5.6 Selection of cooperators

Contents - Previous - Next

5.6.1 Scheduling of Tasks and Key Actors

The final step in identifying individuals to be interviewed or individual fields for experimental purposes is the selection of participating farmers. This selection is made by FSD team members but often may be improved by consulting with local authorities, agents, etc. The importance of a team approach to this selection (i.e., interdisciplinary selection process) cannot be overemphasized.

Participating farmers need to be selected at the beginning of every season or at the start of any new research initiative. Whether these participating farmers or cooperators are part of a repeat core of collaborators in the village or are new selections is a question that each FSD team must resolve based on the needs of the research programme.

5.6.2 Details in the Selection of Cooperators

The cooperating unit may be a dwelling unit, a farming household, or specific members within a household. A representative farmer is one member of a group of farmers within a recommendation domain, having similar circumstances for which one can make more or less the same recommendations (see Section 4.5).

In selecting farmers to participate in formal interviews or in hosting trials, the FSD team may use several criteria. As with the selection of research areas and research villages, the most important criterion for choosing farmers is their representativeness. The farmers selected must be as representative of the area's average farmer as possible, to allow for extensive use of the results. However, they also may be chosen to be representative of farmers in a particular recommendation domain or target group. In the latter situation, farmers with certain characteristics are preferred. These characteristics may be: the gender of the head of the household, a wealth factor (e.g., number of cattle owned), the type of draught used and whether the draught is owned, specific characteristics of the type of land farmed or animals owned, etc. These factors can be used to stratify a sample to ensure representation by both male and female heads of households, by wealthy and poor farmers, etc.

For certain types of work, problems can arise from using a farm household (i.e., a group of people eating from one pot (see Section 4.3) consisting of a unit composed of multiple dwelling units, with their own lands, and linked by extended family and/or other ties) as the research unit. The choice of an appropriate research unit may be an important consideration in selecting cooperating farmers. The dwelling unit may be more appropriate than an entire family compound, if members of the dwelling unit operate somewhat independently in their agricultural activities. This question is related closely to the one of intra-household division of labour and decision making (see Section 4,4.1 ), For some studies, the head of the household may not be the most appropriate cooperator, particularly if some other family member has responsibility for, and decision making power over, the particular activities of interest (Box 5.3).

BOX 5.3: THE APPROPRIATE COOPERATOR MAY NOT ALWAYS BE JUST THE HOUSEHOLD HEAD

Households may contain independent decision makers, thus complicating the issue as to whom attention should be focussed on with respect to different technologies, For example, in Senegal, fields under the control of households have traditionally been divided into common and individual fields. The common fields controlled by the household head provided food for all members of the family. Increasingly, because of stresses and strains in complex household units (i.e., traditionally composed of more than one nuclear unit -- husband, spouses and other dependents), an increasing proportion of fields is controlled by individuals, However, because of poorer access to labour and improved inputs -- which are usually channeled through the household head -- yields of crops are often lower on such fields than on the fields under the control of the household head, For example, with respect to groundnuts, Venema [1978] found that yields were lower on fields controlled by individuals (i.e. particularly when they were women) than on those controlled by the household head). He attributed this partly to lower labour inputs and poorer timeliness with respect to field operations (i.e., particularly dates of planting and first weeding),

Consequently, in such a situation, strategies for improving the productivity of groundnuts perhaps would need to be differentiated according to who controls the field, thus influencing the choice as to who is the most appropriate cooperator. Obviously' however, in the process of such selection, it is important to have the explicit support of the household head, if the main cooperator becomes someone other than the household head himself/herself.

The interest, willingness, and ability of farmers to cooperate are also important selection criteria. Farmers may be interested and willing to cooperate, but they may not have the resources to participate, or they may face conflicts in the use of resources if they do participate. Also, the head of household may be willing to participate, but not some other members of the family who will face an adverse situation if the family participates in the FSD effort. Thus, FSD teams selecting farmer participants must make sure that all members of the household, who may be affected by the household's participation, are willing to cooperate.

Logistical considerations are also important in selecting trial or interview participants. To facilitate visiting participants, particularly at their fields, the FSD team may choose to select farmers from one or two geographical groupings of fields or select participants to reflect groupings or communities within the village. Farmers belong to communities within villages that are linked by family, social, and/or economic ties. Participants in such communities are often located in the same general geographic area within the village, and selecting several community members not only will reduce logistical problems, but can provide additional information on linkages between households for shared resources, work, consumption, etc.

