Target and research area selection:...
 Identification and development...
 Planning on-farm research
 On-farm research and analysis

Group Title: Intra-household dynamics in farming systems research : the basis of whole farm monitoring of farming systems research and extension : a position paper
Title: Intra-household dynamics in farming systems research
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00071937/00001
 Material Information
Title: Intra-household dynamics in farming systems research the basis of whole farm monitoring of farming systems research and extension : a position paper
Physical Description: 16 leaves : ; 28 cm.
Language: English
Creator: Flora, Cornelia Butler
Publisher: s.n.
Place of Publication: S.l
Publication Date: 1984
Genre: non-fiction   ( marcgt )
Statement of Responsibility: by Cornelia Butler-Flora.
Funding: Electronic resources created as part of a prototype UF Institutional Repository and Faculty Papers project by the University of Florida.
 Record Information
Bibliographic ID: UF00071937
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 13274380

Table of Contents
        Page 1
        Page 2
    Target and research area selection: Recommendation domains
        Page 3
        Page 4
    Identification and development of a research base
        Page 5
        Page 6
    Planning on-farm research
        Page 7
        Page 8
        Page 9
    On-farm research and analysis
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
Full Text





by Cornelia Butler-Flora

Chair Technical Committee, FSSP


Professor of Sociology

Kansas State University

Manhattan, KS 66506

Farming Systems Research and Extension has as its goal increasing

agricultural welfare of the farm household as well as increasing the

food capabilities of regions and nation-states while protecting the

natural resource base. In order to do this effectively, procedures

must be derived that come up with recommendations that farm families

will adopt and which do not disadvantage in the long term or short

term members of the family or particular segments of society.

The major contributions that farming systems research has made

to date particularly with farmers involved in market oriented,

commodity based cropping systems has implicit within it a limited

number of methodological choices, based on static assumptions regarding

intra-family provision of the input to the farming system, including

land, labor, capital, and management as well as a sytematic and equal

division of the benefits of overcoming production constraints within

the household. This paper and the project it is to fit into hope to

examine those assumptions and to try to give practitioners of

farming systems research and extension more alternatives when they

begin to address the implementation of farming systems research

and extension projects in the field. By doing so we hope that

the costs and benefits of various choices at each stage of the

design and implementation of farming systems research and extension

can be made explicit, thus allowing for farming systems research

to be even more effective in reaching its goals.

The methodological goals of this paper are completeness and

parsimony. The case studies to be designed, which this paper hopes

to inform, will attempt to address the issues involved in integration

of the division of labor and resources within the household by

gender in terms of the minimum amount of information that needs to

be recorded and analyzed in each stage of farming systems research:

(1) Target area selection (delineation of subareas or recommendation

domains), (2) Problem identification and development of research

base, (3) Planning on-farm research, (4) On-farm research and

analyses, (5) Extension of results and (6) Evaluation of the farming

systems research and extension activities. The hope is that

consideration of the whole farm at each stage of the process in

a more explicit fashion will highlight the interactions between

sub-systems, particularly those that vary by gender, leading to

more effective recommendations in program implementation. In this

exercise we hope to identify key decision points and to present

alternatives at those points where understanding of intra-familial

division of resources might impact microlevel programs choices

from designation of domains to laying out of field trials to

evaluation of field trials to extension programs. This paper will

attempt to suggest some of those decision points and alternatives

that might be involved.

The paper is only a beginning. It is a hope that other

scholars and practitioners of farming systems research and extension

can help contribute to this effort to conceptualize decision points

and alternatives so that case study material can be designed

illustrative of those choices to be made. In this paper I draw

particularly from the work of Kathryn McKee and her presentation at

the third annual Farming Systems Symposium, "Methological Challenges

in Analyzing the Household in Farming Systems Research: Intrahouse-

hold Resource Allocation."

Target and Research Area Selection: Recommendation Domains

The population to whom recommendations apply are defined

agronomically, topographically, politically, and economically.

Generally the areal designation of the project is determined by a

political process. That area in turn then delimits the eco system

constraints that exist. Within a given area, there are different

types of farmers, just as there may be different cropping patterns

and soils of different fertility and potential. Recognition of

the different types of farmers or farm families and their potential

helps in defining the kinds of solutions that are possible once

major constraints are identified.

Beyond the all important constraint of social class, which

relates to size of operation, tenure status, etc., the recognition

of differential family composition is highly related to farmers

access to the factors of production including land, labor and

capital. What is the household composition? Are households headed

by men or women? In the case of female headed households, are

they defacto heads, i.e., at that moment their husbands are gone

due to temporary off farm migration, or dejure female headed,

i.e., no male is legally connected or responsible to that house-

hold. This suggests different location of decision making and

different access to resources as well as off farm remittances.

