W IA ,: ...... .... -- '- --- '" -- ------------.-----
. -L ----.---t -
-at the Uriv sit M F -
GENDER ISSUES IN FARMING SYSTEMS
RESEARCH AND EXTENSION
FROM RECOMMENDATION DOMAINS TO INTRA-HOUSEHOLD DYNAMICS AND BACK:
ATTEMPTS AT BRIDGING THE GENDER GAP
Paper presented at the Conference on
GENDER ISSUES IN FARMING SYSTEMS RESEARCH AND EXTENSION
held at the University of Florida, Gainesville, Florida
February 26 to March 1, 1986
by AMALIA M. ALBERTI*
*Research Specialist, Center for Social Research, Visayas State
College of Agriculture, Baybay, Leyte, Philippines 7127-A
One of the concepts to emerge with the Farming Systems(FS) approach to
agricultural research and extension is that of the recommendation domain
(RD). Defined as "a group of roughly homogeneous farmers with similar
circumstances for whom we can make more or less the same recommendation"
(Byerlee, et al., 1980), the underlying assumption is that farmers of
households within the same RD will have similar responses to proposed
technology (Shaner, et al., 1982: 44). RDs are intended to focus the
research process and expedite dissemination of the recommended technology
thereby facilitating the extension phase.
The debate continues in the farming systems literature and among
farming systems practitioners about both the more relevant criteria and the
preferred timetable to identify and elaborate recommendation domains. The
position of those of us who maintain that the early delineation of RDs
precludes considerations that are not readily evident or initially salient
(Cornick and Alberti, 1985) is countered by others who maintain that the
early identification of RDs permits their progressive refinement (Franzei,
1984). Still others (Norman and Baker, 1984) point out that in the last
analysis both the target groups identified and the nature of the technology
recommended tend to reflect the expertise of the the team members in a
particular Farming Systems Research(FSR) project.
In this paper I take the position that, first, RDs sensitive to gender
issues are difficult to develop due to scant documentation of women's
participation in agricultural and farm-related activities in local areas,
and, second, if developed, difficult to implement due to several features
common to many FSR projects. Indeed, it seems to me that the greater the
pressure for prompt elaboration of RDs, the greater is the likelihood that
women's roles, as well as their concerns within the FSR context, will be
overlooked because of insufficient time to draw them out. The long term
solution to satisfactorily address gender issues in FS, however, lies less
in attempts to develop appropriate RDs and more in efforts to revise the
FSR framework so that gender issues are deliberately and self-consciously
entertained unless excluded. Until these changes occur, several key
questions are proposed to'assist FSR practitioners in assessing what gender
related issues are potentially relevant in a particular FSR site and
whether they can be addressed feasibly within the existing project
OBSTACLES TO DEVELOPING GENDER SENSITIVE RECOMMENDATION DOMAINS
Among the more common techniques suggested for the initial stages of
problem diagnosis leading to the formation of recommendation domains are
reviews of secondary data, informal interviews with persons such as local
officials, residents, and extension workers, and an exploratory survey of
farmers sometimes combined with or followed by a formal survey (Harrington
and Tripp, 1984; Shaner, et al., 1982). The obstacles to uncovering the
extent of women's involvement in the total or select phases of a farming
system imbedded in each of these techniques is discussed briefly.
LIMITATIONS OF TRADITIONAL TECHNIQUES
Secondary Data Reviews
Much of the literature on women in agriculture published within the
last ten years underscores the extent to which the involvement and
contributions of women in this area have been underrepresented (Deere and
Leon, 1981; Lewis, 1981). Nevertheless, secondary sources such as census
data and local agricultural reports that continue to ignore or
underestimate female contributions abound. When FSR staff consult these
materials they are likely to accept the data as factual unless they are
aware of the possibility that female participation in agriculture may be
masked or otherwise distorted. Only when sensitive to this bias may they be
persuaded to seek additional corroboration before dismissing gender as a
potentially relevant variable.
Informal Interviews and Exploratory Surveys
Local officials and extension agents can often provide extensive
site-specific information that a FSR staff member would be hard-pressed to
otherwise obtain so efficiently. For information on female involvement in
agriculture, however, this is less likely the case for several reasons.
First, cultural values may intervene. When female agricultural activity is
associated with poverty, not only are male officials unlikely to discuss
such activity on the part of female members in their own household, but
they may well be reluctant to discuss such activity on the part of female
residents in general presuming that it would reflect negatively on the
socioeconomic status of the community.