Once the FSD team decides on the criteria for farmer selection that are most appropriate for their situation, they must contact the potential cooperators and make their choice of cooperators. Most farmers will be somewhat suspicious of, and curious about, researchers entering their village. Because of the complicated social structure of most small villages, it is often important to identify village leaders and seek their support or approval. Village leaders who may provide necessary or helpful support include traditional leaders, such as the village headman, and community leaders (e.g., an influential school teacher). Gaining the support of such community leaders will often:

Suggestions of ways researchers can gain local leadership support include:

Although it is important to obtain the support of the village leadership, it is also important that the FSD team itself has a high degree of autonomy to avoid such leadership having undue influence on the types and levels of activities undertaken, the farmers involved, etc.

Farmers to be surveyed or to host trials usually are selected from some kind of a list of farmers in the village. Devising such a list is often a challenge (see Sections 8.4.4 and 8.5.3).

Once an FSD team has selected a group of farmers to participate in a survey or to participate in an on-farm trials programme, there is a tendency to work with the same group over a period of years. This has both positive and negative effects. Working with the same farmers provides information over time, which is useful in identifying changes in the system. On the other hand, the farmers may become atypical of the average farmer in the area because of their contact with the FSD team. Thus, it may be important to change the group of cooperating farmers for some surveys (e.g., particularly single purpose surveys) and for some trial work. Whether a change in cooperators is necessary depends on the objectives of the trial or survey and numerous other factors, such as whether the farmers are still representative. It is also important to consider updating the list of farmers from which cooperators are chosen, either randomly or purposively, on a regular basis. This is particularly relevant it. the list is several years old and if substantial in- or out-migration has occurred in the village.

Finally, in selecting farmers for on-farm trials, it is important to match farmers with trials that suit their farming system in terms of resources, preferences, and management abilities. How well a farmer will carry through on a trial may depend on perceived benefits associated with the trial. Thus, the more appropriate the trial is to the farmer's circumstances, the better are the chances that it will be completed,

5.7 Descriptive/diagnostic

The objectives of the descriptive and diagnostic activities are to:

To put it more simply, in order to introduce change, it is first necessary to understand what farmers are doing, and why they are doing it in a particular way. Then, in the diagnostic activity this information can be used to identify areas within the farming system where improvements can be made.

5.7.1 Scheduling of Tasks and Key Actors

Descriptive and diagnostic research begins even during the study conducted to evaluate and select sites for research. FSD team members also will want to exploit all pertinent information already collected at the time FSD work is initiated. However, many formal descriptive and diagnostic studies will begin after research sites have been identified fully.

Descriptive and diagnostic research is never completed fully (Box 5.4). However, because FSD work must proceed to subsequent stages of research (i.e., design and testing of technology) as rapidly as possible, formal descriptive and diagnostic research is likely to peak in the earliest phase of FSD work in an area, At a minimum, however, informal descriptive and diagnostic research continues to re-evaluate the hypotheses formulated before technologies are designed, Descriptive and diagnostic research also can be on-going for special subject areas even years after work was started.

In the classic model, FSD team members generally have written up descriptions and diagnosed priority constraints. Station-based or other visiting scientists have assisted with diagnostic activity in selected subject areas. Far too often, descriptive and diagnostic research has ended with narrow disciplinary oriented perspectives of constraints (Box 5,5). It is imperative that all disciplines, including farmers, collaborate to provide an integrated analysis at this stage (see Section 10,2). Increasingly, FSD teams are asking farmers to take a more proactive role in identifying constraints (e.g., see matrix scoring and ranking in Section 8,4.4) and potential solutions (see Section 5,8).

BOX 5.4: CONTINUOUS DESCRIPTIVE/DIAGNOSTIC WORK IS NECESSARY WHEN RAPID CHANGE IS OCCURRING

The first diagnostic-stage analyses of the Adaptive Research Planning Team in Western Province' Zambia (ARPT-WP) tended to be formal and deal with the local fanning system as though it were static and as though problems identified today would remain as such into the future [Kalonge and Lof, 1994]. This diagnostic approach failed to account for significant evolutionary changes occurring in the economic, social, and political environment of the country and in the adaptive responses within farming to these changes

Zambia is implementing a structural adjustment policy to deal with its economic problems. One response of farmers to these changes is to diversify their mix of enterprises and move away from an emphasis on previously subsidized cash crops. Several specific changes are noted or predicted maize farmers face a sharp increase in fertilizer prices; rice growers face declining profit possibilities because of high transportation costs; and new opportunities are emerging for cassava production.

To address research needs in this changing environment, the ARPT-WP is developing an approach with emphasis on informal and or-going diagnostic procedures Explicit assessments using participatory rural appraisal (PRA) methods are incorporated into the routine activities of farm research groups (see Section 9 8 6) The purposes of this diagnosis are to monitor changes in the environmental circumstances of farmers and in household adaptations to these changes and to evaluate or forecast the impact of technological interventions in these changed circumstances. The aim is to be able to address issues for both the near and long term

5.7.2 Procedures for the Descriptive and Diagnostic Stage

These consist of the following:

Just to identify the problems or areas of potential improvement is not sufficient (Box 5.6). To help farmers solve a problem, a researcher would need to have a clear understanding of the source of the problem.