In determining target populations, the recognition of degree

to which labor is interchangeable or noninterchangeable by sex

and the degree of labor flexibility throughout the production

cycle is crucial. For example, women farmers may have access to

totally different markets than male farmers even though they raise

the same crop. The different markets to which they have access

in turn may very well determine the qualities they seek in the

crops they produce and the degree to which certain innovations

will be acceptable or unacceptable to them. Totally different

recommendations might be given to women producing cowpeas for a

local market versus men producing cowpeas for a regional market.

The interactions of enterprises by gender of the family

members involved is important. For example, do women feed the

cowpea residue to the animals they may raise for household

consumption, savings and fuel? Do men plow under the residue

from their cowpeas? In such cases, the meaning of crop residue

is different for the two groups within the same household because

of the differences of male based versus female based enterprises

and different interactions among them within the same household


Differential access to inputs and markets by gender,

differential enterprises and enterprise mixes by gender, and

differential task allocation by gender should be considered as

part of household composition that can distinguish recommendation

domains. The importance here is not that highly complex data be

generated in orderfor individual solution derived for each farm.

Instead the researchers should know the choices they are making.

For example, a project to increase cowpea production may in fact

have to choose the male system or the female system. It cannot

be assumed that improving one will improve the other.

Identification and Development of a Research Base

Once the kind of farmers to be reached within the target area

are identified, the problems and opportunities of those farmers

must be analyzed. It is important at this stage that the potential

differentials in access to resources by gender and access to

markets by gender be included in the analyses of constraints and

potential solutions to them. For example, in some areas where

drought is a problem water may be a major constraint to agricultural

production. This is overcome by timely planting, which is facili-

tated by timely plowing. Timely plowing in turn requires access

to animal traction, which requires access to animals immediately

after the first rains. When plowing is done by men and animals

controlled by men, female heads of households are disadvantaged.

Often the tradition of male plowing comes from the traditions of

herding animals in the off season, involving rounding up the

animals, who are often weakened by the hungry period, taming them

enough to submit to the harness. Getting large spans of cows,

bulls and oxen to pull together requires a great deal of physical

strength. Women are disadvantaged in terms of both tradition and

strength, yet are increasingly the ones responsible for planting.

In contrast, when there is a market for meat and ability to feed

animals through grazing and stall feeding, including throughout

the hungry season, stall feeding might be institutionalized. That

maintains the animals strength and gentleness, allowing women to

then plow, increasing their ability to produce more grain and feed

during the next hungry season and to improve their status. To

what degree are constraints differentially affecting the different

target population including male farmers compared to female farmers,

male enterprises compared to female enterprises within the same

farm household? How are constraint solutions going to affect either

of these two kinds of enterprises according to whether a man or

a woman is involved?

Labor availability is often a major constraint. Constraint

solution must involve attention to the different demands of different

enterprises within the farm household at different times of the

year. Flexibility of labor by gender related to the different on-

farm and off-farm enterprises need to be taken into account in order

to see the trade-offs of solving a constraint for one enterprise or

cropping system within the total farm versus its potential impact on

the whole farm.

Differential uses by gender of farm household resources may

make overcoming a constraint differentially possible for male

versus female farmers. For example, using manure to increase

soil fertility may conflict with using manure as fuel. Overcoming

a constraint in a male system may increase the constraint in the

female system. Thus a whole farm consideration of particularistic

problems within that part of the enterprise a project chooses to

address is particularly important.

Problem identification must be made taking into account the

tools the farming system practitioner has to offer, which are

often commodity based, in the context of the whole system which

by the nature of limited resource farmers tends to be multi-

commodity based in order to reduce risk and equalize cash flow.

At this stage the decision on what information to collect needs to

be very carefully thought out. In the past social scientists have

erred by collecting data on "everything". One of the goals here

is to collect information on the correct things that will best allow

an early decision on recommendations to implement in the next stage.

Planning On-Farm Research

Shaner has five steps to planning on-farm research, all of

which should involve whole farm considerations, including differential

resource availability by sex.

(1) Reviewing the priorities given to problems and opportunities and

hypothesis for solutions:

Different enterprises with a farm may have different problems

and give different priorities to their solution. Solutions are

not always seen in terms of maximizing the profit (indeed, farm

may only be marginally market-responsive), but may be in terms of

reducing demands on the rest of the farming system. Focus there

must be on interactions among the enterprises, which may vary by

sex. The review of the priorities should involve discussion with

both male and female members of the farm household as priorities

are worked out.

(2) Reviewing research findings:

What is known about the potential solutions to the constraints

in terms of technology that is on the shelf? What is known about

past differentials in the impact of that technology on different

enterprises within the household? In particular this stage must

pay attention to unanticipated effects of the introduction of such

problem solutions on other enterprises within the farming system.