Assuming that these local officials and extension agents are almost
exclusively male, attempts to adjust for these gender-related "blind-spots"
by speaking with their wives or other female household members may not
yield substantially different results (Alberti, 1980). To the extent that
these women partake of the elevated social status of their households, they
are unlikely to make "public" their own involvement in farm-related tasks
or imply difficult socioeconomic conditions within the community by
referring to such activity on the part of other women unless it is to
Second, it has been found that male farmers routinely underestimate
the degree and undervalue the importance of female involvement in
farm-related .activities in which they too participate (Bourque and Warren,
1981; Deere and Leon, 1981; Alberti, in process), and to ignore or be
unaware of the extent of female involvement in farming activities that they
do not share in. Hence asking male farmers about the participation of
females in agriculture will not necessarily elicit accurate information.
Finally we must consider the reluctance of the national male FSR staff
members, especially if they are from the local area, to ask questions that
are deemed "inappropriate" by local standards. Moreover, cultural norms may
restrict male field staff members' access to women for interviews. As yet
another possibility, male FSR staff members may resist interviewing women
because of their own attitudes about female participation in agriculture.
The advantages and limitations of formal surveys have been widely
discussed. Within the farming systems literature Chambers (1980, 1983,
1984) is perhaps their most outspoken and graphic critic as he conjures up
visions of "30 pages of questionnaire ... which if asked are never coded,
or if coded never punched, or if punched never processed ... examined ...
or analyzed..." that a number of us have also seen (1980: 4).
Vis-a-vis women's involvement in agriculture there are two points I
would raise about formal surveys. The first is that preparing a
questionnaire assumes that we know what we need to know and how to ask it.
While this ought eventually to be the case, for surveys conducted during
the initial stages of project development this is not always true for
women's issues precisely as a consequence of some of the limitations just
discussed. .Secondly, many formal surveys are designed to be administered
to either the male or female head of household, but not both. Generally,
the household member available when the interviewer arrives responds.
However, the survey form frequently lacks an item to indicate who was
actually interviewed and whether that person was male or female. Hence,
even if relevant questions about women's involvement in agriculture and
farm-related activities are included, it is impossible to disaggregate male
and female responses and analyze them for consistency'and comparability.
LOCATING WOMEN IN THE FARMING SYSTEMS CONTEXT
Given these constraints, we turn to the issue: What site-specific
information might be readily available that would expedite developing
recommendation domains sensitive to gender differences? By "readily
available" I refer to information that could be elicited over a few cays
through informal conversation with local residents, teachers, and other
persons working in the area, as Rapid Rural Appraisal procedures recommend
(Chambers 1980; Beebe, 1985), in conjunction with a field trip around the
project area. The field trip is essential to provide visual information to
accompany verbal accounts. Lines of inquiry otherwise not considered may
be opened when the information from these two sources does not concur.
The information obtained from responses to the following questions
ought to enable FSR practitioners to contextualize the situation of women
in the FSR setting in broad strokes. At the same time, it would facilitate
a quick assessment of whether the FSR project, as it exists or could
feasibly be modified, can viably address the gender issues relevant to that
site. Where addressing those issues is possible, this would then ideally
be followed by collecting the kind of information needed to inform analyses
cf intra-household dynamics in FSR (Flora, n.d.; Feldstein, 1985).
What are the Local Cultural Norms Regarding Female Agricultural Activity?
Is More than One Culture Represented in the Project Area?
In many parts of Latin America, particularly indigenous regions of the
Andes, women work side by side with men in the fields. In other areas such
as Honduras women are seldom seen working in the fields beneath the direct
rays of the sun and may well be embarrassed if they are. Asian women such
as those from Bangladesh are rarely field workers while many of their
Indian counterparts assume the major role in most if not all phases of rice
The differences referred to here are largely the result of cultural
variations whose dominant mode of expression may be religious or ethnic, or
some combination of the two. What is important is that when we know that a
certain portion or subportion of the population of an area shares a
particular cultural orientation, we are also in a position to make certain
assumptions about the kinds of roles women are likely to assume within an
agricultural setting and how forthcoming information about those roles is
likely to be. For example, if visible productive activity on the part of
women is highly circumscribed, we can expect that even when women do engage
in such endeavors, they will be extremely difficult to document.