The information sought during the diagnostic activity usually involves collecting more details on specific questions that come out of the descriptive activity. These details most commonly need to be collected in the field. While examining problem areas in detail in the field, researchers also should try to assess how widespread and severe particular problems are among the target group. Questions like: how often the problem occurs over years, how many farmers it affects. does it have a severe impact on production and incomes etc., need to be answered. This type of information is necessary for deciding what problems or opportunities are most important and for setting the priorities of the research programme. The issue of setting research priorities is discussed in more detail later (see Section 6.5.2).

BOX 5.5: DIAGNOSING AND RANKING OF PROBLEMS CAN BE BIASED

Busch et al [1983] in a survey of 1,400 American agricultural scientists discovered that the major determinant of why they concentrated on specific research projects was that they enjoyed doing them and that demands raised by clients ranked 13th!

Even if researchers do want to respond to clients' needs, the approach is often wrong, Box [1989: p. 61] reports a revealing incident as follows:

"When we had just met, Virgillo stood up and said: 'Lucas, I understand you want to know. You are a scientist and you want to know, But there is only one way to know what I know about cassava. Speak with me; don't speak to me like others did. Ask me about my life and I will tell you about cassava.' "

This lack of truly interactive communication combined with a degree of arrogance, often results in researchers diagnosing and ranking problems differently from farmers, and from each other when more than one discipline is represented, This is well illustrated by Hawkins [1994] based on work in China, Ghana, and Tanzania, Also, of course, women often are likely to rank problems differently from men, Inappropriate problem diagnosis and ranking are obviously undesirable, because the chances of 'irrelevant' research resulting are high.

BOX 5.6: DIAGNOSING THE PROBLEM IS NOT SUFFICIENT

Farmers in Southern Africa often do not plant in a timely manner, even though planting opportunities are scarce. It is not enough just to go to the farmers and tell them to plant in a more timely manner. It is first necessary to understand why they are not doing so. It may be because most of the farmers are sharing draught power, so that they work first on one field, then on the next. Or it may be that farmers do not have enough labour to look after the cattle and kraal them every night, so they let them wander in the bush. Consequently, when it rains, the farmers have to go and look for the cattle. So, to help farmers solve the problem of delayed planting, a researcher would need to have a clear understanding of the source of the problem.

5.8 Design

The design work involves proposing new technologies to address constraints or opportunities identified in the description and diagnostic work. Usually, the process involves the development of ideas and little field work.

5.8.1 Scheduling of Tasks and Key Actors

The design of new technology or interventions follows the identification of constraints and the prioritization of these constraints. An unfortunate and tar too frequent situation occurs when design focuses on ideas that interest particular researchers but are not justified based on identified constraints.

Because there is no simple formula for generating new technological solutions, creative input must be brought to bear in whatever manner possible. In FSD work to date? visiting scientists and cooperating colleagues from other agencies often have contributed ideas to FSD stall.. In many cases, in fact, it is beneficial to have subject matter experts who can best know what is on the shelf in order to make first-round suggestions to FSD team members.

Sometimes off-the-shelf proposals are not appropriate, given farmer circumstances and FSD researchers need to look elsewhere for creative input. Increasingly, forums are being developed within FSD work in which farmers can discuss options and suggest new ideas that could be tested (e.g., see discussion on farmer group methodology under Section 9.8.6).

Experiences indicate that design should be done carefully but should not result in long delays in the FSD process. In most cases, many unknowns about new technological options can be determined in an appropriate manner during testing followed by an iterative test-redesign-test process in which the options are fine-tuned.

5.8.2 Procedures for the Design Stage

The steps involved at this stage are:

After gathering ideas on the priority topics, researchers and farmers need to work together to select the most promising ones for testing (Box 5.7). Researchers can use their scientific knowledge to select the theoretically best options and use the descriptive data to gauge which options are most likely to fit farmers' resources within the target areas. However, the farmers' judgement also will be Important in determining which systems are most practical and socially acceptable within the target area. Thus, the design activity should include farmers, input and interaction among scientists of different disciplines. Generally, this activity is quite informal among scientists, and informal discussion sessions with groups of farmers are good approaches for obtaining farmer input. Gathering a broad range of ideas during the design activities will increase the interest of all parties involved in the testing and will lead to a higher success rate among the technologies that are tested eventually.

Occasionally, it may be necessary to undertake a more formalized trial/survey activity before technologies can be designed for testing. However, these take time and should be undertaken only when absolutely necessary, because they delay implementation of the testing stage. The two types of more formalized activities that can arise are as follows:

BOX 5.7: FARMER INVOLVEMENT FACILITATES RAPID UNDERSTANDING AND SOLUTION IDENTIFICATION

McCorkle [1986; 1989], amongst others, has been involved in a number of empirical studies that highlight, in the interest of efficiency, the need to:

She has also emphasized another advantage of using such knowledge which is that it is not only technical in nature but also is sociological. Knowledge of the latter can be particularly important in deriving locally relevant solutions.

In one paper, McCorkle [1989] gives two examples from her own experience (i.e., striga from Niger, and salty soils in Tunisia) where researchers could have saved a great deal of time and resources through consulting farmers at the appropriate time. In the same paper, with an example from Bolivia, she illustrates the need to complement farmers' knowledge with that of researchers. The Bolivian example involves livestock diseases. An earlier paper [McCorkle, 1986] developed the case for ethnoveterinary research and its potential value to livestock scientists.

5.9 Testing/implementation

Once the best ideas have been selected, they will need to be tested. The ultimate objective of FSD is to produce new technology options that will be used by farmers to increase their productivity and incomes. It is very important that researchers keep this objective in mind during the testing activities, because it determines the type of testing that is carried out.

5.9.1 Scheduling of Tasks and Key Actors

When moving from design to testing, care must be taken not to overextend the testing programme. It is very easy to design more solutions than can be tested in a satisfactory manner.

FSD team members need to be certain that the experimental interventions can be implemented in a manner that will not cause undue risk to environment, farm operation, or community well being. Otherwise, there is usually little reason for delays at the design stage (see Section 5.8). Nonetheless, a firm basis is needed for interventions selected. In the case of divisible inputs (i.e., chemical fertilizers), the appropriate level(s) to test may be in doubt. Where 'best bet' information can be obtained from station-based research or other secondary information, this might be used and testing could begin, Otherwise, technical and economic trials, which are part of the design stage, may be required before proceeding to farmer testing.

Proceeding to farmer testing should be an interdisciplinary team decision. In farmer-implemented trials, particularly within a farmer group format, the decision to accept a new option for testing is really that of the farmers. This acceptance or rejection is part of the testing process.

Evaluation of test results is primarily the responsibility of the FSD interdisciplinary team and the collaborating farmers. The integrated analysis of these results needs to include an assessment of technical feasibility, economic viability, social acceptability, and above all a farmer view point.

The role of station-based researchers and other resource individuals from outside of FSD is often critical in order to complete these evaluations when FSD staff or the farmers lack a particular expertise ,

Evaluation of test results is followed by a decision on further on-farm research or complementary station research and perhaps proposals on necessary adjustment of the policy/support system, which collectively will hopefully eventually culminate in the release and dissemination of new technology recommendations.

BOX 5.8: TECHNOLOGIES MUST BE APPROPRIATE TO THE RESOURCES AND PRODUCTION ENVIRONMENT OF FARMERS

In the semi-arid areas of Southern Africa, it is well known that winter ploughing will help to increase crop grain yields. But farmers who use animal traction do not apply the system, because the soil is too hard for their animals to plough during winter, For them, the option is not practical.

Most farmers know that applying phosphate fertilizer will increase their crop yields. When fertilizer is given to them free, they are very happy to apply it, However, when they have to purchase it, they know that unless the rainfall is very good, the increase in grain yield will not be enough to pay for the fertilizer. So they do not buy it. In some years, it is not profitable.

5.9.2 Procedures for the Testing Stage

Steps involved at this stage include:

To try and ensure acceptability, representative farmers must be included not only in the initial design but also in testing activities. Farmer groups -- both research- and extension-oriented -- can be used to implement testing of technologies by farmers on their own (see Section 9.8.6). The opinions of these farmers need to be carefully considered, and, if necessary, the technology should be modified before it goes to the dissemination phase.

However in drawing up recommendations, it is important to bear in mind that there is a great deal of variation in the natural (i.e.' technical) environment that farmers face and in the socio-economic characteristics or resources they possess. In spite of this variation, standardized technological packages often are recommended. It is not altogether surprising that, where technological packages have been disseminated, many farmers have adopted components rather than the complete package. In such cases, often little advice is available on what farmers should do. For example, should they use top dressing of fertilizer when they don't weed? The return from the limited research resources can be improved by:

Thus, in recognizing the diversity of farmers, FSD can help in developing targeted and conditional clauses for proposed improved technologies. In a sense, these guidelines indicate how greater numbers of farmers can more closely approach the optimal situation.

Contents - Previous - Next