularly when external
resources are involved, such as in Lirhembe. Three
explanations are set forth to explain differentials.


Although administrative field staff attitudes were not
a prime focus of this research, the elimination of income,
land size, and performance of differentials suggests that
the most important factor in explaining differentiation are
prejudicial attitudes and ideological bias. A number of
agricultural instructors who were interviewed made
prejudicial comments about women farmers. More important,
though, was their expressed avoidance of women because of
customary patterns whereby men speak to men and women to

In the African way, we speak to the man who
is the head of the house and assume he will
pass on the information to other household
members. Being men, of course, it is easier
for us to persuade men.

Previous sections have documented the preferences of
technical staff for speaking with men.

Rather than rely on staff attitudes, the sample data
allow a test of alternative explanations for bias against
women, based on actual staff behavior. It is not
unreasonable to assume that staff, facing time and resource
constraints, would concentrate on the better off (who have
more cash to purchase seed, fertilizer, and labor), on the
owners of larger plots (who have more land on which to
experiment), and on the high performance, innovative farmers
(who would make officers look as if they achieved goals).
If none of these factors explain the differences in service
delivery, then the impact of prejudicial attitudes on
women's productivity becomes more apparent.


A common finding in extension studies is a tendency for
farmers' access to services to increase with wealth. As
shown earlier, though, female-managed farms span all
economic categories and have access to various cash-earning
activities as well as husbands' salaries. The economic

standing scale was dichotomized into low and high, with low
representing near su sistence living and very little access
to cash.

In comparing visits by agricultural instructors, 57
percent of low-income female managers had never been
visited, in contrast to only 39 percent of jointly managed
low income farms. That poor women receive the fewest visits
is fairly predictable. More surprising is how the bias
against women was maintained at high-income levels. A full
39 percent of higirlincome female-managed farms never
received a visit, compared to just 16 percent of the jointly
managed high-income farms.

This same tendency is borne out in looking at the mean
number of services received on a summary scale. Low-income
female-managers received an average of 1.1 services, while
their wealthier sisters received 1.4. In contrast, low-
income farms with 'a man present received an average of 1.5,
but their wealthier counterparts, 2.'4. Poor jointly managed
farms did better than wealthier female-managed farms. The
differences between low-and high-income female-managed farms
are not so marked, i suggesting that the consequences of
agricultural policy for women managers as a group are not as
differentiated as for jointly managed farms.


Female managed farms also span all land sizes, as they
are holding land for migrant husbands who will return upon
retirement. There is a slight tendency for female managers
to inflate low acreage because approximately a quarter of
them are widows. Yet widows are, in fact, slightly more
likely to get services because of their age and the prestige
that it brings, their length of time in the community, and
the reduction of sus icion about unrelated men (staff) and
women together. In an analysis of those farms over five
acres, numbering only 23 households (and thus demanding
cautious interpretation), 38 percent of female-managed farms
had never been visited by an extension officer, compared
with ony 7 percent of jointly managed farms. Land, like
income, does not hold as an explanation for differences in
service delivery.

Farm Performance

Extension officers understandably seek responsive
clients who are willing to diversify and innovate. In the
table below, a summary service scale allows an examination
of the relationship between management type and service
access, comparing farms with one versus two or more cash

crop innovations.

Table 20. Cumulative Service Scale by Farm Management Type Controlled for
Number of Cash Crops Grown.

0-1 Cash Crop 2+ Cash Crops
Female Joint Female Joint

2 services 92%/46 72%/50 63%/22 50%/29

3 services 8%/4 22%/15 34%/12 29%/17

4-5 services -- 6%/4 3%/1 21%/12

N=119 N=93

Female managers who diversify cash crop production are
clustered in the low-service categories. Among farms with
two or more cash crops, slightly over a third of the female
managers had access to a broad variety of services, in
contrast to half of the jointly managed farms. However,
even in that third, only a minute proportion of female
managers (only one case, in fact) had four or more services
extended to them. High-performance female managed farms
were not much better served than jointly managed farms that
showed little adaptiveness. Where only one cash crop or
less was grown, a quarter of the jointly managed farms had
access to three or more services, which was true for only
1/12 of comparable female-managed farms. Extension staff
did not seem to avoid poor performance farm households, as
long as a man was present.

In an examination of early adopters, that is, the 50
farmers who grew hybrid maize as early as five or more years
before the research, almost a third of female managers, or
31 percent, had no administrative support or advice for such
a move, while only 3 percent of the jointly managed farms
were so neglected. Timely adaptiveness, just as crop
diversification, actually underestimates female managers'
capabilities, given the labor constraints of female

Female managers' performance as measured by their
readiness to innovate and by productivity has been
documented in other studies about Kenyan agriculture. Peter
Moock found that women's maize yields per acre in Vihiga
Division (south of this study's research site) were similar
to men's, and even surpassed men's when access to education
and extension was controlled. Similarly, John de Wilde


noted "quite a few women" among progressive farmers, as did
studies in neighboring Muganda (see note 13).


Women are conceptually and politically invisible to the
administration. They rarely attend barazas, and the
continuing notion of male headship over households, however
symbolic given extensive male absence, precludes full
attention to women. But women's behavior, both as
individuals and collectively, !also is part of the
explanation for their invisiblity. Women rely on their own
labor and mutual aid groups, rather than on government. And
women's alternative institutions that compensate for non-
access to services appear to operate as well as those of the
agricultural administration except in Shitoli where
intensified and saturated services increase overall
dependency on government. Yet women's groups are invisible
to the administrationi.

Only rarely do farmers initiate individual contact with
agricultural instructors, and very few women managers are
among those who do. jIn the full sample, only 50 had ever
initiated contact, and seven were women from female-managed
farms. Among female managed farms, 88 percent never
initiated contact, iin contrast to 68 percent of jointly
managed farms. Women seldom make such contacts because they
fear that requesting special attention induces suspicion and
incurs obligations. Women have traditionally depended on
themselves and other women in the agricultural sphere, a
pattern reinforced by program implementation and male

Agricultural administration is ensnarled in local
politics, and those with political influence are likely to
derive benefits. Only infrequently have women -- largely
elite women -- banded together politically and become more
visible. And the way they have defined women's issues has
not helped female managers and farmers. For example, during
the 1972 M.P.'s campaign, wealthier women approached him as
a delegation to request services for women. The manner in
which services were'requested ranged from a very diffuse
request that the M.P. "remember the work women do" and "how
they need help," to more specific requests for donations for
church activities, a nursery or women's center. These women
delegates, from the local elite and well educated for the
area, were I past recipients ofj domestically oriented
government programs at five times the rate of ordinary
women, programs that included marriage training and
"homecraft" such as cooking, sewing, and knitting. Many
were schooled in the postwar era, when female education
emphasized homemaking skills. Given their training, elite

women tend to define "women's issues" within the realm of
home-, child-, and church-related areas, rather than take a
more comprehensive focus on income-earning activities
derived from agriculture.

Owing to male out-migration, women are a majority of
the electorate in the M.P.'s electoral constituency. Though
clan and subclan groups, which coincide with residential
areas, receive priority attention in campaign strategies,
blocs of women that cut across those key groupings are also
important. The M.P. needed women's electoral support, and
women were aware of the potential profitability of linking
themselves with him. However, their political visibility
resulted in the creation of a women's sewing and knitting
class in the cooperative staffed by a home economics
assistant. While the cooperative did benefit those women
who joined, this domestic-oriented service was enjoyed by
only a tenth of the women in the sublocation who had time.

Unless they become more visible, individually and
collectively, women will be by-passed in agricultural
administration. But it is important to consider which women
are visible, what interests they represent, and how
effective their strategies are. In a brief, two-year
episode of political integration in the late colonial era,
Chief William mobilized women on a location-wide basis to
use their labor for land consolidation efforts and to expand
his political base of support. Women were granted some
authority in the form of "women subchiefs" who represented
women in barazas. Women developed a complex organization
called Umoja, large-scale income generation through
collective agricultural labor, and even women's courts,
where women judges mediated infractions among women. 32
Women's brief integration through Umoja in the late colonial
political structure coincided with administrative outreach
to women's groups. Elite women's contemporary political
activity does not.


Common in bureaucracy is the convenience of continuing
earlier policies and only reforming them incrementally. The
wholesale importation of older Western notions about men as
'appropriate' farm managers and women as 'appropriate'
homemakers, though convenient, was in error and ignored the
facts. Following from these assumptions was the tendency to
separate farm clientele administratively and to segregate
staff by sex, creating two obvious problems: the minute
proportion of women field staff--2 percent in Kenya (3
percent for Africa as a whole), 33 and the dilution of
agricultural content. Kenya's experience bears out Uma
Lele's observation that "the goal of extension services has


frequently been not the increase in farm level productivity
of women but rather finding ways to reduce their
participation through promotion of more homebound
activities." 34

The neglect ofi female managers and small farmers
generally is sometimes dismissed as the result of cash crop
rather than food crop emphasis. However, maize has doubled
as a food and cash crop promotion since World War II.
Still, female maize producers experience bias. One wonders
how much more aggravated bias might have been if purely cash
crops were the focus.

The conveniences of staff and of using existing
procedures also forestall attention to all farmers. Male
field staff find it convenient to speak with men, to use
men's communication vehicles such asl barazas, and to rely on
elite farmers (some of whom are women), for their sporadic
demonstration plots. Loan staff also find it convenient to
rely on title deeds as the security for loans. These
strategies born of inertia result in serious inequities and
an underutilization of existing resources.



Merely increasing or intensifying services would have
little effect on addressing ideological bias, women's
invisibility, or inappropriate administrative conveniences.
The Kenya government's official reasons for constraints in
agriculture--knowledge, technology, and credit 35 --are only
a small part of the picture. Beyond this, a gender-specific
approach is needed which recognizes gender disparities in
access to and control over resources and gender-specific
communication patterns. Women producers must be viewed as
legitimate agricultural modernizers. In practice, general
neutrality perpetuates differentiation and ultimately,
ineffective implementation. Home economics, which does not
view women as managers or producers, cannot substitute for
transforming the agricultural administration to serve its
key clientele.

The suitability of "more of the same," or simply
intensifying or saturating an area with services, is highly
questionable. Added extension staff, while they improved
the likelihood of home visits, did not do so proportional to
their numbers. More staff operate in the same type of
administrative structure, which provides few incentives for
effective or equitable administrative performance for
farmers. The costs of continually expanding permanent staff
are too great in resource-scarce contexts like Kenya.
Similarly, the costs of saturating services are virtually
out of the question for the whole rural countryside.
Lirhembe, while innovative and redistributive, was near
collapse due to loan defaults, grade cow death from tick
diseases, and the breakdown of the tractor and vehicle, too
costly and sophisticated to repair locally. But it is not
merely government budgets that bear the cost of
administrative intensification. Another high cost of
intensification is the generation of farmers' dependency on
government to increase productivity rather than on their own
indigenous resources.

Planners in countries like Kenya should consider
alternatives which range from improved management and
supervision to reduced reliance on extension officers and
their home visits in ways which spread responsibility among
men and women professionals, paraprofessionals, and
community groups. Strengthening existing women's groups
would build on indigenous resources to permit information
and labor sharing, economies of scale, and guarantees for
loans. These groups, broadly representative of all women,
should also link with the local political process.

The details of t ese strategies--virtually cost-free
save administrative inertia--are outlined below and
summarized in Box 1, in their various components: staff and
staff training, extension and farmer training, credit, and
cooperatives. Of course jogging administrative inertia
is no easy matter, involving as it does
institutionalized bias and existing procedures.
Administrative will and leadership are essential. Extra
funds may be required to put procedures in place; or a
redistribution of funds inside agencies may be necessary.
However, the costs involved are far outweighed by the
elimination of wasteful expenditures and ineffective
procedures as well as by the gains resulting from the

The alternative strategies below, designed mainly to
affect the point of contact between field workers and
farmers, should be seen as accumulating from improved
management procedures to slight reforms in the
administrative structure, to a more thoroughgoing
transformation of that structure which involves more
considered shifts of policy. They do not address the
appropriateness of various service packages, which bear
ongoing examination, but rather to examine what options for
extending benefits more evenly and equitably exist at
different levels of change. Furthermore, the
recommendations for net restricted to Kenya; some of them
could be genera sized to countries with similar


Suggestions offered in this section do not require an
increase in staff or changes in administrative structure.
They primarily relate to field extension methodology and the
more efficient utilization of available resources. 36

Staff and Staff Training

Kenyan agricultural instructors work in an
administrative structure which gives them little incentive
to perform effectively, efficiently, or equitably. An
improved management strategy would alter the composition of
field staff and lii t their discretion by establishing
procedures which di ect field workers to serve those now
disadvantaged while at the same time making staff
performance matter in an incentive structure which rewards
or penalizes specificibehavior.

Presently, there are gross imbalances in the gender
composition of staff. Because male staffs' avoidance of
women managers is a primary reason for the inequities
I i

outlined in this case, more female staff should alleviate
this problem.