When more than one cultural group is represented in an area additional
factors may come to bear on the situation. Is one group dominant and the
other subordinate? Is the participation of women in agricultural and farm
related tasks the same for both groups? Are the norms regarding such
involvement the same? If the norms vary, which norms do agricultural
extensionists and field workers represent?
In culturally complex settings it is important to specify the cultural
group or groups to which a recommendation domain applies. This should help
clarify and explain what would otherwise be unanticipated responses to a
recommended technology. Factors that might be involved include
differential access to extra-household labor by ethnic group, or different
production objectives despite use of the same traditional technologies.
Does Women's Participation in Agriculture Vary by Social Class? If so, in
There is an ever-growing consensus that the way households, and the
women within those households, participate in the farming system is highly
contingent on social class. Women from land poor households who engage in
farming.tasks tend to work longer hours at those tasks and generate
proportionately lower returns.than other women. Oftentimes they are the
women who have been left behind while their male partners migrate in search
wage employment. Women from landless households are clearly the most
vulnerable as they are increasingly dependent on an ever more tenuous
agricultural.wage labor market that relegates them to more restricted and
marginal employment opportunities even as it expands commercially (Hart,
1978; Stoler, 1977; Sen, 1985; Stolcke, n.d.; Young, 1985; Horn, et al.,
1984: Chaney and Lewis, 1980).
While these trends may be widespread they are not universal. Knowing
whether they are valid for a particular setting should give us some clue of
how candid men, or women, or both are likely to be about female involvement
in agricultural and farm-related activities.
Do Women Specialize in Food Production and Subsistence Agriculture?
Despite broad variations in patterns, the preeminent role of women in
in the production of food for home consumption appears to cross continental
bounds (Chaney and Lewis, 1980).
In Latin America the evidence is widespread that the majority of women
who directly engage in agricultural production at the household level do so
primarily with basic crops intended for home consumption though they may
also market small portions of those crops. If the household also raises a
cash crop, it is likely to be under the care of the male head of household,
even when women contribute labor to its cultivation. The more the
household's agricultural activities are commercially oriented, the less is
the likelihood that the women of the household will be directly involved in
agricultural production. However, when laborers are present, women of the
household are usually expected to provide the support services surrounding
food preparation, and are occasionally called upon for managerial
activities (Deere and Leon, 1981; Bourque and Warren, 1981; Alberti, in
In Asia the scenario is distinct. Despite broad variations in the
extent of women's direct involvement in rice-based agricultural economies
due to ethnic and religious differences, women are always involved in the
processing of rice and frequently bear major responsibility for its
transplanting, weeding, and harvesting. When the household's access to
rice fields is insufficient to meet its own consumption needs, women as
well as men are likely to seek work as agricultural laborers with rice as
the preferred medium of payment. Participation in the harvest of kin and
neighbors, if not the planting as well, is another strategy geared to
insure a ration of rice (Hart, 1978; Sen, 1985; Dey, 1985). In each
instance the overarching objective is to obtain food that can be
immediately used by the household.
In contrast with rice cultivating areas, using the Philippines as an
example, the cultivation of cash crops such as coconut and tobacco and
commercial varieties of root crops such as cassava and camote is dominated
by men. Root crops grown for home use, however, are often under the
immediate control of women (Cornick and Alberti, 1985).
Until recently, the situation in Africa presented what had probably
been the most consistent association between crops and gender. Even now,
food crops are grown almost exclusively by women, though some women,
particularly those near urban areas, have begun to cultivate cash crops as
well (Ferguson and Horn, 1985: 3). In contrast, men continue to
concentrate their efforts in cash crop production.
Getting a sense of the pattern that predominates for a given FSR
project should help us to identify the crops and animals that women tend to
work with-as well as to assess the FSR project's capabilities in those
UTILIZING THE INFORMATION WITHIN A FARMING SYSTEMS FRAMEWORK
Having asked the.questions, we now turn to how we may fruitfully use
the information obtained.
First, the knowledge gained should enable us to better identify the
variables that are particularly relevant vis-a-vis women in farming systems
in the project area. Second, it can provide us with guidelines to estimate
the validity of the information and data that does exist. Third, it
highlights the kind of information that is available while giving some
indication of what is lacking. This should help us to assess what
additional information is needed and to appraise how sensitive its
collection may be.
For example, the knowledge that there are two ethnic groups within the
FSR project bounds should immediately prompt us to question whether their
attitudes toward agriculture, and women's involvement in agriculture, are
the same. If they differ we should be attuned to the importance of
systematically distinguishing responses by ethnic group.
