A PROPOSAL FOR
THE INTEGRATION OF WOMEN FARMERS INTO FARMING SYSTEMS
RESEARCH/EXTENSION METHODOLOGY: HOW & WHY
12 March 1985
Susan W. Almy
Adjunct Associate Professor
Center for African Studies
Grinter Hall 470
University of Florida
Gainesville FL 32611
Abstract page 1
The Project 5
Project Sites and Personnel 9
Scheduling and Reporting 11
Appendix A: Women Farmers and Agricultural
Support Services in Kenya 17
Appendix B: FSR/E and Women's Labor Exchange
Appendix C: The Farming Systems in the Njoro
Area and Possible Interventions 21
Appendix D: Preliminary Agreements with Egerton 24
Bibliographic References 28
Curricula Vitae of Florida Participants:
Susan W. Almy 32
THE INTEGRATION OF WOMEN FARMERS INTO FARMING SYSTEMS
RESEARCH/EXTENSION METHODOLOGY: HOW & WHY
The proposed project is designed to integrate women farmers into a
Farming Systems Research and Extension (FSR/E) project in East
Africa, in order to develop better methods for such integration and
to evaluate its effects on technology development and extension and
hence, ultimately, food production. The project will be the joint
effort of five researchers, two from the University of Florida, USA,
and three from Egerton Agricultural College, Kenya, and will take
place over three years and four months in two sites in Kenya, one
highland resettlement and one semi-arid traditional. The project
will seek to answer whether direct collaboration with women farmers
under the FSR/E methodology makes a difference in the types of
technology developed or in the extent of its diffusion and types of
farmers to which it is acceptable. It will also test the advantage
to be gained in working with individual farmers vs. farm work
exchange groups, a form traditional to many East African women. The
project is designed to provide clear evidence as to whether attention
to gender is important to increasing food production through FSR/E
and, if so, the most efficient methods for achieving it.
Linkage with present and developing Farming Systems programs in the
region is essential to project impact. Thus the project has already
been discussed with the CIMMYT/East African Farming Systems Training
Programme as well as USAID's Farming Systems Support Project (FSSP)
at the University of Florida. Permission for active field
collaboration will also be sought from the Kenyan Ministry of
Agriculture by Egerton College. In addition, a small workshop to be
convened four times during the project term to discuss project
methodology and results with participants of other FSR groups will
serve as a focus for the diffusion and adaptation of the project
objectives and methods to other countries in the region.
Back-up support from Egerton College will involve assistance from
its regular Farming Systems team, other relevant faculty, and the
Linkage Committee in working with the Ministry of Agriculture. The
University of Florida will provide technical support from the Farming
Systems program, the Women in Agriculture group, and the Food in
Africa program, and grant administration through International
Programs (the Institute of Food and Agricultural Sciences), in
collaboration with the African Studies Center.
The project budget is USS493,760 for the forty-month period of
August 1985 to November 1988, not including University of Florida
The food situation in Kenya, as elsewhere on the continent, is
deteriorating. Average food production per capital in Kenya between
1977 and 1979 was 94% of the figure for 1969-71; in sub-Saharan
Africa as a whole it decreased to 91% (World Bank 1981). Kenya's
arable land is divided equally among zones of high, medium, and low
agricultural potential (known respectively as zones 2, 3, and 4),
based primarily on a combination of temperature (altitude) and
rainfall (Senga 1976:71,74). Within the last two generations, Kenyan
small-holders have intensified their traditional use of zone 3 and
moved permanent homesteads and cropping patterns into zones 2 and 4.
Increasingly since 1970, the government has supported this move. As
the population density on farms has grown, and traditional permanent
cultivation methods have been applied to marginal areas in zone 4 and
new, inappropriate areas in zone 2, crops and animal yields have
begun to decline. The 1984 drought only accentuated that process.
As an ISNAR report (1981:4) states:
In these circumstances, the greatest challenge is for
agricultural research to develop improved production
technologies to provide for and sustain increased
productivity by' small-holder farming in areas of both high
and marginal potential. This approach should also be
accompanied by a wider application and support of proved
technologies for increased production in the higher
potential areas [zone 23, where the limits of potential
productivity are now far from being reached.
The ISNAR report (pp.13,21-22) underlines that although there exist
such technologies for the higher zones, little research on those
appropriate for small farms as vs. large yet exists, and that this
must be the immediate goal of the 1980's. In order to create or
select appropriate technologies, as well as to succeed in diffusing
them, the researchers must know the needs and goals of the target
farmers. The report calls for the greater integration of
socio-economic research within the agricultural research system to
achieve this goal, including at the farming systems level
(pp.55,72-73). The present project fits within this area of endeavor,
and is based on the assumption that specific and detailed attention
to the gender of farmers is a critical factor in the success of
technological adaptation and diffusion to East African
During the last decade, agricultural ministries in many parts of
Eastern Africa have come to accept that the majority of their
agricultural labor force is female. Dixon (1982) gives figures of
50-70% for the region, and the percentages are higher in the small
1. The original version of this project was joint with Dr. Christina
Gladwin (UF), who is now completing development of an intra-household
decision-making method for identifying technology for Florida women
farmers, and will later seek funding to test it in Kenya.
farm sector, where the men often migrate to paid employmen*
Until the intrusion of a cash economy, women in most places managed
and worked their farms on their own, under the general guidance of
their husband or another male relative who retained final authority
over land and labor, and who contributed by clearing virgin land and
sometimes by making infrastructural improvements such as drainage
ditches and storehouses (Barnes 1983, Clark 1980, Fortman 1982,
Munroe and Munroe 1977, O'Leary 1983, Pals 1983, Sangree 1965). With
increasing population densities and a cash economy, the situation has
become more complex: some men have removed themselves to urban and
large-farm jobs, returning rarely to approve farm decisions, and
others have turned to farming for a cash income, joining their wives
or farming plots independently (Barnes 1983, 1984; Obbo 1980; Spring
et al. 1983; Weisner and Abbott 1977). Nevertheless, women remain
the major source of farm labor and, probably, that of day-to-day
management decisions in the fields.
Although many agricultural policy makers, researchers and
extensionists have come to accept these facts, the way to their use
remains murky (Gladwin, Staudt and McMillan 1984). Because the vast
majority of these people are men, and both regional and colonialist
traditions provide for a separation of the sexes in matters of
business and non-kin encounters, the cases of active integration of
women farmers into agricultural support systems -- extension,
research, marketing cooperatives -- are limited to those few where
women employed by those systems have initiated it (Appendix A).
In Malawi, for instance, Spring (1984a) helped forge a link between
soybean cooking courses and extension-assisted farm trials to teach
the women participants how to grow this new crop; the trials in turn
revealed to the extensionists a number of unexpected problems with
the package of recommendations they had begun to teach male farmers.
The Malawi extension service is now making efforts to increase direct
work with women, advised by the women's unit. In Zaire, Fresco
(1982) developed on-farm trials with women farmers to explore useful
cassava recommendations; she worked with the women simply because
almost all cassava farmers in the region were women. When Haugerud
(1984) joined a Rwanda/CIP potato research program, she persuaded
them to include women potato farmers among their collaborators in
on-farm trials. In Western Kenya, trial projects to involve women's
groups in growing and harvesting firewood began several years ago,
although their first attempts ran into difficulties because they
ignored the fact that men customarily have primary rights to tree
products. Throughout Kenya, U.S. Peace Corps volunteers began
providing technical assistance to women's groups' productive
activities in 1984, although the drought precluded immediate work in
agricultural activities (Barnes 1984a,b).
There are doubtless many other specific examples of providing for
access of farm women to agricultural support systems. But the
development to date, except perhaps in Malawi, seems haphazard' and
personalistic. Many male agriculturalists remain convinced that the
farming systems are changing to include more men farmers, who will be
more easily influenced to use new technologies, or that it is more
proper, feasible and efficient to influence the women farmer through
her husband, who is, after all, the ultimate decision maker. The
former conviction does not seem to hold up overall: instead,
increasingly more rural men are seeking urban jobs to supplement
meager farm earnings and/or to use their formal educations. The
latter conviction needs to be tested, if the direct inclusion of
women in agricultural programs is to gain the full support of policy
makers whose primary concern is the national food supply. Is the
technology adapted in collaboration with women farmers different from
that adapted in collaboration with men? Do farm trials established
with women farmers reach and convince a wider or different segment of
the population than those with men? Or is working with women farmers
simply an equity issue?
At the same time, adequate methods of working with farm women need
to be explored and evaluated. Men's and women's communication
networks are often quite different: different places and functions of
meetings, different conventions of discussion, different sizes and
compositions of groups. It may be found that concentrating on
methods appropriate to women will produce ones more appropriate to
the majority of men (as vs. male influentialss") as well: the Malawi
group (Spring 1984a) noticed that women did not convert recommended
spacing correctly into practice, and distributed marked sticks to
measure the distances, a tactic which from their other data (Spring,
Smith and Kayuni 1983) would have benefitted male farmers as well.
Other extension methods -- such as the utilization of the traditional
women's labor exchange groups (Appendix B) -- will be more specific
to one sex. If reaching women farmers directly can make a
significant difference to raising agricultural production, then
methods of doing this that can be utilized by men as well as women
workers must be identified and tested.
The Farming Systems Research and Extension (FSR/E) methodology is
considered most suitable for the exploration and testing of these
questions. First, it is becoming widely accepted internationally as
an important interface between the technology developers and the
users, that is, between researchers, extensionists and farmers.
Second, it was developed to enable the adaptation of agricultural
research to farmers' needs and local conditions. Work with the
extension system alone, as in the T & V system now being applied to
Kenya, cannot help if the technology existing is inadequate to the
target population, and a FSR/E project serves both to develop
adequate technology and to ensure that extension improvements are not
being masked by inappropriate content. Third, several countries in
the Eastern African region have begun to install their own FSR/E
systems, and a tested women's component has an opportunity to be
incorporated by them before their own methodologies have become too
In collaboration with Egerton Agricultural College, Njoro, Kenya, a
project has been developed to meet these ends by integrating a
women's component into the College's on-going FSRE23 program and by
studying its effects. The objectives of the project are to:
1. test the need for the incorporation of women farmers in FSR/E
in terms of its qualitative (differences in type of technology
adapted) and quantitative (differences in numbers of farmers
adopting the new techniques) effects;
2. develop and test methods for incorporation of women farmers
that will take advantage of gender-linked differences in
communications and work patterns;
3. use project data to convince FSR programs in Kenya and
elsewhere in Eastern Africa to revise their procedures to
ensure the incorporation of women farmers, using project
methods where appropriate;
4. increase Egerton capacity to research and train in FSR/E; and
5. develop and extend agricultural technology improvements
acceptable to the bulk of farmers in two ecological zones of
Kenya is the logical country for this work for several reasons --
the high proportion of women farmers, the strong basic agricultural
research establishment (now increasing its commitment to FSR), and
the presence of the central office of the CIMMYT/Eastern Africa
farming systems training program. Egerton College provided the
researchers for the first demonstration farming systems surveys
carried out by the CIMMYT/EA office in 1977, has developed its own
local FSR program with CIMMYT help, and plans to host an East-West
African workshop to compare FSR strategies in August 1985. The
College, in the process of obtaining University status, also gives
the most field-oriented college-level agricultural training in Kenya,
and supplies the middle rank of agricultural support service staff.
It also collaborates actively with the National Plant Breeding
Station at Njoro and with district veterinary and agricultural
Egerton has indicated that it will negotiate both research
clearance and further collaboration arrangements with the Ministry of
Agriculture once funding is established. Contact with the Ministry
of Agriculture, its regional research stations and their new FSR
teams will be maintained directly, through College channels, and
through the CIMMYT/EA network. Liaison with other national FSR/E
programs in the region will also be developed, in order to stimulate
2. The acronym FSR is used to indicate a primary emphasis on
technology development, with few direct provisions for extension.
related projects, improve methodology, and review and publicize
results. Four small workshops for these purposes will be convened
annually during the life of the project.