(3) Seeking whatever help is needed and available from regional,

national and international sources:

Availability of technological solutions to market based problems

that involve the male farming enterprises may be much greater than

those for females and thus easier to implement. The need to look

harder for the solutions to the female based problems should not

deter the seeking of that technology. For example, there may be a

great deal of information available on wheat or barley varieties,

which are often a male crop. There may be very little information

available and the potential variety adaptation or changes in

cultural practices because of potential positive impact of the

total farming system.

(4) Deciding whether to accept current environmental conditions as

they are or to attempt some degree of change:

Are the required inputs available? Often lack of rural

purchasing power has slowed the development of input markets for

them in rural areas. If a policy recommendation to change this

is to be made, such recommendations should include recognition

that women farmers as well as male farmers need access to these

new resource pools.

Perhaps special channels to insure female farmers access may

be necessary. If changes in conditions are to be changed, are

there specific mechanisms to include women? The planning of on

farm research is particularly important, because this is the

point at which the decisions on what data to gather in order to

measure progress and evaluate the results is important. Data-

gathering decisions in the past have tended toward focusing on

field-specific results, rather than whole farm-specific results.

It is important that at this stage the planning of the on-farm

research include measuring the allocation of family labor over

time, its flexibility and availability at different stages of the

agricultural cycle. Monitoring data should include resource

control by sex and their availability at different points in the

farming cycle. Monitoring on-farm trials should include analysis

of economic and noneconomic incentives that impact the acceptance

of particular innovations. For example, the results of on-farm

trials are generally expressed in terms of yield per hectare or

net monetary gain per hectare, using farm gate prices at harvest.

More sophisticated measures should be built into the design that

take into account the interactions of different enterprises within

the farming system and the differential implications of those

interactions by sex. Allocation of resources by sex and the

differential incentives by sex within the household may determine

whether the labor needed will be provided. Will demands for

weeding a male crop interfere with planting a female crop?

Monitoring must include the usual attention to agronomic factors

but should also include attention to such factors as labor provision

by sex and, when possible, management input by sex.

On-Farm Research and Analysis

Once the research has been designed, selection of the farms for

on-farm research is crucial. Finding farmers who are willing to

lend their fields or animals to this process at times seems difficult.

Researchers often feel more comfortable convincing farmers who have

similar characteristics to the researcher than those who are very

different. Female farmers are often implicitly felt to be different

from male researchers. Generally we find that no matter what the

proportion of female farmers in a particular area, even in enter-

prises where women producers dominate, within a farming system that

may be male dominated, participating farmers tend to be males rather

than females. Only if female farmers as well as male farmers

participate in on-farm research can the unintended implications of

the technology recommended be truly understood differentially by

gender. Some farming systems practitioners have gone so far as

to suggest that we simply do away with the survey analysis

altogether and simply get on with the on-farm trails in order to

find out what the real constraints are. Thus it is crucial in

the on-farm research analysis that farmers of different charac-

teristics and implicitly different access to resources and

differential access to markets be chosen. The female farmers must

be chosen in large enough proportions that the existence of

differentials by sex can be determined or the possibility that

they do exist be eliminated.

Although the selection of the crop or animal system to be

addressed may be already biased by sex, it is important to look at

both male and female fields when attempting on-farm research for

researcher managed trials, and farmer managed trials. Contact

with female farmers may be difficult, because the researcher relies

on already established linkages, particularly the extension

service, in identifying farmers. The bias of extension toward

male farmers, which extends from the bias of educational systems

toward training males in agriculture and females in home economics

skills, is well documented. Thus practitioners must be aware that

there is a choice in terms of the sex of the cooperating farmer

and that choice may influence the degree to which the results of

the on-farm trials are generalizable back to the recommendation


With both male and female farmers, monitoring on-farm trials

must include both the direct and indirect outputs that stem from

the interaction of the particular crop or animal system being

experimented upon and other enterprises or subsystems within the

household. These indirect interaction effects may ultimately

determine the large acceptance of the recommendation, the increase

in yield that might be anticipated, or increase in net income from

that particular crop. Complete identification of those system

interactions that would have occurred in the previous stage must be

continued in the monitoring that takes place. Finding measures

for these interactions may be difficult. Even in monitoring on-

farm trials on male farmer's fields data should be gathered from

female members of the farm household and the converse should be

true in on-farm trials with female farmers. An example cited by

Kate McKee is useful illustration:

With the introduction of higher yielding cereal variety,
more labor will be demanded for harvesting, processing
and storage activities (which generally account for high
share both of total farm labor expenditure on the crop
and of total value added from the crop). If non-family
labor cannot be hired because of unavailability or
unaffordability, the women and children in the household
may be expected to provide most or all of the additional
labor, since as a rule they are considered to have primary
or sole responsibility for tasks such as food or processing
and storage. Yet this exacerbated labor peak may coincide
with other major demands on their time, such as harvesting
of another crop, pursuit of off-farm employment, or fuel
wood and water collection. If the additional labor demands
are fulfilled by these family members, production of other
essential household goods and services may thus be reduced.
Mechanization can create an untenable situation. The use
of machines and/or draft animals to perform land preparation
tasks (which are often the responsibility of adult males)
may permit expansion of land under cultivation thereby
creating additional labor demands for unmechanized weeding,
harvesting and processing tasks (which are often the domain
of women and children).