In addition, in most cases female recruits are tracked
primarily into home economics, which neither aims to serve
women as farmers nor has the budget and staff to reach more
than a minuscule percentage of "farm wives." Male extension
staff generally receive little to no home economics
training, the content of which carries over into
agriculture. 37

Administrative leaders should consider establishing
targets for minimum proportions of women in agriculture
staff training, along with extra cash incentives, if
necessary, to recruits and to the colleges, to alter
exclusionary practices of the past. More female staff in
the context of a majority female farm clientele and gender-
specific communication patterns will increase the likelihood
of serving women and women managers. Management might also
consider retraining home economists, who already receive
some agricultural training, although this move would deplete
an already sparsely staffed division if it is to retained.

Worries are frequently expressed that female staff
would not accept postings away from families. Management
should consider the importance of stationing staff--male or
female--near their families and furthermore, increasing the
stability of that placememt. As it is, transportation
problems for agricultural instructors limit their time with
farmers, and frequent staff rotation reduces instructors'
accountability for change within given areas.

Still, the overwhelmingly male staff is responsible for
serving women managers, for which special incentives will be
necessary. Management should institute a performance
appraisal system which rewards effective and equitable staff
performance. For example, a monitoring system could be
established to assure equitable performance. Or targets
could be set to increase contact with disadvantaged
clientele, such as women managers, along with monitoring
mechanisms to document the realization of goals. Staff who
do not meet goals would receive poor evaluations with
consequences for their salary increases and mobility

Such a reform might add some paperwork to an already
overburdened system. But these monitoring mechanisms could
be easily mainstreamed into existing reporting procedures.
For example, a system could be devised to review reports on
who is visited and ensure that men and women are visited
proportionately. Moving beyond the penalties implied in
such procedures, staff might also be rewarded monetarily,
with bonuses for equitable delivery. Such a reform would be


cost-free if the increments of upcoming salary increases
formed the fund for this two-tiered payment system, or if
status--in the form of awards or titles--constituted the

Of course, both new and current staff should be aware
of gender inequities, the legitimacy of women producers, and
the desirability of increasing their productivity in new and
in-service training programs. The trainers themselves must
be exposed to and encouraged to incorporate this type of
information in the curriculum. Such information goes beyond
offering a home-econom cs course option to trainees.

Extension and Farmer Training

Home extension allows individual, personal contact
between farmers and field specialists, adapted to unique
farm conditions. But it is a costly means to deliver
information,| and the discretion available to field workers
makes rife the possibilities of both preferential and
discriminatory treatme t.

Group extension increases the prospects of accurate,
relevant, and visually convincing information transmission
with potentially more equitable outcomes. Problematic,
though, arel the communication channels used to announce
demonstration plots and the groups for whom demonstrations
are made.

Administrators may consider promoting the use of more
group extension, after first assuring that staff understand
the community social structure. Those staff must carefully
consider the kinds of communication channels in which
messages are transmitted to assure that dissemination occurs
through women's networks as well as men's. The focus should
shift away from individual home visits; group extension,
such as demonstration plots and continuous work with local
groups, should be used with both separate women's groups and
men's groups, or integrated groups, depending on community
structure. I In many cases, this involves identifying
existing neighborhood groups both for dissemination and
demonstration. 38

Farmer training provides more intense instruction on
specialized topics. Under the current approach, training
staff are competent and monitored effectively. Yet farmers
perceive training as available "by invitation only," and
hence as something Lo confer status. Farmers must be
assured that any and all are entitled to receive training.


Credit is virtually unattainable for all but a select
few. Were low-risk, in-kind loans, such as the Guaranteed
Minimum Return, to have a reduced land minimum, a larger
number of farmers should have access. All credit
recommendations add cost to the extent that access expands
among small farmers.

Institutional procedures for acquiring credit make it
extremely difficult for women managers to obtain capital
from this source. Policymakers should consider adopting
minimum quotas to assure that women farm managers receive
AFC loans.


Most cooperative recruitment is aimed at men, and
consequently, women managers or wives either do not join or
they use benefits less frequently than they are entitled to.
Lingering but increasingly irrelevant notions of "male
heads," regardless of where they are located or what they
contribute, reduce the likelihood that women will seek
membership or that there will be cooperative out-reach to

In some cases, one could recommend recruitment of men
and women as an improved management procedure. In the case
of Kenya, however, cooperative laws which require members to
hold land title deeds perpetuate bias against women, given
the way that land reform has excluded women from formal

Systematic recruitment of women farm managers, based on
their producer status rather than land-ownership status
should be encouraged. Also, when partners live together,
joint membership in cooperatives should be required to
encourage both women's participation and their visibility.
Along with that, cooperative benefits and crop payments
should be allocated directly to producers, including women.
Life staff training, information about gender inequities and
the legitimacy of women producers should be incorporated
into cooperative training.


An intermediate change strategy goes further than
simple improved management to reform structures slightly by
expanding or changing the composition of operating staff,
the manner of disseminating services, and by promoting minor
shifts in policy. This does not necessarily imply greater
cost, only more energy in reform efforts. Alternative B







Establish targets for
minimum proportions
of women into agri-
cultural training.-- -

Increase stability of
staff placement.


Male staff avoid
women farm managers.



Frequently transfer- None
red staff lack spec-
ific area expertise
and accountability
for results.







Low (unless
utilized) ---

Low None

Recognize-desirabil----High-staff turnover-- None
ity of both men and decreases quality of
women staff to be performance; distant
stationed near fam- transfers the ratio-
ilies. nale for low female

Devise incentives for
male staff to serve
women farm managers,
such as MONITORING to
assure equitable per-
formance and/or TAR-
GETS to increase con-
tact with disadvan-
taged clientele.

Reward staff monetar-
ily for equitable de-

Incorporated informa-
tion about women pro-
ducers and inequities
into training.

EXTENSION Promote group exten-
STRATEGY sion with women's and
men's groups.

Overwhelmingly male
staff avoid certain
farm managers and
lack incentives to
serve them.

Overwhelmingly male
staff avoid certain
farm managers and
lack incentives to
serve them.

Staff lack cues to
improve equitable

Group outreach ex-
tends quality out-
reach in equitable

None, but

None, hut

Some, for

ity groups
and commun-
ication net-






Some ---. None





Low -

None (If sal-
ary increments
the funding



Disseminate informa-
tion about training
through women's and
men's networks.

Avoid substituting
home economics train-
ing for agriculture.

Reduce land minimum
requirement for
Guaranteed Minimum
Return loans.

Establish quotas to
assure that women
farm managers receive
AFC loans.

Recruit women farm

Require joint member-
ship for partners
living together.

Allocate co-op bene-
fits and payments

Incorporate informa-
tion about women pro-
ducers and inequities
in training.

Information reaches
men primarily.

Home economics di-
lutes full agricul-
tural training.

Small farmers can-
not qualify now.

Women farm managers
rarely get loans.

Women not tapped in

Recognizes women as
de facto producers,
managers, members,
and representatives.

Benefits serve as
incentives for act-
ual producers.

Co-op managers lack
cues about problems.








Some for























Some, associ-
ated with ex-
panded credit.








(all of the
above PLUS)




Create network of
paid, part-time para-
professionals, both
men and women.


Professional staff
cannot adequately
cover current terri-


Some -iden-
tify farm













(all of the
above PLUS)


Reward professional
staff monetarily from
community revenues.

Solicit evaluation
feedback from com-
munity-committees on
staff performance.

Bring training-staff
to the community.

Group loans to-pro-
ducers lacking title
deeds or salaries.

Lending facilities
located closer to
farmers (or mobile

Make co-op management
and participants-re-
flect community pro-
ducer composition,
such that half or
more women.


Require staff to work
with paraprofession-
als and representa-
tive community mem-
bers to establish
collaborative goals
and strategies for
agricultural change.

Staff avoid certain
farmers; lack incen-
tives to serve them.

Staff not account-
able to communities

Distant training
difficult for women
farm managers to

Procedures clue
many producers, such
as women managers.

Less mobile farmers

Currently unrepre-
sentative._ _-

Staff not account-
able to community.


Some -appro-
priate re-

None -

Some -ex-sts-
ing groups
or group

Some, com-

Some, repre-







Top-down -7=Some-= -







Cost to com-


Some travel

ated with ex-
panded credit.

Some transport



... -... ... ...... .... ....................
Integrate agricultur- Segregated staff
al and home economics tend to segregate
staff, farm community.

Develop a "Farming
Systems" approach to
agricultural research
with farmers.

Establish joint title
for land ownership
with corresponding
names on title deeds.

Research not locally
adapted or judged by
farmers' criteria.

Procedures exclude
women producers from
credit and from con-
trol over assets.


None (would
augment re-
search with
local appli-

Some, spouse




Very com-


Very -

Initially high

Some -training
of staff





should be viewed as an add-on to the cumulative set of
changes recommended for Alternative A. Its effectiveness
depends on completing the above changes.


Besides those changes recommended for staff above,
management should consider ways to extend quality staff
coverage even further with a network of paid, part-time
paraprofessionals, both men and women who work with
neighbors in areas coverable by foot.j Already, innovations
are diffused through informal means, but the quality of
technical information being passed is uncertain. Trained
paraprofessionals would improve the quality of that
information in systematic ways. Local paraprofessionals are
familiar in the community, and their residence in the area
would enhance interaction with farmers. Women could be
easily recruited. Besides, the lower salaries of men and
women paraprofessionals, compared with those of
professionals, would represent cost savings compared with
simply augmenting professional staff.

To increase the equitable performance of professional
staff while at the same time broadening their accountability
to men and women in the communities they serve, the upper
tier of monetary rewards for their work along the lines of
the two-tiered system described above would be derived from
community sources. Such a reform, however, has the danger
of politicizing delivery even more to men and to the already
privileged, to the extent community participation structures
are unrepresentative. Therefore,' increasing women's
political participation to produce accountability to women
is essential. Reducing the complications of that
recommendation somewhat but at the same time meeting its
rationale would be the strategy of soliciting evaluation
feedback from representatives, male and female, from the
communities in which agriculture staff work.

Extension and Farmer Training

Physical distance between farmers and training centers,
requiring days or weeks away from the residence, limits
farmer trainees to those households with at least two
adults. Furthermore, cultural constraints on women's
mobility may make female participation virtually impossible.
Policymakers should create training alternatives, such as
bringing training staff to the community to run comparable
programs out of a school, church, or community center.
Travel expenses would add to costs, ibut the result would be
widely expanded access for farmers Currently, training
staff are underutilized, so such a reform would improve the
effectiveness of their professional time. Again, trainers
must recognize that'home economics is no substitute for

agricultural training.


Besides expanding the coverage of current loans through
existing, but improved procedures as outlined above,
policymakers should develop alternative means to deliver and
secure the payment of credit. For example, loans might be
extended to existing or newly formed women's groups, secured
through group liability and consequent peer pressure to
guarantee repayment. Also, administrators should consider
locating lending facilities closer to farmers, especially
for those less mobile, such as women managers.


If women are to be recruited into cooperatives
effectively, (assuming a policy framework which does not
limit cooperative membership to landholders), steps should
be taken to ensure that women are represented equitably in
management, to reflect their numbers in the community. This
should also increase the number of women who attend
cooperative training programs which, like farmer training,
might best be located in the community.


A more thorough change goes furthest to alter
management and their structures in more fundamental ways
which involve the community as well as the administration,
and involve broader policy decisions. These changes work to
empower individuals and involve them in ongoing flexible
program goal definitions. 39 Again, the following
recommendations will only be effective if alternatives A and
B are pursued.


Once properly supervised and made more accountable to
the community, staff should be required to work with
paraprofessionals and representative farmers, reflective of
numbers in the community, to establish goals and strageties
for agricultural change. Women would be active
participants. Communities would share responsibility with
the administration over strategies for change that affect


Rather than separating research from extension and from
farmers, experimental | research should be conducted with
farmers on location land should be judged according to
criteria farmers use when making adoption decisions, such as
risk reduction, appropriate crop mixes, and improved income.
Appropriate Ihere is a farming systems approach which takes
into account different management structures, female and
otherwise, labor c nstraints of women and men, and
differential access to inputs and land.

Policymakers should also consider integrating
agriculture into home economics staff training--already
accomplished to a certain extent--and also integrating home
economics into agricultural training with the ultimate aim
of a single agricultural extension service, with home
economics as a district level specialization. A bifurcation
of services by task, gender, function, and clientele is
unrealistic, inefficient, and ineffective.


The fundamental cause for women managers' low access to
loans is land reform which excluded women producers from
control over and title' to land they till. As such, they
operate much1like tenants lacking secure access to and stake
in this productive resource. Steps should be taken to
establish women's ownership of land, either jointly with
spouses or as individuals, with corresponding names on title

General Policy Change

Besides these specific recommendations, other more
general policies relating to price and participation which
now hinder small farmers and women managers should be

A strategy less dependent on bureaucracy, and thus less
costly, would improve guaranteed prices, now too low in some
cases to cover input costs or to provide incentives for
farmers to !produce more. Low prices for staple goods
frequently placate urban consumers rather than make higher
production attractive lfor farmers. And producers, women
included, must be compensated directly for their work to
provide incentives.' Moreover, either market restrictions
should be lifted, or the number of maize bags that can be
moved across district boundaries should be raised. Among
small farmers in Kenya, these restrictions fall heavily on
women traders. In the best scenario, they hide maize and
pass through!police checkpoints, pay the police to pass with
theAir goods, or, at worst, have their maize confiscated.

Women, both individually and collectively, must be
integrated into local community politics and development
committees in order to learn of and receive the many
politically charged benefits. Kenya's official desire to
increase rural participation must extend to women in
balanced ways, in proportion to their numbers and
developmental importance. Separate women's bureaus must
also see their role as making all policies sensitive to
women rather than merely social services, community
development, or traditionally separate women's activities.
To ensure this policy integration, women's advocates, with
resources and authority to both encourage and penalize,
should be positioned in national, district, and local
offices to work with women's groups toward this aim.

The recommendations in Alternatives A through C move
from the more simple, management-oriented changes to those
that involve the community, necessarily more complex. Yet
such community involvement, besides jibing with Kenya's
goals to increase participation, reflect a more
people-oriented strategy which builds both capacity among
community residents and accountability from staff, and
thus a more holistic development. Virtually all of the
recommendations are cost-free, save the energy necessary to
institute change and overcome the usual bureaucratic
inertia. Taken alone, each would produce outcomes that
improve the effectiveness of agricultural
administration. Adopted together, the recommendations
should have a profound impact on addressing the issues
raised in this case study.


Historically, women in western Kenya have produced food
for the community and contributed considerable resources to
households. With the introduction of new economic and
political forms, agriculture was increasingly
commercialized, employment was made available to men earlier
and more extensively than women and agricultural advice,
training, and loans were oriented primarily to men. An
idealized view of the family was introduced and diffused
through government programs in which husbands are
breadwinners and women are relegated to the domestic rather
than productive sphere. Still, women are active in
agriculture, as workers and managers, and they belong to
thriving associations. Yet early agricultural policy was
explicitly tied to idealized assumptions about the family.
Agricultural services were initially and predominantly
delivered to men, a result of administrative structures
largely comprised of men who implemented policy in a setting
with strong gender-specific communication patterns.

Governments both before and since the seventies have
channelled common agricultural services--such as
agricultural instructors' visits--to men, as well as the
most valuable services such training and loans, the latter
being tied Ito the control of other resources such as land
and wages. In fact, disparities between female-managed
farms and other types of farms widen as the value of the
service increases, both in cost to the administration and in
benefit to the farmer. Expensive strategies such as
intensifying or saturating areas with services do little to
alter these Idisparitids. A bias against female managers
exists, independent 6f other factors such as economic
standing, land size, and performance. Indeed, wealthy and
innovative women farming on relatively large tracts of land
experience discrimination when compared to their male
counterparts. Land reform has reinforced this pattern,
making it difficult for women managing farms to acquire

Despite! Kenya's well-developed extension system, a
pervasive bias exists against an important and large part of
its clientele, women managers. The administrative failure
to reach this large female clientele is a telling critique
on government commitment and ability to affect agricultural
change. By pursuingi a range of low-cost administrative
reforms which recognize and build on indigenous capability,
including those of female managers, Kenya has the potential
to increase productivity and thereby ameliorate the domestic
food crisis.


1 The policy objectives came from two sources. First,
objectives are laid out in the Government of Kenya,
Development Plan for the Period 1974 to 1978 (the period of
this research), Part I, (Nairobi, Government Printer, 1974),
pp. 18ff and Chapter 10 on Agriculture, Land Settlement, and
Cooperatives. Second, objectives are inferred from program
implementation in the field research area.

That Kenya is concerned with equitable development can
be derived from frequent reference to the goals of "social
justice," a "more equitable distribution of income," and the
"equitable sharing of its (economic expansion) benefits."
pp. iv., 1, Development Plan. The plan also makes reference
to the Sessional Paper no. 10, "African Socialism and Its
Application to Planning in Kenya," 1965, a development-with
equity philosophical statement which this plan continues.

2 The quote is from Louise Fortmann, "Women's Work in
a Communal Setting: The Tanzanian Policy of Ujamaa," Women
and Work in Africa, Edna Bay, ed. (Boulder: Westview
Press, 1982), p. 195. See also, Ester Boserup, Woman's Role
in Economic Development (New York: St. Martin's Press,
1970); Jette Bukhe, Village Women in Ghana (Uppsala:
Scandinavian Institute of African Studies, 1979).

3 This and the following comments from Wagner are
found in Gunter Wagner, The Bantu of North Kavirondo, Vol. 1
(London: Oxford University Press, 1949) pp. 41, 20, 45,
156. Kavirondo is the former name of Kakamega. Wagner's
work was done among the northernmost Bukusu and southernmost
Maragoli subgroups of the Balyuia. Idakho location is just
north of Maragoli location. See also another
anthropological study by Walter Sangree, which focuses on
the Luyia group in the southeastern part of the district,
Age, Prayer & Politics in Tiriki, Kenya (London: Oxford
University Press, 1966). Also of interest is a government
time allocation study entitled "A Report on Economic Studies
of Farming in Nyanza Province, 1963," Farm Economic Report
No. 26, 1969.

4 Kenya National Archives, Native Agricultural File,
Center and North Kavirondo(1930-46).

5 Judith Heyer. Survey of Agricultural Development in
the Small Scale Farm Areas of Kenya since the 1920's
(Nairobi: University of Nairobi, Institute for Development
Studies) Working Paper No. 194, October, 1974, pp. 15-16.

6 Kenya Colony & Protectorate, A Plan to Intensify the
Development of African Agriculture in Kenya, Compiled by
R.J.M Swynnerton (Nairobi: Government Printer, 1954), p.53.

7 Swynnerton, p. 10.

8 Everett Rogers, Modernization Among Peasants; The
Impact Of Communication and The Diffusion Of Innovations
(New York: Free Press, 1969, 1962).

9 Judith Heyer, D. Ireri, and Jon Moris, Rural
Development in Kenya, (Nairobi: East African Publishing
House, 1971); selections in James Sheffield, ed., Education,
Employment and Rural Development (Nairobi: East African
Publishing House, 1967); ILO/UNDP, Employment, Incomes and
Equality: A Strategy for Increasing Employment in Kenya
(Geneva: ILO, 1972);1; Joseph Ascroft, Fred Chege, Joseph
Kariuki, Niels Roling, and George IRuigu, "Does Extension
Create Poverty in Kenya?" East Africa Journal, 1972; David
K. Leonard, "Why Do Kenya's Agricultural Extension Services
Favor the Rich Farmers?" Paper presented at the 16th Annual
Meeting of the African Studies Association, Syracuse, New
York, October-November' 1974.

10 This and following historical evidence came from
the Kenya National Archives. Nyanza Province. Central and
North Kavirondo District Annual Report, 1948; Native
Agricultural Files.

11 Information in this paragraph comes from the
following documents: North Kavirondo District Annual Report,
1951, p. 14, and 1952, pp. 45, 70; Special supplement to
the District Annual Report, 1955, p. 11: District Annual
Report, 1959, pp. 26, 28. The Kenya Colony and Protectorate
Annual Reports also report the recruitment of women
agricultural assistants,i as does Lord Hailey, An African
Survey (London: Oxford University Press, 1956), who reports
that 48 female staff were employed in agriculture, p. 894.
The Ikolomani Division' Miscellaneous Affairs file, 1958-59,
reports that district-level male agricultural staff had a
parallel female staff.

12 Bertha F. Strange, The Story of Kenya's Home
Extension Service: 1962-65, (Washington, D.C.: U.S.A.I.D.
Mission to Kenya, 19651), p. 1.

13 See Peter Moock, "Managerial Ability in Small Farm
Production: An Analysis of Maize Yields in the Vihiga
Division of Kenya," (Unpublished Ph.D. Dissertation,
Columbia University, 1973), p. 168; and his "The Efficiency
of Women at Farm Managers: Kenya," American Journal of
Agricultural Economics 58, 5, December, 1976. See also John
de Wilde, Experiences with Agricultural Development in
Tropical Africa (Baltimore: Johns Hopkins University Press,
1967), p. 169; Edgar Bowe and Jon Moris, "Social
Characteristics of Progressive Buganda Farmers," East
African Journal of'Rural Development 2, 1,(1969), p. 57;
Judith Heyer and Joseph Ascroft, "The Adoption of Modern
Practises on Farms in Kenya: Preliminary Results of a 1968
Survey of Farms across Kenya," (Dar es Salaam: University
of East Africa Social Science Conference, Volume V, December

27-31, 1970), p. 336.

14 Republic of Kenya, "Report of the Agricultural
Education Commission," Chaired by Weir (Nairobi: Government
Printer, 1967), p. 13, contains the information on the
number of women officers. Those figures matched the 2%
figure in Kakamega District.

15 All demographic, spatial, and density information
in this study comes from the Republic of Kenya, Kenya
Population Census, 1969, Vols 1-3 (1970), unless otherwise

16 S.H. Ominde refers to the uneven sex ratio in his
Land and Population Movements in Kenya (Nairobi: Heinemann,
1968), p. 109. Wagner provides male out-migration figures
for the 1930s, II, p. 94, and Judith Heyer, "A Survey of
Agricultural Development in the Small Scale Farm Areas of
Kenya since the 1920's," p. 2, (see note 5) claims Kakamega
as the largest labor exporter, as does the 1945 North
Kavirondo District Annual Report, P. 17.

17 A material consumption scale of 5 points was used
to measure economic standing because farmers rarely received
a regular wage income. The scale was dichotomized for
purposes of simplicity; levels 1 and 2 are low, and levels
3-5, high. At the low economic standing level, farmers had
a thatched roof and very little furniture in the house --
not more than a table and two folding chairs plus some
cooking pots. At the high level, farmers lived in a
permanent or semipermanent house and had more furniture than
the basic table and chairs. Permanent houses have concrete
walls and floors, as well as corrugated iron sheet roofs.
Semipermanent houses have corrugated iron sheet roofs; their
mud walls are packed into a frame made of long, wiry
branches fastened together, and their mud floors are beaten
to become hard and level.

18 This figure is calculated from the 1969 Census,
Vol. 1, p. 123, by subtracting the number of "excess" women
aged 25-29 from the number of men in the same age group.
According to the ILO/UNDP, Employment, Incomes and Equality,
a third of Kenyan households have a male head working away
in town (400,000) and another 125,000 are female headed, p.

19 This figure is based on questions and observations
during interviews, but especially on a more intensive study
of the village in which I lived. Respondents are reluctant
to state how much support they receive from distant

20 Interview in Shigokho, April, 1975. All quotes
such as these come from field notes.

21 Achola Pala Okeyo, "Daughters of the Lakes and
Rivers," Women and Colonization, Mona Etienne and Eleonor
Leacock, eds. (New York: Praeger, 1980).

22 For example, at the Kakamega District International
Women's Year Seminar, March, 1975.

23 David K. Leonard, "Organizational Structures for
Productivity in Agricultural Extension," in his Rural
Administration in Kenya (Nairobi: East African Literature
Bureau, 1973), pp. 134, 145. For a penetrating analysis of
the incentive structure in which extension officers work,
see Leonard, Reaching Peasant Farmers: Organization Theory
and Practice in Kenya (Chicago: University of Chicago Press,

24 Because this is not a random sample and thus does
not reflect a normal distribution, The Chi-Square Tests of
Significance are technically not appropriate. The size of
the sample may mean it approximates a normal distribution,
and therefore tests of significance have been included for
exploratory purposes. Table size of less than 212 indicate
missing data or are subsamples, as specified in the table
title. Numbers have' been rounded off to the nearest

25 This represents one-sixth of the cost of school
fees to send a child to Standards 5, 6, or 7 in a government
primary school (1975). The price in 1975 for a bag of
hybrid maize seeds -- enough for one acre -- was Shs. 23.

26 In 1975, minimum acreage for the GMR was 15 acres,
although I knew of several cases where farmers with 9 acres
received loans. The figure comes from the Development Plan.
Vol. I, p. 212 along with a statement that the government
wants to change this disparity.

27 On resource constraints, see Warren F. Ilchman and
Norman Uphoff, Political Economy of Change (Berkeley:
University of California Press, 1971), pp. 32, 58-59: Robert
H. Jackson, "Administration and Development in Kenya: A
Review of Problems Outstanding," Development Administration;
The Kenyan Experience, Goran Hyden, Robert Jackson, and John
Okumu, eds. (Nairobi: Oxford University Press, 1970), pp.
326; Guy Hunter, Modernizing Peasant Societies: A
Comparative Study in Asia and Africa (London: Oxford
University Press, 1969), pp. 172-73, 187; Albert Waterston,
Development Planning:, Lessons from Experience (Baltimore:
Johns Hopkins University Press, 1965). The frequent use of
extension-farmer ratios to indicate quality or improvement
is also suggestive of this approach.

28 Rogers, 1962, pp. 40, 285-292.

29 The 1975 prices of grade cows started at Shs. 1200/

- or about $200, and local cows were half that price. The
price of a grade cow was equivalent to secondary school fees
for two children at a government supported school in 1975.

30 The hybrid maize-bean adoption is designated as
both a food and cash crop. The cash crops include the
hybrid maize-beans combination, coffee, European vegetables
(tomatoes, cabbages, or onions), passion fruit, and sugar
cane. The food crops include the hybrid maize-beans
combination, European vegetables, root crops (sweet potato
or cassava), millet crops (finger millet or sorghum), and
nut crops groundnutss or monkey nuts).

31 The cumulative services refer to 1 visit, 2 or more
visits, farmer training, demonstration plot attendance, and
loan application or receipt. Added to that for Lirhembe
members are maize and fertilizer loans, tractor services,
and grade cow receipt.

32 A full analysis of this organization can be found
in my "The Umoja Federation: Women's Cooptation into a
Local Power Structure", Western Political Quarterly, 33, 2
July, 1980, pp. 278-290.

33 For the African figure, see Burton Swanson and
Jaffar Rossi, International Directory of National Extension
Systems (Urbana: University of Illinois, 1981), p. 272.

34 Uma Lele, The Design of Rural Development: Lessons
from Africa (Baltimore: Johns Hopkins University Press,
1975), p. 77.

35 Development Plan, P. 18.

36 Approaches like this can be found in Robert
Chambers, Managing Rural Development (Uppsala: Scandinavian
Institute of African Studies, 1974) and Leonard, 1977.

37 Kenya now requires male agricultural instructors to
take a home economics subject during training. Such courses
are also offered in in-service training, I am grateful to
Janice Jiggins for this information as well as other
information about recent changes in Kenya.

38 Kenya is one of the few countries to have
instituted a woman-sensitive Training and Visit system which
requires the identification of women-contact farmers and
women in contact groups. (Again, I am grateful to Janice
Jiggins for this point.) However, although easily
adaptable, most Training and Visit systems still perpetuate
male preferential patterns. For information on Training and
Visit systems see Agricultural Extension: Training and Visit
System, by Daniel Benor and James Q. Harrison, and A Systems
for Monitoring and Evaluating Agricultural Extension
Projects, by Michael Cernea and Benjamin Tapping (both,


Washington DC, The World Bank, 1977).

39 This "people-oriented" approach as opposed to top-
down approaches in outlined more fully in Felipe Alfonso and
David Karten, Bureaucracy & The Poor (West Hartford,
Kumarian Press, 1980) and David Korten, "Community
Organization and Rural Development: A Learning Process
Approach" Public Administration Review Vol. 40 Sept-Oct
1980 pp. 400-510

I recognize that most development agencies & donor
organizations currently operate in ways incompatible with
this approach. However, some private voluntary
organizations have used this approach successfully.



A geographically purposive sample of 212 farm
households was obtained in the Idakho research site between
December, 1974, and June, 1975. It represents 10 percent of
the total number of households in the geographic areas
targeted. My initial concern was to assure that varying
distances from the road and main paths, and thus from
agricultural instructors and services, would be covered.
These geographic areas coincide with clan and subclan
identities. Once spatial areas were designated in order to
obtain geographic and clan representativeness, I selected
farms that would be representative of varying economic
standings and age groupings. Numbers were based on my
approximations of their proportion of the population. I did
not know in advance, however, about the farm management type
until the interview had begun. The correspondence of female
managers to the proportion of female heads in Kakamega
supports the notion that my choice of farmers was chance-
like in method. The sample is not, however, a random one,
and the universe of this sample is restricted to one
location. The sample does not purport to generalize to all
of Kenya or Africa, but rather to illustrate sex differences
within a sample which may be suggestive for other parts of
Kenya or Africa with agriculturally based cultures in areas
of high land scarcity and rates of male out-migration.

Though scientific sampling techniques were not
utilized, I am confident that the sample judiciously
represents a reasonable cross section of farmers in high and
ordinarily serviced areas in western Kenya. The basis of
this confidence is my six-month residence in Shikulu and my
participation in community life, where I gained in-depth
knowledge of that subclan and geographic area.

A female research assistant from Shikulu and I
conducted the interviews. My assistant translated questions
and responses from Lwidakho to English. The interview
format generally proceeded as follows. We entered a farm
compound, greeted household members, introduced ourselves,
and requested an interview. (No respondents declined to be
interviewed.) We then proceeded to ask a systematic set of
questions of each farmer, beginning with demographic and
crop information, land size, husbandry practices, and
sources of information about crops and practices. Farmers
live in dispersed settlement patterns, and their fields
surround houses. This afforded us the opportunity of
visually verifying answers to questions relating to land
size and crop husbandry practices. The next section of the
interview involved questioning farmers about their
awareness, receipt, and evaluation of agricultural services.
Following that, we asked questions about community
participation and electoral activity. For female

respondents, questions were then asked which were designed
to tap attitudes toward women and women's problems. Probing
about marital situations depended greatly on the willingness
of respondents to discuss such matters. No systematic
responses of personal marital relations have been tabulated
because of the unevenness of answers.

Great efforts were made to make the interview as
informal, conversational, and reciprocal as possible. No
questionnaire was utilized; instead, notes were taken during
the interview, which I then typed at the end of the day.
When the interview was completed, respondents were invited
to interview and probe the interviewer. This exchange
permitted deeper insight into the respondent and a more
holistic understanding of the farm. The length of the
interview depended on the openness of respondents. A
typical interview took 40 minutes; minimal length interviews
took 20 minutes, and the lengthiest were 3 hours.



Ascroft, Joseph; Chege, Fred; Kariuki, Joseph; Roling,
Nieles; and Ruigu, George. "Does Extension Create
Poverty in Kenya?" East Africa Journal, 1972.

Benor, Daniel, and Harrison, James Q. Agricultural
Extension: Training and Visit System. Washington,
D.C.: The World Bank, 1977.

Boserup, Ester. Women's Role in Economic Development. New
York: St. Martin's Press, 1970.

Bowe, Edgar, and Moris, Jon. "Social Characteristics of
Progressive Buganda Farmers." East African Journal of
Rural Development 2, 1, (1969).

Bukhe, Jette. Village Women in Ghana. Uppsala:
Scandinavian Institute of African Studies, 1979.

Cernea, Michael, and Tepping, Benjamin. A System for
Monitoring and Evaluating Agricultural Extension
Projects. Washington, D.C.: The World Bank, 1977.

Chambers, Robert. Managing Rural Development. Uppsala:
Scandinavian Institute of African Studies, 1974.

de Wilde, John. Experiences with Agricultural Development
in Tropical Africa. Baltimore: Johns Hopkins
University Press, 1967.

Fortman, Louise. "Women's Work in a Communal Setting: The
Tanzanian Policy of Ujamaa." In Women and Work in
Africa, edited by Edna Bay. Boulder: Westview Press,

Hailey, Lord. An African Survey. London: Oxford
University Press, 1956.

Heyer, Judith. A Survey of Agricultural Development in
Small Scale Farm Areas of Kenya Since the 1920's.
Working Paper no. 194. Nairobi: University of
Nairobi, Institute for Development Studies, 1974.

Heyer, Judith, and Ascroft, Joseph. "The Adoption of Modern
Practises on Farms in Kenya: Preliminary Results of a
1968 Survey of Farms Across Kenya." In Documents of the
East Africa Social Science Conference. Dar es Salaam:

Heyer, Judith; Ireri, D.; and Moris, Jon. Rural
Development in Kenya. Nairobi: East African
Publishing House, 1971.


Hunter, Guy. Modernizing Peasant Societies: A Comparative
Study in Asia and Africa. London: Oxford University
Press, 1969.

ILO/UNDP. Employment, Incomes and Equality: A Strategy for
Increasing Employment in Kenya. ;Geneva: ILO, 1972.

Ilchman, Warren F., and Uphoff, Norman. Political Economy
of Change. !Berkeley: University of California Press,

Jackson, Robert H. "Administration and Development in
Kenya: A Review of Problems Outstanding." In
Development Administration: The Kenyan Experience,
edited by Goran Hyden, Robert Jackson, and John Okumu.
Nairobi: Oxford University Press, 1970.

Kenya, Republic of. African Socialism and its Application
to Planning in Kenya. Sessional Paper no. 10, 1965.

Agricultural Education Commission.
Report of the Agricultural Education Commission.
Nairobi: Government Printer, 1967.

__ Ministry for Finance and Economic
Planning. Development Plan for the Period 1974 to
1978. Nairobi: Government Printer, 1974.

Ministry for Finance and Economic
Planning. Kenya Population Census 1969, Nairobi:

_. Ministry of Economic Planning and
Development. A Report on Economic Studies of Farming in
Nyanza Province, 1963. Farm Economic Report no. 26.

Kenya Colony and Protectorate. Nyanza Province. North
Kavirondo District. Annual Report, 1948. Nairobi,
Kenya National Archives, Native Agricultural File.

Annual Report, 1951.

i ___. Annual Report, 1952. _
_i ;

S. Annual Report, 1959.

_. Annual Report,
1955. "Special Supplement." Nairobi, Kenya National
Archives, Native Agricultural File.

_A Plan to Intensify the
Development of African Agriculture in Kenya. R.J.M.
Swynnerton, comp. Nairobi: Government Printer, 1954.

Korten, David C. "Community Organization and Rural
Development: A Learning Process Approach." Public
Administration Review (Sept.-Oct. 1980).

Korten, David C., and Alfonso, Felipe. Bureaucracy and the
Poor. West Hartford: Kumarian Press, 1980.

Lele, Uma. The Design of Rural Development: Lessons from
Africa. Baltimore: Johns Hopkins University Press,

Leonard, David K. "Organizational Structures for
Productivity in Agricultural Extension." In his Rural
Administration in Kenya. Nairobi: East African
Literature Bureau, 1973.

___ Reaching Peasant Farmers: Organization
Theory and Practice in Kenya. Chicago: University of
Chicago Press, 1977.

"Why do Kenya's Agricultural Extension
Services Favor the Rich Farmers?" Paper presented at
the 16th Annual Meeting of the African Studies
Association, Syracuse, October-November 1974.

Moock, Peter. "The Efficiency of Women as Farm Managers:
Kenya." American Journal of Agricultural Economics 58,
5 (December 1976).

Managerial Ability in Small Farm Production:
An Analysis of Maize Yields in the Vihiga Division of
Kenya. Ph. D. diss., Columbia University, 1973.

Nairobi. Kenya National Archives. Native Agricultural
File, Center and North Kavirondo (1930-46).

Okeyo, Achola Pala. "Daughters of the Lakes and Rivers."
In Women and Colonization, edited by Mona Etienne and
Eleonor Leacock. New York: Praeger, 1980.

Ominde, S.H. Land and Population Movements in Kenya.
Nairobi: Heineman, 1968.

Rogers, Everett. The Diffusion of Innovations. New York:
Free Press, 1962.

Modernization Among Peasants: The Impact
of Communication. New York: Free Press, 1969.

Sangree, Walter. Age, Prayer and Politics in Tiriki, Kenya.

London: !Oxford University Press, 1966.

Sheffield, James, ed. Education, Employment and Rural
Development. Nairobi: East African Publishing House,

Strange, Bertha F. The Story of Kenya's Home Extension
Service: 1962-65. Washington, D.C.: U.S.A.I.D.
Mission to Kenya, 1965.

Staudt, Kathleen. "The Umoja Federation: Women's Cooptation
into a Local Power Structure." Western Political
Quarterly 33, 2 (July 1980).

Swanson, Burton, and Rossi, Jaffar. International Directory
of National Extension Systems. Urbana: University of
Illinois, 1981.

Wagner, Gunter.i The Bantu of Nort Kavirondo.
London: Oxford University Press, 1949.

2 vols.

Waterston, Albert.
Press, 1965.

Development Planning: Lessons from
Baltimore: Johns Hopkins University


Women's Roles and Gender Differences in Development: Cases for Planners
Prepared by the Population Council.

The Nemow Case, by Ingrid Palmer
February 1985 $6.75 ISBN: 0-931816-16-5

Sex Roles in the Nigerian Tiv Farm Household, by Mary E. Burfisher and Nadine R.
February 1985 $6.75 ISBN: 0-931816-17-3

Agricultural Policy Implementation: A Case Study from Western Kenya, by Kathleen
February 1985 $6.75 ISBN: 0-931816-18-1

Kano River Irrigation Project, by Cecilia Jackson
Pub. date July 1985 $6.75

The Ilora Farm Settlement in Nigeria, by Heather Spiro
Pub. date July 1985 $6.75

The Impact of Agrarian Reform on Women, by Ingrid Palmer
Pub. date July 1985 $6.75

The Impact of Male Out-Migration on Women in Farming, by Ingrid Palmer
Pub. date July 1985 $6.75

Kumarian Press offers a discount on purchases of full sets of these volumes. For more
information, to ask for our catalog or to place an order, write or call

630 Oakwood Avenue, Suite 119
West Hartford, CT 06110
(203) 524-0214

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

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