The denial of female involvement in agriculture by government and
local officials may be better understood once we know that women engage in
agricultural activities only under conditions of poverty. This knowledge
in turn ought to suggest that we need to exercise caution to elicit the
desired information while avoiding offending the persons questioned.
The shortcoming of these illustrations is that real life situations
rarely fall into compartments that vary so neatly along a single dimension.
Rather, multiple variables combine and fuse, whether systematically or
erratically, resulting in ever more complex relationships. Their salience
is heightened as they interact with some of the more common features of
farming systems projects. Let us examine some of these characteristics and
the way they interact with gender concerns.
FARMING SYSTEMS CONSTRAINTS AND THEIR IMPLICATIONS FOR GENDER ISSUES
Site selection for farming systems projects often results from
political and economic decisions that occur outside project bounds (Shaner,
et al., 1982; Harrington and Tripp, 1984). Marginal areas are less likely
to be selected. Not only do they tend to lack political leverage, but
projects in such areas are more prone to failure due to the restricted
access to resources of their residents. Women who engage in agricultural
and farm-related activities, however, are frequently concentrated among the
resource poor who are commonly located in more marginal areas.
Despite occasional efforts to the contrary farming systems projects
are frequently commodity-oriented either as the result of project mandate,
or team member expertise, or a combination of these factors (Norman and
Baker, 1984). But a commodity orientation is frequently aligned with a
commercial orientation. As has been discussed, however, women are more
likely to cultivate food crops with a view to household consumption.
Hence, when a FSR project has a commodity orientation it may implicitly
ignore women by excluding the crop, or crop focus, of most concern to them.
Lastly, FSR projects tend to adapt already existing technology, or
shelf technology, to a particular situation, rather than to develop new
technology for a specific situation. They justify their approach on the
basis of insufficient resources and a time frame inadequate to allow for
additional research. However, existing technologies have tended to be
capital intensive, and until recently, to give demonstrated results only
when adopted as an entire package, rather than in steps over time. Hence,
to the extent that women who engage in agricultural and farm related
activities are concentrated among the resource poor, they may be unable to
adopt the new technology because of insufficient cash resources. Or if
they have the resources, they may be unwilling to adopt the new technology
because it is inappropriate to their goals when they are subsistence rather
than commercially oriented.
To paraphrase Chambers, then, these factors interlock (1980: 3). Nor
does the conflict end there. As Harrington and Tripp remind us: "Domains
are formed so that researchers can effectively deal with the majority of
farmers in a particular area" (1984: 14). However, the only majority that
women tend to constitute as household level agriculturalists is that of the
rural poor. Nevertheless, even among them, some women are partnered,
others single, some the only farmer in the household, and still others only
sources of labor. Though women who directly engage in farming and
farm-related activities are unlikely to be wealthy, it is likely that there
is considerable variation in their access to resources, even among those
broadly label as "poor".
Women in agriculture tend to share a disadvantaged position in
male-oriented agricultural research and development programs. The way they
experience that disadvantage, however, is mediated by their culture,
resources, and civil status, and, hence, varies. It is difficult for
recommendation domains that depend on homogeneous circumstances in key
variables to locate issues that relate to "women" equally despite their
diversity, for, indeed, there are few. What a true incorporation of gender
issues in farming systems implies is a revision of the farming systems unit
of analysis from the household to the male -and female household heads
within the farming systems household for the stages of problem diagnosis
and design. The information thus provided would enable the farming systems
practitioners to make conscious though difficult choices about where the
FSR resources will be channelled knowing full well and in advance whether
and how those choices are likely to differentially affect men and women.
Alberti, Amalia M.
1979 "Metodologia Apropriada para el Estudio de la Mujer Rural en
los Andes de Ecuador". Paper presented at the workshop
Metodolocia AproDriada Dara Estudiar a la Muier Rural. Quito:
CEPLAES/ Ford Foundation.
1982 "Observations on the Productive Role of Women and Development
Efforts in the Andes". Proceedings of the First Women in
Development Workshop sponsored by HIID/MIT. Cambridge,
In "Gender, Ethnicity, and Socioeconomic Differentiation in the
process Andean Highlands of Ecuador". Ph.D. Dissertation.
California: Stanford University.