In the on-farm trials, particularly farmer-managed trials,

the points where the farmer does not follow the recommended

practices at the recommended time should be particularly examined

in depth. Sexual differentials in decision making provision

compared to labor provision might be highlighted by these incidents.

Who allocates whose labor may help explain why farmer managed

trials, which are always frustrating to the researcher, have the

problems that they do. Differential input by gender should be part

of the follow-up of these on-farm trials.

On-farm trials should help elucidate incentive structures for

the household and within the household particularly in the case of

farmer managed trials, attention should be given to the allocation

of the inputs to the production process and the benefits gained.

Who benefits and who loses in these cases? This is particularly

important when the division of obligations for family maintenance

is highly gender specific. In these cases men and women allocate

resources under their control to activities that best enable them

to fulfill these obligations rather than to the activities that

are most productive from an aggregate household perspective.

Differential extra-household responsibility should be paid particu-

lar attention in the case of polygamous households, but should also

be an aspect of monitoring on-farm trials in other areas as well.

McKee again is helpful in presenting the reasons for monitoring

incentives as part of an on-farm trial process. It also should

be part of the ex-ante design of on-farm research.

The analysis of incentive structures, within the
household helps to explain the abundant evidence of women's
search for and protection of independent income sources as
well as their preference to allocate their labor resources
to activities over which they control the product, e.g.
dairy processing for household consumption and sale, rather
than unpaid weeding of cereal controlled by their husbands.
An example of the distributional consequences of a
new livestock technology may illustrate how the extent to
which a new technology is viewed as an improvement will
vary according to the individual family members perspective.
If men control livestock maintenance and cash sales of
animals, on the whole they would favor intervention which
would increase live weight and quality of stock. Women,
if they receive income from sell of milk, cheese and other
dairy products, as is often the case, would stand to gain
most from the intervention which increase milk quantity
and quality. Both men and women would benefit from
increase calf survival or interventions which permit an
increase in the number of animals under current land and
labor constraints. If improved technologies are to be
developed and tested it is clearly important to understand
who would have an incentive to participate (if additional
labor is necessary to meet cash outlays or if purchased
inputs are required).

Monitoring on-farm testing requires great sensitivity not only

to the statistically manipulable variables of agronomic inputs and

economic outputs but to a larger set of variables involving house-

hold reallocations to the changes in the system, particularly the

differential input demands of the new recommendations. Most on-

farm research to this point has not addressed these issues directly

although in programs such as that of ICARDA we see a shifting of

the indicators used in on-farm trials in response to the

understanding of the household situation that comes from longer

interaction with the farm family and various enterprises that

comprise their farming systems.


Extension of result are often biased toward male agriculturalists.

1 .8

They are also often biased toward larger farmers. This is

complicated by the fact that the extension agent is often little

recognized given little prestige, and is, further, not a part

of the research process. Ideally women extension agents should

be involved in extending the results of farming systems research

recommendations to male and female farmers. Information flow to

female farmers can occur in a variety of ways: (1) Male extension

agents can include women in as part of their larger clientele.

This simple solution to the problem is often the most difficult,

as many societies have strong norms against men interacting with

women. (2) Female agricultural agents can be trained. This is

an ideal solution but often difficult to implement, as the period

of training is long and the motivation for women in many societies

to become agriculture extension agents is small, particularly when

such activities away from home might be seen by parents as

interfering with their daughters marriagability. Getting agricul-

tural colleges to adjust to female students is often more difficult

than getting farmers to change their technology. Thus, this must

be seen as a long term solution to getting knowledge to female

farmers. (3) Perhaps more practical in the short run is to train

the Home Economics extension agents or paraprofessionals in the

agricultural technology necessary for women. These paraprofessional

or female agricultural agents will be highly attuned to the variety

of demands on female farmers and, if they can be given short courses

in agricultural technology, will perhaps be most adept at extending

this technology to female farmers.

t "*> *

I A -r


A number of decisions are made at each stage of farming systems

research and extension that by necessity narrow the focus of the

analysis to make it manageable. Hopefully, this project can serve

to make farming systems practitioners aware of a wider variety of

alternatives at each decision point so practitioners can make a

conscious choice of whether or not a gender biased research or

extension decision will have specific cost and benefits vis-a-vis

the ultimate goal of the project. It is hoped through the case

studies and through further examination of existing literature to

document decision points and provide examples of alternatives that

are possible for the farming system practitioner.

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