The linkage with the University of Florida provides similar
methodological guidance and the opportunity to carry project results
to Western Africa, through the AID-funded FSSP network. (Because of
regional differences both in farm labor force structure and
communication networks, project results are not expected to be fully
applicable there.) The University of Florida is particularly suited
for such project support because of its confluence of tropical,
African, and women's agricultural expertise. The University has the
only Center for Tropical Agriculture in the continental United
States, and major academic, research, and extension programs in
African Studies and Farming Systems. In 1983 the integrative Food in
Africa Program was established at the Center for African Studies, and
the Women in Agriculture group coalesced into a formal program
serving the Centers for Tropical Agriculture, African and Latin
American Studies. Key resource people include Anita Spring, founder
of the Malawi Project; Christina Gladwin, a pioneer in the use of
decision models in the incorporation of US farm women in FSR; Della
McMillan, a specialist in the relation of household dynamics to FSR
in West Africa; Susan Poats, the African specialist for the FSSP,
engaged in developing methodology for women in FSR; and Peter
Hildebrand and Chris Andrews, experts in FSR methodology.
With two important differences, the project will be treated as a
normal FSR/E project, basically following CIMMYT/EA guidelines,
including recommendation domain delineation (in the new site),
exploratory (informal) survey, verification (formal) survey,
farmer-managed trials with researcher supervision, and evaluation
Firstly, at least half of the participating farmers in all phases
of the project will be women who carry out the majority of farm work,
whether or not they are solely in charge. In informal surveys to
provide baseline and problem-identification data, both men and women
will be interviewed separately, together with their spouses, and in
groups. The team will prepare itself to note differences in topics
and depth of discussions in these various groupings. The formal
survey will be addressed to the primary farm worker and analyzed by
sex. Both surveys will include specific attention to which family
members carry out which tasks within the farming system, how this
relates to time demands at home and off-farm, how decisions are made
to modify the system, who supplies the land and inputs other than
labor, and who controls the use of the products. Examples of how
such information will be used and types of interventions that may
result are found in Appendix C.
In deciding on interventions for testing, priority will be given,
first, to problems most limiting to the total production of the farm,
and, second, to problems most affecting women, where anticipated
effects on production are similar. On-farm trials will be carried
out with female and male collaborators. The final list of
interventions chosen for trial by the team will be discussed with
each collaborator, and s/he will choose which to test. The choices
will be used to confirm or reject hypotheses generated from the
diagnostic data that particular technologies are of more concern to
one sex or the other. Yields of the same type of trial will be
analyzed by sex, and also using Hildebrand's (1984) Modified
Stability Analysis technique, which will provide an independent test
of whether male and female farmers falls into different categories
with respect to environment and general management practices, and
therefore need distinct types of technology. Together this data will
provide the answers to whether incorporation of women farmers into
diagnosis and trials produces qualitative differences in technology.
The second difference from the normal FSR/E project is the emphasis
on extension to women. During the informal survey, data will be
collected on women's labor exchange groups in the area (and men's, if
there are any), other farm-related groupings and institutions of
local importance, contacts with agricultural support service
personnel by different family members and their purposes, and, from
the support personnel, important recent extension messages provided
to the area. The work groups will be an a 2riori focus because they
represent a major difference between men and women farmers and have
received little technical or other support (Kayongo-Male 1983). The
formal survey will include a few questions on groups participated in,
education and agricultural support staff contact by family member,
and specific knowledge of a few key extension messages[33. This
information and that on present farming technology will be used as
baseline, to gauge the depth of the male-female extension gap, and to
indicate collaborator farmers. Further field visits will be made to
women's work groups and individual and group extension meetings (both
FSR/E and regular service) to observe interactions among women
farmers and between men and women farmers and agricultural support
Modifications in interaction with trial collaborators will then be
built into the on-farm trials, beginning with group vs. individual
sponsorship. Half the trials with women in the first year will be
carried out with individual farmers, and half with women's groups --
that is, on the land chosen by the group, with group labor and
participation in all visits. If men's groups related to farming
exist or can be formed for the purpose of following the trials, men's
trials will be similarly split. Any other major modifications in
standard FSR/E or Extension communications behavior indicated by the
diagnosis and first year's work will be tested similarly in
succeeding years, as numbers of trials will not permit further
sub-division of the sample. Minor modifications enabling
communication (ones easily adopted with minimal training and costs)
will be introduced simultaneously to all trials.
At each trial visit, numbers of active and passive participants and
3. This technique was used by the Malawi group in preliminary surveys
nation-wide and revealed clear differences in accuracy and level of
knowledge of planting, fertilization and other extension
recommendations between farm men, their wives, and farm women without
husbands present (Spring 1984b).
completion of plot tasks will be recorded for comparison. Final
results and collaborators' opinions of them will be compared. In
addition, spot checks will be made twice a season of the extent of
neighborhood awareness of each trial. Two or three questions
relating to the present appearance of the trial plots and/or some
occurrence in the last agent visit to it will be designed and applied
rapidly and informally to five houses at varying distance from the
site, on a different radial each time to avoid response
contamination. Sex of respondent and housing quality (as wealth
proxy) will be noted. A month after the trial's completion, a third
such check will include questions on relevance of the trial results
for the respondent's own farm. A formal survey after the final trial
season will check on actual adoption of trial interventions, and the
origin of the adopters' decisions to apply them to their own farms.
The extent of interest in and acceptability of trial results by
different farmers according to the type of technology and nature
(gender, group) of trial collaborator form the quantitative data to
test the importance of the incorporation of women farmers in FSR/E.
If the women-oriented or women-tested interventions spread more
broadly, or to different, significant groups of small-holders (eg.,
farms operated principally by women vs. those operated jointly or by
men), then the effort to revise patterns of interaction with farmers
in FSR/E to include more women participants is justified. If not,
present patterns reaching women through their husbands are adequate.
The project is not expected to introduce any new costs into FSR/E
methodology, aside from one-time experimental ones connected to
method testing and evaluation. It is possible that group trials will
require more time than individual ones; if this turns out to be the
case, the time required will be calculated and measured against any
benefits relating to greater farmer participation in evaluation and
to local extension.
SE12ect Sites and Personnel
Because of the novelty of the enterprise and the difficulty of
acquiring necessary research/extension cooperation from the Ministry
of Agriculture without a prior demonstration, it is intended to start
cautiously within the Egerton College area, and spread out to a
comparison site in the second year once the methodology is fully
developed, original participants are experienced in it, and
cooperation with Ministry groups can be negotiated. The second area
should be an environment with some basic technology already developed
for it, with high levels of women farmers, and considered an
agricultural problem area: perhaps in the ambit of the Kakamega
regional research station, where FSR work has already gone on, in the
The Egerton College Farming Systems Team (1983), with financial
support from CIMMYT, is already working in the Njoro area (Central
Kenya), in Mutukanio, a valley at 2220m altitude some 2 kilometers by
10 with varying soil and rainfall conditions. Some 2000 families
purchased 1-2 hectare plots and moved there between five and ten
years ago, coming from widely different regions and farming
traditions. The staple crops are a maize/bean/potato mix, with
vegetables and dairy cattle growing in importance. According to the
team, about 80% of the primary farm workers are women, but they have
communicated principally with the men. A fuller description of the
N3oro site is found in Appendix C. The Egerton/Florida team will work
in a similar area nearby and in close collaboration with the main
The existing team has had to work slowly because all are full-time
teachers in the College, and fieldwork must be done after hours and
during holidays. They have 15 members, from Crop Sciences, Animal
Sciences, Animal Health and Economics Departments, but 7 are
currently away finishing degree courses at the University of Nairobi.
The group carried out informal and formal surveys in 1983 and
1984143, and designed trials in maize-beans fertilization, potatoes,
vegetables, and fodder, but because of the 1984 drought only the
first of these were planted, and few trials survived to maturity.
The results of these three or four trials, however, were enough to
convince neighboring farmers that their land would benefit from
fertilizer. The Mutukanio area is likely to be atypical of the
surrounding highlands in that its people are in the process of rapid
change, both social and agricultural, to adapt to their new homes.
However, its. ecological and social diversity may also provide
benefits in testing agronomic and communication strategies over a
wider range than could be obtained elsewhere.
The sub-unit composing the team for this project will consist of
4. The results of the latter are still in analysis due to the lack of
a computer at Egerton. The present project will provide training and
a micro-computer to be left at the College, which should avoid future
five people, an Egerton agronomist newly returned from his UF
doctoral program, an Egerton home economist involved in a concurrent
breast-feeding study and wanting to concentrate on women's time
allocation, an Egerton animal scientist specializing in small
livestock, a Florida anthropologist/socio-economic research
methodologist who has worked previously in Kenya in rural and
agricultural development and most recently in small farmer and
university programs in Northeast Brazil, and a Florida agricultural
economist who has worked on African refugee and women's issues.
Egerton has only one rural sociologist (no anthropologist) and cannot
release an economist for the project at this time.
Team members will work nearly full time in the project, with up to
10% of their time being provided back to the College for guest
teaching. The College will have to be compensated for the time of
their three staff, so that they can hire replacements on a temporary
basis. Three of the proposed members are female and two male.
The group will work together with the regular College FSR team in
developing and carrying out experiments and evaluations in the Njoro
area. The preliminary draft of the project has been discussed with
the team at Egerton, and they welcome it as a way to expand their own
work. Perdiems are included in the budget for members of the regular
FSR team and for other faculty from soils, animal sciences, economics
and other crop scientists to be involved in the fieldwork on a
Scheduling and ReportinS (If. Chart 1)
The project team will come together in August 1985, carry out
informal survey work in September and discuss the results with the
regular team and other researchers into October. An attempt will be
made to plant some trials in the short rains beginning in October,
weather permitting, although the analysis of technological needs by
sex and preparatory work with the women's/men's groups will not be
complete. Full trials will begin with the long rains in March 1986.
It is estimated that sixty to eighty trials, some on the same farm,
with their attendant data collection needs should be within the
competence of the five-person team in the first year.
Secondary fieldwork related to soils and water problems, time
allocation, farm budgets, extension agents and groups will be carried
out from October to December, and the formal survey of 5% of
households will be carried out in January, when the farmers are least
occupied. The first two-day regional workshop, to debate methodology
to be used in the project, will be held in December/January during
the holidays, and draw on people working on the incorporation of
women farmers into FSR or general agricultural programs, five from
elsewhere in the region (including the Malawi and Rwandan programs)
and five from within Kenya.
Survey results will be analyzed through February and final
decisions made by the team, College and farmer collaborators on
trials for the long rains. Field days and discussion of results will
occur in August-September, and plans will be made for the short
rains. At the end of the short-rain cycle, a second workshop will be
convened of a sub-set of the first, consisting of those participants
actively engaged in similar projects. This second workshop will
evaluate the work group strategy for men and women participants, the
preliminary agronomic results and results of secondary communications
strategies. Recommendations will be incorporated into project plans;
e.g., the group strategy will be continued or not at Site I on this
The Njoro division of the Extension Service already collaborates
with the College FSR Team. It is hoped that throughout the first year
of operations, they can be progressively involved in running the
trials and collecting the on-going data. At the same time, Egerton
will be negotiating with the Ministry of Agriculture to allow similar
collaboration with both Extension and the local research station in a
second site, probably in Western Kenya. Given this agreement, the
project team will divert the majority of its attention to carrying
out the same 1985-86 time-table in 1986-87 in the second site,
beginning with informal survey in October 1986. Whether or not work
groups have proved an effective strategy in Njoro, they will be
tested under the different conditions of the second site during its
The project will continue at the two sites during 1987-88, testing
revised technologies for the farming systems and for communications.
A third smaller evaluative and planning workshop will be held at the
end of 1987. During the third year, when Extension personnel will
0 E n1
S0 I -co
0f4 0 0
to w 1
0 41 cu
w HH 0 92.
Op 4 0 0
0EH C. i Ha- .H H
-H 0 H0
6- 0 U 0 0
o o c Q c H w|
r0 0 -H --O-0
o H rl +1 C > 1 0 o
H H H H
n g d u ri ^- e 4
5 oCo < a or
have taken over much of the on-farm work at the second site, the
project team will give priority to a general evaluation of the
project's effects on technology type, acceptance rates and types of
accepters, and production increases, both through analysis of data
already collected and random formal survey of the two areas. If,
because of the requirements of the T & V System, the Extension
Service is unable to assist, the project will probably have to be
limited to the first site.
The final workshop will be held in November 1988 after analysis of
the results of the 1987-88 crop years. Preliminary drafts of final
reports will be presented by project participants, as will reports by
participants of the previous workshops from other projects. The
final workshop, like the first, will be enlarged to include
representatives from agricultural programs in the region not
immediately involved in incorporating women farmers. Supplementary
funding will be sought at that time to bring in representatives from
regional ministries of agriculture and from agencies outside the
Throughout the course of the project, regular reports and seminars
at Egerton will allow for faculty participation in its design.
Training modules (including slide-tape and discussion outlines) will
be produced on various aspects of the project for use in the
College's teaching and in project dissemination. The software
evolved during the project for analysis of household systems and
flows of agricultural information among households will be documented
and diffused as appropriate. Written reports and presentations will
also be made to the Ministry of Agriculture and the CIMMYT/EA
Reports will also be made back to the University of Florida, to the
Farming Systems Support Project (which serves all USAID Title XII
programs), the African Studies Center's Food in Africa Program, and
the Women in Agriculture Program, both building from and adding to
these programs' broad experience in the areas of small farmers' and
women farmers' development. These programs will provide back-up
support in terms of bibliographic help, advice and occasional site
visitations by faculty experienced in related research, and technical
advice on the software and training modules, some of which will be
suitable for international use in the FSSP's training program.
ApEEpendix A: Women Farmers and Agricultural Support SErvices in Kenya
Since the 1970's the Kenya agricultural support system, and
particularly the research services, have been slowly changing their
orientation from the large to the small farmer. Most of its
scientific agronomic knowledge of food crop cultivation had been
adapted from Northern Europe, and is not suitable for the drier and
hotter lowlands where many small farmers are being forced to move to
find land. Research internal to Kenya until the last decade has also
concentrated almost exclusively on cash crops for export rather than
subsistence commodities. Within the last year (1984), the Kenyan
government has begun to give more attention to farming systems
research and to the links necessary between research and extension.
The conjunction of East African and British cultures that occurred
during the colonialist period created several new traditions in
governmental strategy that were not badly prejudicial to the early
export crop programs, but are so to the present food crop
initiatives. These new traditions included, in particular, the idea
that real work should be done only by men, and a method of program
implementation and training that stressed hierarchical, formal
In a region in which women carried out the bulk of farming, but men
frequently planted and felt responsible for a few prestige crops
convince farm men to take primary responsibility for the new,
cash-oriented crops, especially those which resembled men's crops by
being long-cycle and requiring less daily care. Although women
quickly were drawn into cash-crop work as helpers to their husbands,
it was not usually vital in the introductory years that they
understand in detail what their husbands had been told to do, nor
even be convinced beforehand that it was a good idea. Today,
however, in some areas many women must manage cash crops on their
own, as their migrant husbands can rarely return to help (e.g.,
Weisner and Abbott 1977); thus further advances even in cash crop
production are coming to depend more and more on government attention
to women farmers.
This is most obviously the case in food crop production. Staple
grain crops have always been grown by women in Kenya, as elsewhere in
East Africa. Men helped by clearing heavy bush and sometimes breaking
up the ground, but the women then took over all operations through
harvest, after which the men might return to uproot and burn the
stalks for the next season (Barnes 1983, Clark 1980, Fortman 1981,
Munroe and Munroe 1977, O'Leary 1983, Pala 1983, Sangree 1965). As
improved maize varieties were introduced in the 1940's before an
African extension system had been set up, this crop spread
"spontaneously" (Bernard 1972:81) through the traditional networks,
and also became typed in most areas as a women's crop.
Later attempts to improve maize, millet and legume productivity
have occurred within the framework of the national governmental
system, i.e., through farm men rather than farm women (Brain 1976,
Fortman 1981, 1982, Mbwagwa 1981, Moock 1976, Staudt 1976). This and
the evolution of a cash market for maize and other food crops have
led to a partial breaking down of the sex-exclusion (Barnee 1983),
but more detailed study of the division of labor in particular cases
reveals that this breakdown can be superficial. For example, Pala
(1983) shows that in most households of a Luo area, where maize is an
important cash crop, women continue to do the breaking of soil clods,
sowing, weeding, harvesting, and storing of the crops, while their
husbands primarily help with bush clearance, plowing and granary
construction. Under such circumstances, the male-orientation of the
national extension and research services is likely to be a major
constraint to the attempt to increase food crop production.
That this orientation still exists today has been noted by the
authors cited above in both Western and Eastern Kenya. Staudt (1975)
provided numerical data that farms in Western Kenya managed by women
alone, rather than couples, were much less likely to be visited by
extension agents, recruited to Farmers' Training Centre courses, or
learn about agricultural loans. Women's agricultural labor groups
are the poorest of the registered community development organizations
in the country (Ng'ethe 1983), and hence least selected for training
Research biases are less easy to identify without a specific
study. From the studies cited above on Kenyan farm women, it would
seem that research on improved methods of land preparation and
overall considerations such as higher yields and length of growing
season would appeal more to farm men, whereas research on yield
stability, harvest and processing quality and improved methods for
planting, tillage and weeding would attract women more. But women
will also prefer higher yields given sufficient yield security, and
under certain conditions may favor reducing cropping seasons: a Luo
woman with poor access to a man with plow would be very interested in
land preparation techniques or short-season varieties to circumvent
this lack. The need for sex-linked technology can only be identified
in an action program such as the one here proposed.
Over the last decade, Kenyan agricultural agencies have become
increasingly aware of the need to address women farmers, but the
biases built into the system have changed slowly, and methods for
achieving the goal have been lacking. Today, the Training and Visit
approach being applied to the extension service makes it even more
difficult to make those changes at extension stage, since the
approach stresses key contact farmers and pre-set messages rather
than two-way communication.
To some extent, the lack of women's training and assistance has
been a response to lack of qualified woman-power. Fourteen percent
of the entering class of 1975 at Egerton College, which trains
middle-level agricultural staff, was female, and despite a special
effort women were still only 22% of the 1983 class (Anderson 1983).
Mutiso (1979) states that many women who have been trained drop out
of the extension service because of the conflict between a stable
family life and the constant transfers the service requires.
Egerton, which should have a large percentage of the women
agricultural researchers at the graduate level in the country, had
only 12 women faculty in 1983. Only one of these was in a research
discipline related to women's traditional farming responsibilities,
soil science (Anderson 1983).
The situation of women agriculturists, researchers and
extensionists in Kenya raises a range of important questions. First,
are there significant elements within the food-crop farming systems
that are being missed primarily because of the greater difficulty of
men establishing communication with women about a women's crop? Such
a possibility is made more likely by another inheritance from the
colonial system, a dependence on lecture-style teaching in extension
service classroom and baraza. This teaching method is used in
women's Community Development courses (Kayongo-Male 1983) and in the
Farmers' Training Centres (Barnes 1984b). Women, who in general tend
to avoid men's meetings or at least leave the talking to the men, are
likely to be even more inhibited under such methods from presenting
their opinions on proposed agricultural changes than are the men.
Second, are the women-headed farms (22% across the country,
according to a survey reported in Barnes 1983) excluded from useful
crop information and inputs under the present system? Staudt, Moock
and Mbwagwa have answered yes for certain places and topics, but
broader information is needed on that. Also, women farm managers may
be excluded from direct information and inputs, but do they obtain it
indirectly from relatives and friends? In the same way, their needs
for research may or may not be voiced by (male) relatives who receive
the extension agents or attend the barazas. Is such indirect
communication adequate to diffuse technical information? For
example, when a new maize variety is distributed to men with an
explanation of why it must be weeded at certain intervals, fertilized
and harvested differently, how much of this information arrives
intact to the women?
Such second-hand information often arrives distorted, particularly
when it is provided by a non-participant in the activity described.
In Malawi, Smith (1983) noted that the wives of men collaborators in
FSR/E trials often did most or all of the work on them, yet were not
present when the agent explained the procedures to their husbands,
and these would routinely forget some of what they had been told in
relaying the necessary information. In the same program (Spring
1984b), many fewer women farmers than men knew the correct agronomic
recommendations of the Rural Development Programme. It seems likely,
then, that technological information is also not reaching women
farmers adequately under current systems.
Third, 78% of women have a husband at home to receive extension
information: are these women able to tap into that flow for their own
use in their field activities? How often does the husband work with
the wife on the same fields, or do they work at separate activities,
on separate crops or even on separate plots planted to the same crop
(as noted in Gillette 1980, O'Leary 1983, and detailed in Burkina
Faso by McMillan 1983a,b)? If they work separately, to what extent
are they able to use the information and inputs acquired by the
husband on their own fields as well?
AEEP~~id B: FSR/E and Women's Labor Exchange Groups
The use of farmers' groups for on-farm trials is not entirely new
to FSR/E. Trials have been implemented in different places both with
and without the participation of farmer groups as collaborators. In
Guatemala, ICTA's on-farm trials usually occurred through the
participation of farmer cooperatives, which seemed to provide faster
diffusion of improved seeds and practices, and greater access to
markets, labor, inputs, and even sometimes land (Gladwin 1984). In
Malawi, the WID trials were initiated through women's home economics
extension courses, but carried out individually by the students.
However, no one has yet evaluated the difference in results
deriving from use of group or individual collaborators. Will groups
of women open up more than individuals to men and/or women
researchers in the diagnosis phase? Will groups working together on
a joint trial, even if on one member's land, discuss the results more
among themselves, with the researchers, and with other people in the
neighborhood, than will individual women collaborators? Since few
Kenyan cooperatives have women members, and even fewer women's
cooperatives exist, will the less structured women's work groups
traditional to Eastern Africa extend their functions to work to
resolve problems of market, labor, and input access delineated in the
Work groups are mentioned in numerous ethnographies from Kenya and
throughout Eastern Africa, groups of three to ten or more women who
regularly exchange labor for planting, weeding, and harvesting. No
further description or analysis is ever given; they are as invisible
as the woman farmer was two decades ago.
In recent decades, in many areas, the groups have begun to hire
themselves out locally for farmwork to earn cash; some have worked to
raise money to buy a joint plot on a settlement scheme, as in Njoro,
or a store in a town (Wachtel 1975). In Kiambu District, Ng'ethe
(1983) found that 25% of the registered community development
(Harambee) groups were women's work exchanges. The different types
of Harambee groups are set up to contribute work and funds to
community education and health projects, and to joint economic
ventures such as crafts shops, poultry runs and marketing. They are
registered by the government, which sometimes provides training in
organization and technical aspects and assistance with non-local
The groups vary in size from a dozen to thousands of members, and
in formality from business-like organizations to groups of a few
friends. The majority of the women's work groups are on the small
and unorganized end of the scale, and receive no assistance,
technical or otherwise, from government. Technical training for
women has tended to focus on leadership and home care and on educated
women from non-farming groups (Kayongo-Male 1983). Staudt (1978b:451)
observed in Western Kenya that farm women in areas with active work
exchange groups were much more innovative in agriculture: information
and stimulus about new techniques passed rapidly through the groups.
It is this capacity that we hope to build on.
AEEfBdix C: The Farming S stems in the Naoro Area and Possible
Njoro lies well within the western highlands region once
monopolized by European plantations, in an area specializing in
wheat, maize and dairy cattle. Much of the immediate area is still
in in large farms, held both privately and in public reserves for
forest protection, seed production, and future expansion of
small-holder settlements. The Mutukanio Company is one such
settlement, carried out under government auspices by some two
thousand farm families which bought shares of a hectare each in the
Company between 1973 and 1978. The agricultural adaptations made in
the Mutukanio Valley can be significant in guiding the future
sttlements planned by the government as land pressures elsewhere
increase. The following description stems from the Egerton (1983)
report and from my own previous experience in Eastern Province and
observations in loco in November 1984.
Mutukanio lies at about 2200-2250m along a narrow valley. Before
the sale it supported extensive grazing as part of a larger ranch.
Part of the valley floor is taken up by a asphalt road along which
small shopping centers are rapidly being developed by local farm men
(including, in 1984, new seed-fertilizer outlets). Most of the land
is sloping to some extent, and part of the valley floor suffers from
poor drainage. The soils are slightly acidic, less red and more
sandy than those to the east of the Rift, and more prone to erosion
(viz. Ahn 1977). This erosion has been accelerated by the original
decision of some of the new settlers to cut down all vegetation,
although replanting is beginning to take place. The top of the
steeper ridge bordering on the valley is protected by a government
forest belt, which also furnishes grazing and wood for local use.
Rainfall is normally year-round, with peaks over 30mm average a
month in April and August and sloughs under 20mm in December,
January, September, and October. The major cropping season is the
long rains in April to August; a tendency to dry spells in June hurts
yields in this period. Secondary plantings may be made in August or
late October. Normal droughts in Kenya follow a cycle of one in every
eight to ten years, and Mutukanio, at only 940-990 mm annual
rainfall, must suffer from them. During the major 1984 drought most
of the Valley's production was lost, but parts of the western slope
received enough rain to harvest maize. There are apparently no
irrigation systems. Due to the elevation, temperatures are low, from
15.5 to 18 degrees centigrade average from July to January.
The typical enterprise mix consists of maize, beans, English
potatoes, dairy cattle, hens, kales, cabbages, and tomatoes. Some
farms have wheat, sheep or goats, or other additions. Over half the
land goes to the first two in an intercrop; the maize is a
government-released hybrid purchased yearly, and the beans are a
mixture of local and commercial varieties. Seed potatoes are both
purchased from a government company and selected and propagated on
the farm, and sometimes are grown year-round. Livestock is usually
cross-bred with European lines; cattle are owned by the men and often
herded cooperatively with several other farmers, away from cultivated
areas, but they also eat crop residues. The kales are local
varieties propagated by cuttings, while other vegetable seed is
purchased. Varietal and propagation problems cause late blight in
potatoes and black rot in kale; the beans are attacked by rusts and
aphids, and stalkborers and cutworms are occasional problem in the
maize. Chemical control in crop protection, weeding and storage is
not used except for occasional treatment of future seed corn. The
farmers have previous experience with chemical fertilizer but have
felt the land is too recently opened to need it; the 1984 Egerton FSR
trials changed that opinion for some. Manure goes to the vegetable
After the 1983 diagnosis, the Egerton F5R team identified possible
interventions in the areas of varietal and planting improvements in
potatoes, kales, and fodder, cattle hygiene, relay cropping of early-
and late-maturing maize varieties to escape the June dry spell,
maize-bean intercropping trials, and relay cropping staple crops in
August or October. (Only maize-bean fertilizer trials were carried
out that year because of the drought.) The emphasis of these
interventions was on the farm and its enterprises, rather than an the
farm household, even though many needs of the latter were identified
in the report. Analysis of the same report from the perspective of
the farm manager and worker suggests additional interventions that
should be tried.
The Njoro farm family operates from a strictly limited land base.
Its primary concern is survival, with secondary goals of paying for
its children's education, improving its standard of living, and
possibly in the future amassing enough capital to buy more land
elsewhere. A major nearby urban market (Nakuru) makes marketing a
problem only to those who must sell large quantities of maize at one
time: for this they must go to the marketing board, which pays off
slowly. No credit is available for inputs purchases, so either small
livestock or the food needed for family consumption must be sold.
(Cattle do not seem to be sold for this purpose, so perhaps should
not be emphasized in interventions.) The farm is a long-term
investment: hence present and proposed farming systems should be
evaluated for their effects on soil conservation and improvement.
Seasonality is a major problem in local agriculture. Labor needs
peak in March to May during land preparation, planting and first
weeding, and for some families in August and October with secondary
crop planting. Men help with the land preparation in March and in
April children on school holiday lend a hand, but the women alone
bear the brunt of the weeding in May, and this is the month when
anyone with cash hires others to help. (Women also do the second
weeding and harvesting.) Cash is also needed now for school fees.
But cash also suffers from seasonality. It comes from sales of
milk, surplus maize, beans and potatoes, hens, eggs, sheep and
goats. Maize, beans, and potatoes are all in short supply at this
season, and the milk money, coming regularly, is probably absorbed by
routine household expenses year-round. Other drains on convertible
produce have already been made in September for more school fees and
in March for any input purchases. Better on-farm storage facilities
and/or small livestock care are thus bound into the resolution of the
May labor (and food) shortage. Svaings schemes by women's groups
that would allow them to sell smaller amounts of food and animal
products outside the critical seasons and bank them together could
also be considered, if production could be improved.
This solution assumes, however, that others locally will be
desperate enough in May to hire out to help weed when they should be
weeding their own plots. A parallel solution would be the better
spacing and reduction of the weeding operation. This might be
achieved by different spacing and intercropping patterns and
intervals and greater cover, including possibly the relay cropping of
early- and late-maturing maize varieties proposed by the Egerton team
and leguminous green manure intercrops. The settlers come from a
variety of farming traditions and it should be possible to isolate a
number of intercrop patterns from the already existing mix for
analysis and comparison with respect to labor needs and timing.
Effects on residue for livestock feeding and on the following
season's land preparation needs should also be assessed. The
original plan for 1984 proposed such intercropping trials but only in
order to measure yield effects, and the idea was postponed,
apparently because of secondary importance.
Finally, it is not clear from whence domestic water comes, how much
labor is consumed in acquiring it, and whether carried water is used
to save key crops in dry spells. Water is the women's
responsibility, though the men used to be. responsible for digging
water channels. On these lines, the possibility of group water
storage and gravity flows for supplementary irrigation in dry spells
needs to be explored.
Ahn, Peter M. 1977. "Erosion Hazard and Farming Systems in East
Africa". In Sgil Conservation and Management in the Humid
Trpics, D. J. Greenland and R. Lal, Eds. Chichester: John
Wiley & Sons. Pp. 165-176.
Anderson, Mary B. 1983. "Kenya: Egerton College: a Case Study".
Cambridge MA: Harvard Institute for International Development.
Barnes, Carolyn. 1983. "Differentiation by Sex among
Small-Scale Farming Households in Kenya". Rural Africana
1984a. "Women Farmers in Kenya". Speech to
the Social, Agricultural and Food Sciences Group, University
of Florida, December 10.
1984b. Personal communication.
Bernard, Frank. 1972. East of Mount Kenya: Meru Agriculture
in Transition. Munich: Weltforum Veriag.
Brain, James L. 1976. "Less Than Second-Class: Women in Rural
Settlement Schemes in Tanzania". In Women in Africa:
Studies in Social and Economic Change, Nancy J. Hafkin and
Edna G. Bay, Eds. Stanford CA: Stanford University Press.
Clark, Carolyn M. 1980. "Land and Food, Women and Power, in
Nineteenth Century Kikuyu". Africa 50(4):357-369.
Collinson, Michael P. 1982. "Farming Systems Research in
Eastern Africa: The Experience of CIMMYT and Some National
Agricultural Research Services, 1976-81". Ann Arbor: MSU Int'l
Devel. Paper No. 3.
Dixon, Ruth B. 1982. "Women in Agriculture: Counting the Labor
Force in Developing Countries". Po2. and Devel. Review
Egerton College Farming Systems Research Team. 1983. "Report of
Informal Survey of On-Farm Research with Farming Systems
Perspective in Mutukanio, Njoro Agricultural Division, Kenya".
Report to CIMMYT Eastern African Economics Program Nairobi.
Fortmann, Louise. 1981. "The Plight of the Invisible Farmer: The
Effect of National Agricultural Policy on Women in Africa". In
Women and Technological Change in Developing Countries,
Roslyn Dauber and Melinda L. Cain, Eds. Boulder CO: Westview Press.
1982. "Women's Work in a Communal Setting:
the Tanzanian Policy of Uagmaa". In Women and Work in
Africa, Edna Bay, Ed. Boulder CO: Westview Press.
Fresco, Louise. 1982. "Women and Cassava Production: An
Approach to Improving Agricultural Productivity in Rural
Zaire". Report UNDP/FAO: Kikwit, Zaire. November.
Gillette, Cynthia. 1980. "Maize Production in a Traditional
African Society: Economic Rationality Revisited". Human Org:
Gladwin, Christina. 1984. Personal communication.
Gladwin, Christina, Kathleen Staudt, and Della McMillan. 1984.
"Reaffirming the Agricultural Role of African Women: One Solution
to the Food Crisis". Presentation to the AFAA Conference on
Food Security in Africa, Minzini, Swaziland, 19-28 April.
Haugerud, Angelique. 1984. Letter to Susan W. Almy, November.
Hildebrand, Peter H. 1984. "Modified Stability Analysis of
Farmer Managed, On-Farm Trials". Agronomy Journal
1981. A Report to the Government of Kenya: Kenya's National
ASricultural Research System. The Hague, Netherlands: ISNAR.
Kayongo-Male, Diane. 1983. "Helping Self-Help Groups Help
Themselves: Training of Leaders of Women's Groups". J. Ea. Afr.
Research and Devel.:13:88-103.
Mbwagwa. R.K. 1981. "Adult Education in Agriculture: a Case
Study of Taita/Taveta District". Afr. J. Sociol.
McMillan, Della. 1983a. Changing Patterns of Grain Production
in a Resettlement Scheme in URRer Volta. Washington DC:
Southeast Consortium for International Development. Center for
Women in Development.
1983b. Land and Farming Systems". Warming
Systems ProK2et Newsletter 1(1). USAID.
Moock, Peter. 1976. "The Efficiency of Women as Farm Managers:
Kenya". Amer. J. Agqr Econ.:831-835.
Munroe, Robert L. and Ruth H. Munroe. 1977. "Land, Labor, and
the Child's Cognitive Performance among the Logoli". Amer.
Mutiso, Roberta M. 1979. "Career-Role/Family-Role Conflict
among Women Agricultural Extension Officers in Kenya". J. Ea.
Afr. Research and Devel. 9(1/2):67-77.
Ng'ethe, Nyuguna. 1983. "Politics, Ideology and the
Underprivileged: The Origins and Nature of the Harambee
Phenomenon in Kenya". J. Ea. Afr. Research and Devel.
O'Leary, Michael. 1983. "Population, Economy and Domestic
Groups the Kitui Case". Africa 53(1):64-76.
Pala Okeyo, Achola. 1983. "Women's Access to Land and their
Role in Agriculture and Decision-Making on the Farm:
Experiences of the Joluo of Kenya". J. Ea. Afr. Research
and Devel. 13:69-87.
Sangree, Walter H. 1965. "The Bantu Tiriki of Western Kenya".
In Peoples of Africa, James Gibbs, Ed. Pp. 41-80.
Senga, W. M. 1976. "Kenya's Agricultural Sector". In
Agricultural Development in Kenya: An Economic Assessment,
Judith Heyer, J. K. Maitha, and W. M. Senga, Eds. Nairobi:
Oxford University Press.
Smith, Craig. 1983. "Relevance of Agricultural Station Trials
to Women Farmers". In Women Farmers in Malawi: their
Contributions to Agriculture and Participation in Development
PErgJegt, by Anita Spring, Craig Smith and Frieda Kayuni.
Lilongwe: Women's Agricultural Development Project USAID.
April. Pp. 44-51.
Spring, Anita. 1984a. "Trials and Errors: Using FSR to Reach
Farmers Who Are Often Neglected". Speech to the Social,
Agricultural and Food Sciences Group, University of Florida,
1984b. Profiles of Men and Women Smallholder
Farmers in the. Lilongwe Rural Development ProEct, Malawi.
Final Report of the Women in Agricultural Development Project
in Malawi to USAID. Gainesville FL. March.
Spring, Anita, Craig Smith, and Frieda Kayuni. 1983. Women
Farmers in Malawi: their Contributions to Agriculture and
Participation in Development ProlEcts. Lilongwe: Women's
Agricultural Development Project USAID. April.
Staudt, Kathleen. 1975. "Women Farmers and Inequities in
Agricultural Services". Rural Africana 29 (Winter):81-93.
1976. Agricultural Policy, Political Power
and Women Farmers in Western Kenya. PhD dissertation.
1978a. "Rural Women Leaders: Late Colonial and
Contemporary Contexts". Rural Africana 3:5-21.
1978b. "Agricultural Productivity Gaps: a
Case Study of Male Preference in Government Policy
Implementations". Development and Change 9(3):439-458.
Wachtel, Eleanor. 1975. "A Farm of One's Own: The Rural
Orientation of Women's Group Enterprises in Nakuru, Kenya".
Rural Africana 29:69-80.
Weisner, Thomas S., and Susan Abbott. 1977. "Women, Modernity,
and Stress: Three Contrasting Contexts for Change in East
Africa". J. Anthro,. Research 33:421-451.
World Bank. 1981. Accelerated Develgoment in Sub-Saharan
Africa: An Agenda for ACTION. Washington DC: IBRD.