1985 "Rapid Rural Appraisal: The Critical First Step in a Farming
Systems Approach to Research". USAID/Philippines.
n.d. "Factoring New Information into Decisions About On-Farm Trials".
1980 "Rapid Rural Appraisal: Rationale and Repertoire in Rural
Planning". Paper presented to the World Bank Agricultural
Sector Symposia, January 1980, under the title 'Shortcut
Methods in Information Gathering for Rural Development
1983 Rural Development: Putting the Last First. London, U.K.:
B. P. Ghildyal
1984 "Agricultural Research for Resource-Poor Farmers: The 'Farmer-
First-And-Last' Model". Paper prepared for the National
Agricultural Research Project Workshop on National Agricultural
Research Management held at the National Academy of
Agricultural Research Management, Rajendranagar, Hyderabad.
Chaney, Elsa M., Martha W. Lewis
1980 "Women, Migration, and the Decline of Smailhoider Agriculture".
USAID: Office of Women in Development. Washington, D.C.
Cornick, T. R., Amalia M. Alberti
1985 "Recommendation Domains Reconsidered". Paper presented at the
Farming Systems Symposium: Farming Systems Research and
Extension: Management and Methodology, Kansas State University,
Deere, Carmen D., Magdaiena Leon de Leal.
1981 "Women in Agriculture: Peasant Production and Proletarianization
in Three Andean Regions. ILO Rural Employment Policy Research
Program: Unpublished Working Paper.
Dey, Jennie M.
1985 "The Concept and Organization of a Research Network on Women in
Rice Farming Systems". In the Report of the Proiect Desian
WorkshoD on Women in Rice Farming Systems. Los Banos,
Feldstein, Hilary S.
1985 "Intra-household Dynamics and Farming Systems Research and
Extension: Conceptual Framework". (mimeo)
Ferguson, Anne, Nancy Horn
1985 "Situating Agricultural Research in a Class and Gender Context:
The Bean/Cowpea Collaborative Research Support Program".
Culture and Agriculture, No.26: Spring:l-10.
Flora, Cornelia Butler
n.d. "Intra-household Dynamics in Farming systems Research: The
Basis of Whole Farm Monitoring of Farming Systems Research and
Extension: A Position Paper". (mimeo)
Frankenberger, Timothy R.
1985 "Adding a Food Consumption Perspective to Farming Systems
Research". Paper prepared for the Nutrition Economics Group,
Technical Assistance Division, Office of International
Cooperation and Development, United States Department of
Franzel, Steven C.
1984 "Modeling Farmers' Decisions in a Farming Systems Research
Exercise: The Adoption of an Improved Maize Variety in
Kirinyaga District, Kenya. Human Oroanization, 43: 3:199-207.
Harrington, L. W., R. Tripp
1984 "Recommendation Domains: A Framework for On-Farm Research".
CIMMYT Economics Program Working Paper. CIMMYT: Mexico.
1985 "Formulating Recommendations for Farmers, Researchers and
Policy-Makers: Issues in the Design and use of Socio-economics
Research". Paper presented at the ESCAP CGRPT Centre Conference
"Towards Recommendations for Research, Policy and Extension:
Methodological Issues in Socioeconomics Analysis of Coarse
Grains and Food Legumes", Bandung, 18-23 November 1985.
Horn, Nancy, Brenda Nkambule-Kanyima
1984 Resource Guide: Women in Agriculture in Botswana. Michigan:
Bean/Cowpea Collaborative Research Support Program.
Lewis, Barbara C.
1981 Invisible Farmers: Women and The Crisis in Agriculture.
Washington, D.C.: Office of Women in Development, Agency for
W., D. C. Baker
"Components of Farming Systems Research FSR Credibility and
Experience in Botswana". Agricultural Technology Improvement
Project, Department of Agricultural Research, Ministry of
"Women Workers and the Green Revolution". In Lourdes Beneria,
ed.,Women and Development: The Sexual Division of Labor in
Rural Societies. New York: Praeger Publishers.
Shaner, W.W., P. F. Philipp, and W. R. Schmehl
1982 Farming Systems Research and Development. Boulder, Colorado:
n.d. "Social Inequality and Gender Hierarchy in Latin America".
Position paper presented at the SSRC Workshop.
1985 "The Creation of a Relative Surplus Population: A Case Study
from Mexico". In Lourdes Beneria, ed., Women and Development:
The Sexual Division of Labor in 'Rural Societies. New York: