Journal of farming systems research-extension

Material Information

Journal of farming systems research-extension
Running title:
Journal for farming systems research-extension
Abbreviated Title:
J. farming syst. res.-ext.
Association of Farming Systems Research-Extension
Place of Publication:
Tucson Ariz. USA
Association of Farming Systems Research-Extension
Publication Date:
Physical Description:
v. : ill. ; 23 cm.


Subjects / Keywords:
Agricultural systems -- Periodicals -- Developing countries ( lcsh )
Agricultural extension work -- Research -- Periodicals ( lcsh )
Sustainable agriculture -- Periodicals -- Developing countries ( lcsh )
serial ( sobekcm )
periodical ( marcgt )


Dates or Sequential Designation:
Vol. 1, no. 1-
General Note:
Title varies slightly.
General Note:
Title from cover.
General Note:
Latest issue consulted: Vol. 1, no. 2, published in 1990.
Electronic resources created as part of a prototype UF Institutional Repository and Faculty Papers project by the University of Florida.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact Digital Services ( with any additional information they can provide.
Resource Identifier:
22044949 ( OCLC )
sn 90001812 ( LCCN )
1051-6786 ( ISSN )

Full Text
I hitc-ratim, FPR With On-Farni Rocarch
I- Nc\\ A-c Extcnsion in FSR
J011011011 Lt7lldCC4'
29 Adjusting and Transt'Cri-ing Technologics J.I. Allita
39 Gcndcr Issucs in FSRE Patricia Ladipo
il Intc-ratin- Wonicii into FSR S. C'bakraborti, ct al.
59 Farm Mall'i-cillclit Int,01-niation S\.Stcllls
69 Informal Rescarch in Ncpal S.P. Cliand and B.D. Gurni tj
81 Establishim, a Local Coordinatim, Nctwork E. Elw -.ouriand L. Ediravds
91 Risk in Tcclinology and Diffiision R.R. Marsh
109 Stistainablc Aoriculturc Polio, Analvscs
T I. Dobbs ct al.
125 Farnicr Participation in On-Farm Trials
J.L. I) Ill.

for Farming Systems Research Extension
Volume 2, Number 2, 1991
Published by
the Association for Farming Systems Research-Extension

Journal for Farming Systems Research-Extension Editorial Board
Timothy R. Frankenberger, Editor Timothy J. Finan
Office of Arid Lands Studies Bureau of Applied Research in
The University of Arizona, Tucson Anthropology
The University of Arizona, Tucson
Peter E. Hildebrand Donald E. Voth
Food and Resource Economics Agricultural Experiment Station
Department University of Arkansas, Fayetteville
University of Florida, Gainesville
Harold J. McArthur C. David McNeal, Jr.
Office of International Programs Extension Service, USDA
University of Hawaii, Honolulu
The Journal is sponsored by:
The Farm Foundation
The Ford Foundation, New York
The Ford Foundation, New Delhi
United States Agency for International Development United States Department of Agriculture University of Arkansas/International Agricultural Programs Office University of Florida
The University of Arizona
The Journalfor Farming Systems Research-Extension is published by the Association for Farming Systems Research-Extension (AFSRE), an international society organized to promote the development and dissemination of methods and results of participatory onfarm systems research and extension. The objectives of such research are the development and adoption through participation by farm household members of improved and appropriate technologies and management strategies to meet the socioeconomic and nutritional needs of farm families; to foster the efficient and sustainable use of natural resources; and to contribute toward meeting global requirements for food, feed, and fiber.
The purpose of the Journal is to present multidisciplinary reports of on-farm researchextension work completed in the field, and discussions on methodology and other issues of interest to farming systems practitioners, administrators, and trainers. The Journal serves as a proceedings for the annual international Farming Systems Symposium from which selected and refereed papers are included. It also welcomes contributed articles from members of the AFSRE who were unable to attend the symposium. Contributed articles will be judged by the same review process as invited articles. Technical Editors: Nancy Schmidt, John Bancroft, and Daniel Goldstein, Office of Arid Lands Studies, The University of Arizona
Design and Production: Paul M. Mirocha, Delphine Keim, Diedre Muns, Robert S. Breckenridge, and Ingrid Downey, Arid Lands Design, Office of Arid Lands Studies, The University of Arizona ISSN: 1051-6786

Journal for Farming Systems Research-Extension Volume 2, Number 2, 1991
I Integrating FPR With Conventional On-Farm Research Programs:
An Example From Botswana
G.M. Heinrich, F. Worman, and C. Koketso
17 New Age Extension in Farming Systems Research: End of the
Technical Message Era
Jonathon Landeck
29 Adjusting and Transferring Agricultural Technologies:
Three Examples from Per(i
J.I. Mata
39 Looking Beyond the Farm for Gender Issues in FSRE
Patricia Ladipo
51 Integrating Women into Farming Systems Research: A Homestead
Vegetable Production Project in Coastal West Bengal S. Chakraborty, JE. Gleason, B. Mandal, and C.S. Das
59 Farm Management Information Systems: A Role for Literacy
Training in FSRE Programs
K.L. Mintz
69 Informal Research With Farmers: The Practice and Prospects in
the Hills of Nepal
S.P. Chand and B.D. Gurung
81 Moroccan Experience in Establishing a Local Coordinating Network:
Researchers-Farmers-Extension Agents
E. Elmzouri and L. Edwards
91 The Importance of Risk in Technology and Diffusion: The Case of
Small Maize Producers in Mexico
R.R. Marsh
109 Sustainable Agriculture Policy Analyses: South Dakota On-Farm
Case Studies
T.L. Dobbs, D.L. Becker, and D.C. Taylor
125 Farmer Participation in On-Farm Trials: The Case of Lowland Rice in
Southern Senegal
J.L. Posner, E. W Crawford, and Mulumba Kamuanga

Integrating FPR With Conventional On-Farm Research Programs:
An Example From Botswana'
G.M. Heinrich, F. Worman, and C. Koketso2
The term "Farmer Participatory Research" (FPR) encompasses a wide array of differing approaches to adaptive-type research (for examples, see Farrington and Martin, 1987 and Farrington, 1988). All of these approaches, however, share the common goal of incorporating farmers more directly into the process of technology development and application. In general, they focus on agricultural systems involving resource-poor farmers, systems that often have been overlooked by more conventional agricultural research approaches (Chambers et al., 1989).
FPR approaches usually are considered to be most appropriate for areas described as "difficult" (Farrington, 1988) or "complex, diverse, and riskprone" (Chambers et al., 1989). However, these approaches might also be appropriate in more well-endowed environments if they increase the efficiency and cost-effectiveness of on-farm research.
FPR is not a replacement for other types of research. Rather, it is most useful as a compliment to more traditional on-station and on-farm farming systems research (FSR) approaches. It can complement more conventional approaches by ensuring that farmers' objectives, priorities, and constraints are understood within the research system; by helping to ensure that research output is relevant and practical within the farm environment; and by incorporating the energies, resources, ideas, and indigenous technical knowledge (ITK) of farmers into the technology-development process.
1 Agronomist and agricultural economists, respectively, Department of Agricultural
2 Research, Francistown, Botswana.
Paper presented at the Tenth Annual Association for Fanning Systems ResearchExtension Symposium, Michigan State University, East Lansing, October 14-17, 1990. This paper represents the opinions of the authors but does not necessarily represent the views of the Department of Agricultural Research, Botswana.

The need to incorporate FPR in NARSs
There is growing acceptance of the fact that FPR could contribute significantly to the development of appropriate technologies for resourcepoor farmers, particularly in difficult agricultural environments. However, this potential impact is not likely to be realized unless FPR approaches are adopted by National Agricultural Research Systems (NARSs) and applied on a sustained basis and on a fairly broad scale within relevant areas (Norman and Modiakgotla, 1990).
Although the concept of FPRdoes not preclude farmer participation in onstation research work, it is assumed in this paper that most FPR activities will take place on-farm. Thus, the most logical way for FPRto be integrated into NARSs is through on-going, on-farm FSR programs. It is further assumed that incorporating FPR in NARSs means integrating FPR with on-farm FSR programs operated by NARSs.
NARSs exist in most developing countries and have better access to information and resources than do farming communities involved in resourcepoor agriculture. For example, NARSs often have linkages with International Agriculture Research Centers (IARCs) and participate in other international information networks. They have input on national policy issues, the capacity to develop applied technologies, and often possess the infrastructure to establish linkages with farmers and extension systems on a national level.
The NARSs stand to benefit directly from FPRapproaches that improve the efficiency of the formal research process (Norman and Modiakgotla, 1990). As the end users of new technologies, farmers have an important role to play in technology development and application (Norman and Modiakgotla, 1990). FPR approaches can provide an effective method for incorporating farmer participation in the technology-development process at the national level. Isolated FPR activities or projects may be useful at a local level, but because of the magnitude of the political power and human and physical resources associated with the NARSs, FPR activities can best achieve a major impact when incorporated into the NARSs.
The adoption of FPR by NARSs may be slowed by two primary factors. First, adoption involves the introduction of new approaches into an institutional setting that may be somewhat resistant to change. Furthermore, FPR requires that farmers be regarded as partners in the research process, rather than strictly as clients. This means that they are empowered with some degree of decision- making capacity and that researchers must surrender some control
Journal for Farming Systems Research-Extension

over the research process. This is not a concept that usually is taught in the formal education process, and, initially, some researchers may experience a difficulty accepting this change. Hence there may be attitudinal and institutional barriers to overcome. Second, while more and more FPR approaches are being developed and tested, they often are supported by external funding and often require considerable inputs of researcher time (Farrington, 1988). These economic and logistical constraints contribute to making FPR programs difficult to sustain in an institutional setting.
In a review of FPR concepts and practices, Farrington and Martin (1987) state that "more case studies are needed where the incorporation of FPR into NARS has been attempted." The following case from Botswana is an example of an FPR approach developed as an integral part of a regional on-farm FSRE program within the Department of Agricultural Research. Following a description of the on-farm program and its development, some issues are discussed that involve the integration of FPR with more conventional on-farm FSR approaches.
AN EXAMPLE FROM BOTSWANA Historical Perspective
Botswana is a semiarid to arid country with all the characteristics of a "difficult" environment for arable agriculture. Rainfall is low (450 mm in the crop-producing areas) and unreliable, and soils generally are low in fertility and moisture-holding capacity. Some 70 to 75 percent ofthe population lives in the rural areas. Rural incomes generally are derived from diverse sources, such as livestock production, crop production, and off-farm enterprises. Despite the risks involved in arable agriculture, more than 95 percent of households in the rural areas of the Tutume Agricultural District, which was studied in this project, engage in cropping activities every year (Miller and Seleka, 1985).
The Agricultural Technology Improvement Project (ATIP) was funded jointly by USAID and the Botswana Government and administered by the Mid-America International Agricultural Consortium (MIAC) through Kansas State University. It was one of a number of farming systems projects initiated in Botswana in the late 1970s and early 1980s. One of its primary functions was to strengthen the capacity of the Ministry of Agriculture to develop and extend applied technology options for increasing farm producVol. 2, No. 2, 1991

tion through farming systems research in the areas of the Central Agricultural Region and Tutume Agricultural District. The following discussion deals with the on-farm research program begun in 1983 that developed in the Francistown area, which is part of the Tutume Agricultural District.
The ATIP began with the classical farming systems research perspective. Although the primary research emphasis was on descriptive and diagnostic activities in the first two years, testing of some promising on-shelf technology options also was begun during this period. Over time, the emphasis shifted to design and testing activities.
Introduction of FPR Activities
Much of the early trials work in the Francistown area was conducted under researcher management (RM), with either researchers or farmers implementing the work (i.e., RI or FI). Researcher management implied that researchers determined the trial objectives and design and directed trial implementation. However, in the 1985-86 season, ATIP decided to examine the potential of one promising technology option ("double plowing;" Heinrich and Worman, 1988) under farmer-managed, farmer-implemented (FMFI) conditions. To facilitate the testing process, a group of volunteer farmers was invited to implement the trials and monthly group meetings were held to monitor progress, to discuss farmer observations through the season, and to deal with problems.
Heretofore, the RMRI trials work had been focused largely on tillageplanting systems for soil-moisture enhancement and improved stand establishment. Although it was known that myriad other constraints existed, researchers viewed soil moisture and stand establishment as the primary local constraints to crop production. Research was restricted to these topics because researcher time was limited and the RM format allowed only a limited number of trials. Too many topics would have spread research efforts too thin.
Twelve farmers participated in this first group activity. During the monthly group discussions it rapidly became apparent that (1) the farmer-participants were quite capable of conducting trials on their own and (2) the farmers had a wide range of problems and interests beyond the topic being tested. Technologies existed that might usefully address some of the problems.
In following years, therefore, these groups were expanded to allow the examination of a wide range of technology options. A detailed description of how these groups operated has been presented elsewhere (Worman et. al.,
Journal for Farming Systems Research-Extension

1987). A summary of group operations follows:
1. Participation was open to all interested participants within a village. The local extension agent was invited to attend all group activities.
2. At the beginning of the season, researchers presented a wide range of technology options (from many sources) for farmers' consideration. Farmerparticipants also were encouraged to suggest areas where there was a need for improved technology options not presented in the list. When this occurred, more options were sought and added to the list.
3. Farmers individually selected options to address their own particular problems and to fit their specific resource constraints. For example, those farmers with soil-fertility problems might select a fertilizer trial, whereas others who controlled their own draft power could select more intensified tillage for water conservation and those who did not control draft power might choose to try a hand row planter to gain more control over the time of planting. Field activities were conducted on an individual basis. The groups were used only for organizational and discussion purposes.
4. Subgroups of farmers who had selected the same options then conducted trials With those items according to a standard, mutually agreedupon trial design. This allowed proper comparisons and some simple statistical analyses of yield results.
5. Where necessary, small amounts of inputs were supplied by the research teams (e.g., 1 kg of seed for each of four cowpea genotypes in a variety trial). The research team also assisted in pegging trail plots where necessary, monitoring the dates of all field operations in each trial, and weighing plot grain yields after farmers had completed harvesting and threshing.
6. After implementation was underway, the whole group met monthly to discuss progress and observations and to deal With any problems.
7. As the harvest period approached, field days were held often to share interesting results with farmers outside the group, extension personnel, and other researchers.
8. After harvest, a postseason survey was conducted with each farmerparticipant regarding each technology option he or she had tested in order to quantify farmers' opinions and perceptions of the technology and to solicit suggestions for improvement.
Within these group activities farmers controlled the research agenda by selecting topics and technologies to test and by identifying areas where more technology options were required. Researchers acted as a source of technolVol. 2, No. 2, 1991

ogy options and gained information for their own purposes through mutual agreements on trial designs that would allow some statistical analysis.3 Evaluation and adaptation of technology options were done jointly in group discussions and through the end-of-season surveys.
These groups have been described under several different titles, but they currently are referred to as Research-Oriented Farmer Groups (ROFGs) The growth of ROFGs was rapid (Table 1), reflecting a high degree of farmer interest in the activity. In addition, the ROFGs became a powerful tool for examining the potential of a wide range of technology options under farmer management and added a great deal of flexibility to the on-farm research program. Furthermore, it is probable that the regular group meetings enhanced trial implementations, as has been observed elsewhere (Baker et al., 1988.)
Table 1. The Growth of ROFG Activities in the Francistown Region, 1985-89
No. of Total Total valid No.of
Year villages farmers comparisons options tested
1985-6 1 12 12 1
1986-7 3 97 44 8
1987-8 3 143 152 6
1988-9 3 128 140 8
Sources: ATIP Annual Report No. 4, p. 105; ATIP Progress Report F87-6; ATIP
Annual Report No. 6, p. 63; ATIP Progress Report F90-60.
A Farmer Participatory Extension Approach
Subsequent to the development of the ROFGs, ATIP was asked to assist in addressing the problem of low agricultural productivity in a Communal First Development Area (CFDA) near Francistown.
After discussions with the CFDA coordinator and extension personnel, ATIP decided to test an approach for accelerating extension activity through
It should be noted that the ROFG's were used primarily as a source of quantitative data regarding farmers needs, interests, and technology evaluations. Collection of hard data was limited to recording plot grain yields so that productivity levels of new technology options could be monitored when employed under farmer management, across
locations and years.
4 The ROFG's have also been called, at different times, options testing farmer groups and researcher-led farmer groups.
Journal for Farming Systems Research -Extension

adaptation of the ROFG approach,which currently is referred to as ExtensionOriented Farmer Groups (EOFGs). A detailed description of this activity may be found in Heinrich et al. (1990).
The EOFGs functioned similarly to the ROFGs, but with several important differences. Because EOFGs were an extension activity, the EOFG leadership was provided by extension personnel, although researchers still participated to some extent in group meetings. Only previously tested technology options were employed (these groups were not used for technical research). Because the groups were designed for extension, no specific trial formats were required and no hard technical data were collected.
The approach was tested in two villages, with participation open to all. Within the EOFGs, farmers specified areas of technological interest, and a range of available options were discussed that might address those interests. Individual farmers selected options to fit their interests and resources and conducted their own individual trials. The group (extension personnel, farmers, and researchers) met monthly to discuss progress, solve problems, and share information and ideas. Field days were held where farmers had the opportunity to present their field work to other farmers and to extension personnel from other extension areas.
The group format made for more efficient use of extension officers' time and facilitated the sharing of knowledge among the participants. The EOFG approach assisted farmers in the correct application of improved technology and resulted in an increase in the adoption of improved technology as indicated by a marked rise in the purchase of various equipment items through the government subsidy scheme (Heinrich et al., 1990). In addition, the EOFGs provided a good forum for NGOs to interface with farmers and define areas where NGOs can participate in addressing local constraints (Heinrich et al., 1990).
The ALDEP five-year plan (1990, personal communication) assessed the EOFGs as an effective extension tool and adopted them at the pilot-project level within extension service.
The on-farm research program developed in the Francistown region comprised three major parts: researcher-managed trials program, ROFG activities (with FMFI trials), and EOFG activities. The role of each and the portion of
Vol. 2, No. 2, 1991

program resources devoted to each are summarized below.
Rescarchcr-Managed Trials Program
Both RMRI and RMFI trial types were used when the technologies being examined were not compatible with farmers' own resources (e.g. design-stage trials involving heavy equipment, such as subsoilers). RMRI and RMFI formats also were used when a number of very specific "systems" were being compared (using both agronomic and economic criteria), requiring fairly precise implementation of the different systems and the collection of hard technical data, or where a number of different factors and their interactions were being compared, necessitating a relatively complex experimental design (e.g. a diagnostic trial evaluating the relative effects of several components of a technology package, such as an improved tillage system, phosphate fertilizer, and improved weed control).
The RM trials section, therefore, was used largely for diagnosis and design, or in early testing-stage trials when hard data were required or cause-effect relationships needed to be examined. With the high degree of researcher input required, only a limited number of research topics could be examined using this format. For that reason the format was narrowly focused on high priority research areas. It accounts for roughly 60 percent of researcher time, but probably only 50 percent of other program resources.
ROFG Activities
This section added a great deal to the overall program. In particular, it provided a format to involve farmers, researchers, and extension personnel in the joint planning, implementation, and evaluation of technology-options trials. At the same time, it facilitated the sharing of both external scientific knowledge and ITK and ensured that farmers' perceptions and constraints were understood. As a result, the range of topics that farmers and researchers could examine was expanded well beyond those topics given highest priority by researchers.
ROFG activities added considerable flexibility to the total program, owing to the ease with which technology options could be moved into the testing program. This flexibility was extremely useful because it allowed the program to respond rapidly to the needs and interests of farmers, as well as to the requirements of on-station commodity research teams for on-farm testing of their products (Norman and Modiakgotla, 1990).
Journal for Farming Systems Research-Extension

This section of the program was used mainly for evaluation and adoption of new technology options under farmer management and implementation. About 30 percent of researcher time was allocated to this section of the program, although it consumed roughly 40 percent of research resources. The amount of information obtained per unit of researcher time was high because farmers themselves conducted most of the work and contributed many ideas.
EOFG Activities
This section of the program dealt primarily with dissemination-stage activities. Appropriately, it was led by extension personnel but provided a forum for continued extension -researcher interaction as technologies moved into the extension domain. This activity absorbed roughly 10 percent of researcher time and a very minor percentage of research resources.
On-farm research programs have a number of clients, principally farmers, extension personnel, and station-based researchers (Norman and ModiakgotIa, 1990). The various linkage mechanisms employed by the on-farm research program in Francistown are described below.
Farmer Linkages
Researchers met monthly with three different groups each composed of 10 to 30 farmers of various resource levels in three different villages. About 90 percent of the farmers attending these meetings were women. Topics discussed covered the general field situation, farmers' constraints and problems, potentially useful technology options, and farmers' evaluations of various technologies under testing. These meetings provide a very strong channel for communications. Farmer field days were held, which allowed farmers not participating in research activities to view and comment on developing technology options. In addition, postseason surveys were conducted to record individual farmer assessments of various technology options.
On-Station and On-Farm Research Teams
Originally, these began as weak linkages Without formal channels of
Vol. 2, No. 2, 1991

communication. With the development of the ROFGs the linkages become much stronger. The on-station commodity-research programs became a valuable source of new technology options for on-farm teams and farmers, while interest in testing technology products on-farm grew among stationbased researchers as useful information was returned.
At present, the Department of Agricultural Research is in the process of organizing multidisciplinary, subject-oriented research groups on-station (e.g. a National Tillage Research Group, composed of a soil physicist, agronomists, a hydrologist, an economist, a statistician, an agricultural engineer, extension personnel, and others). These groups will include both on-station and on-farm research personnel and will be responsible for designing and implementing research strategies, evaluating research results, and formulating proposed extension recommendations. This approach would formalize the linkage between on-station and on-farm research teams.
On-Farm Research and Extension
One of the principal approaches used for this purpose was the annual regional coordination meeting. This was a meeting of regional and district level staff from the research and extension departments, which also included NGOs involved in agricultural activities within the region. The purpose of the meeting was to present rough work plans for the coming year and to allow for discussion and comment before the various participants had finalized their work plans. Areas of overlap in activities among the various agencies involved in agricultural development within the region were identified. This meeting was informally arranged within the region.
Extension personnel were included in the ROFG activities. Local extension agents attended monthly village group meetings and participated in the planning and conduct of field days. Extension personnel at the district and regional level and extension agents from outside the village areas often attended ROFG field days. Research and extension personnel and NGOs collaborated in the development and testing of the EOFG approach.
The combination of ROFG and EOFG activities within the region could provide a continuum of more-to-less researcher involvement and less-to-more extensionist involvement as technologies move through the testing, adaptation, and dissemination stages. The level of farmer involvement could remain high and relatively constant. The concept of a continuum is important because it avoids the breakdown in communication between research and
Journal for Farming Systems Research-Extension

extension that can easily occur when the two are separated in different government departments.
According to the model that has developed, then, the linkage systems within the region could operate as follows: Farmers would have strong participation in the research and extension process through the ROFGs and EOFGs. The research and extension personnel would interact through the annual regional coordination meetings, through joint participation in ROEGs, ECEGs, and field days, and through other activities that might develop, such as jointly organized farmer-training courses and agricultural shows.
Linkages between the regional field work and on-station research would occur through the multidisciplinary, problem -oriented research groups. These groups have the potential to coordinate joint on-farm and on-station activities, as well as to serve as a liaison with extension personnel at the national level. On-station researchers would also have direct but weaker links with farmers though visits to the ROFGs.
Optimally, station-based researchers would be responsible for on-farm research, so that they would have more direct linkages with farmers and the farm situation. However, this has not developed in Botswana, in part because of the "project" nature of farming systems research and in part because limited research facilities and personnel prohibit the decentralized research system that would be required by this approach.
The example from Botswana illustrates several important points regarding the integration of FPR with more conventional, on-farm research programs. As the case study shows, the FPR section of the on-farm research program was successfully merged with more conventional techniques. It did not detract from, but rather added greater capacity, effectiveness, and efficiency to the total program. This was achieved by adding the research capacity of farmers to the available research resources within the department, thus increasing total research capacity. It would seem reasonable that such an addition could be beneficial both in "difficult" environments and in more well-endowed agricultural environments.
The primary purpose of the groups was to include farmers more directly in the technology-evolution process. In the ROFGs, however, trials generally
Vol. 2, No. 2, 1991

were implemented in a manner that would allow statistical analyses of yield results. Some FPR practitioners might argue, therefore, that researchers were still exercising too much control over the testing process. However, on-farm research programs have clients other than farmers (e.g. on-station commodity-research teams) who prefer to work with quantifiable data. Furthermore, researchers may have interests that go beyond farmers' immediate needs in developing their own improved production systems (e.g. the identification of high-impact technologies for priority extension or the performance of crop genotypes over locations and years).
In developing FPR systems, compromises can be reached so that the interests of the various participants are met. For example, an independent study of the ROFGs by the Rural Sociology Unit in the Botswana Ministry of Agriculture concluded that farmers participated in the the ROFGs primarily because the groups were viewed as a useful source of new information and allowed for the practical testing of new ideas (Ntseane, 1988). On the other hand, the research team in Francistown viewed the ROFGs as a strong channel of communication with farmers. This was reflected in the strong and sustained farmer participation over several years, the useful research results obtained, and the subsequent increase in involvement of on-station research programs in on-farm FSR activities. It is more likely that FPR approaches will be sustained by NARSs when the interests of both farmers and researchers can be accommodated.
Institutions, like farmers, are somewhat resistant to change. Similarly, it is unreasonable to expect institutions to make large investments of resources in activities that have not been shown to be strongly beneficial. Thus it should not be expected that conventional on-farm FSR programs in NARSs would invest initially in large-scale or expensive FPR approaches. Continuing the analogy to resource-poor farmers, the types of changes most likely to be adopted by institutions are those that initially require only small changes in the allocation of personnel and resources. Initiation of FPR activities by on-farm FSR teams working with small groups of farmers on a local level is one way in which such changes could be introduced with relative ease.
The overall on-farm research program that developed in the Francistown area evolved in response to perceived needs among researchers (for expanded research capacity) and observed interest among farmers (for a broad range of technology options). The final structure was shaped by local cultural norms. The point is that it is unlikely that one FPR approach developed for a specific
Journal for Farming Systems Research-Extension

situation could be directly applied to another area without some adaptation. As such, examples of FPR approaches developed in different areas are most useful as a source for developing principles and ideas that then can be adapted and applied by local on-farm FSR teams to suit the local situation. The specifics of how FPR approaches operate in on-farm FSR programs must be developed at the local level.
Baker, G., H.C. Knipscheer, and J. De Souza Neto. 1988. The impact of regular research
field hearings (RRFH) in the on-farm trials in northeast Brazil. Experimental
Agriculture 24:281-288.
Chambers, R., A. Pacey, and L.A. Thrupp, eds. 1989. Farmer first: Farmer innovation
and agricultural research. London, UK: Intermediate Technology Publications. Farrington, J. 1988. Farmer Paricipatory Research: Editorial introduction. Experimental Agriculture 24:269-279.
Farrington, J., and A. Martin. 1987. Farmer Participatory Research: A review ofconcepts
and practices. ODI Discussion Paper 19. Overseas Development Institute, London. Heinrich, G.M., and F.D. Worman. 1988. Double plowing: A low-external-input
technology for increased grain production on hardvelt soils in eastern Botswana.
Agronomic Abstracts, p. 55.
Heinrich, G., S. Masikara, S. Magalela, and G. Moremedi. 1990. Testing an accelerated
extension approach in northeast district: The case of extension-oriented farmer testing groups. ATIP Working Paper 25. Agricultural Technology Improvement Project,
Gaborone, Botswana.
Miller,W., and T. Seleka. 1985. Agricultural baseline survey of Tutume District. ATIP
Working Paper 3. Agricultural Technology Improvement Project, Gaborone,
Norman, D., and E. Modiakgotla. 1990. Ensuring farmerinput into the research process
within an institutional setting: The case of semi-arid Botswana. Paper presented at the International Symposium on Natural Resource Management for a Sustainable Agriculture, New Delhi, India, February 6-10, 1990. (Also published as ATIP
MP90-1, Gaborone, Botswana.)
Ntseane, P.G. 1988. ATIPgroups report. ATIP Working Paper 11. Agricultural
Technology Improvement Project, Gaborone, Botswana.
Cropping Systems Resource Management Unit and the Research and Extension Liaison
Office, Department ofAgriculture Research, The Gambia. On-Farm testing byfarmers groups: Firstyear report. 1989. Paper presented at the ARREV meeting. (mimeo). Worman, F., G. Heinrich, S. Masikara, B. Mabongo, and S. Bock. 1987. 1986 Farmer's
groups technology options testing trial. ATIP Progress Report F87-6. Agricultural
Technology Improvement Project, Gaborone, Botswana.
Worman, F., L. Williams, C. Tibone, and G. Heinrich. 1990. 1989 Adoption study:
Spontaneous technology adoption infarmergroups. ATIP Working Paper 34. Agricultural Technology Improvement Project, Gaborone, Botswana.
Vol. 2, No. 2, 1991

The following two tables present examples of the types of technical data collected in the ROFGs.
Table Al presents an example of summary yield data collected in one year (1987-88). Table A2 presents summary yield data collected over several years. These give some indication of technology performance in wet (1987-88) and dry (1988-89) years and suggest priority areas for extension efforts.
Table Al. Mean Yield Data for Selected ROFG Trials Francistown Area, 1987-1988
Yield Yield
improved trad. check % Increase No. of Trial name Crops (kg/ha) (kg/ha) in yield trials
Double Cereals 488' 288 69 32
plowing Cowpeas 181, 130 39 30
Combined 339' 212 60 62
Groundnut Grouhdnut
seed (v. Sellie) 342b 276 24 44
Row planting 80% Sorghum 173 NS 131 32 10
vs. 20% Cowpea
Phosphate 75% Cereals 550' 366 50 8
fertilizer 25% Cowpea
Cowpea Blackeye 171 13
variety ER7 145 13
trialsd TVX 210 13
B005C 318 13
.,b,c Denote significant differences between treatments at the 0.1%, 1%, and 5% levels of probability. NS=not significant. d Includes only trials that tested all four varieties together.
Journalfor Farming Systems Research-Extension

Table A2. Effectiveness, of Various Technologies Used by
Farmer Testing Groups, Francistown, 1986-1989b
Percent increase in yield by year
Technology 1986-87 1987-88 1988-89 Average
Double plow 69 60 85 71
Phosphate fertilizer' 97 50 20 56
Rowplant vs. broadcast 6 32 30 23
Groundnut seed treatment 24 0 12
Effectiveness is measured as mean percentage increase in grain yield per technology option with time
b Crops were primarily sorghum and millet, although some trials also included cowpeas.
Fertilizer was applied at a rate of 20 kg phosphate per ha. d Groundnut seed treatment involved groundnuts only. In 1987-88, treatment was with Captan; in 1988-89, treatments included both fungicides and fungicides plus insecticides.
Source: ATIP Progress Reports PRF87-6, PRF90-2, and PRF90-6, Department of Agricultural Research, Gaborone, Botswana.
Vol. 2, No. 2, 1991

New Age Extension in Farming Systems Research:
End of the Technical Message Era
Jonathon Landeck'
There are three sets of actors in the West African2 farming systems researchextension (FSRE) theater: farmers, researchers, and extension agents, each with specific roles. Farmers are the clients of FSRE researchers and extension agents and their raison d'ctre. They are the ultimate bosses and the final decision-makers with respect to technology adoption. Research and extension serve farmers by generating and exchanging information with farmers to improve their options in the field of sustainable agricultural production.
Farmers do not really need FSRE in order to continue farming. We in the FSRE business, however, need farmers to stay in business. We prefer, of course, that FSRE benefit farmers, but unless FSRE can figure out how to make the rains fall in a timely manner, it likely will take a backseat to subsistence farmers' immediate concern with dailysurvival. Fortunately, most subsistence farmers will agree to produce the crops that we need for our data sets, pro bono. We plan, nevertheless, for active participation by farmers in FSRE under the rubric of on-farm, farmer-managed adaptive research. Assuming that farmers will participate actively in FSRE, how can we as researchers and extension workers better serve farmers' needs for information exchange?
Consider the multitalented agri-socioeconomic researchers for whom FSRE provides an institutional structure. They serve farmers by doing what they do best: they collect data. Without doubt, we need farm-level data. Such data are the earth, wind, and fire of practical information. FSRE researchers
1Department ofAgricultural and Extension Education and Department of Crop and Soil Sciences, Michigan State University; presently Agriculture Program Leader, Cornell
2 Cooperative Extension of Seneca County, Waterloo, New York. The comments in this paper are made with the agricultural systems of Burkina Faso, Niger, Guinea, and Senegal in mind-nations where I have had considerable contact with farmers.

collect data, breathe life into the numbers, create information, write reports, double-check the results, wring their hands, then voila, the technical message is born. But it is the next step in agricultural technology transfer following the birth or renaissance of a technical message that is the most critical.
In the past, the agricultural technology-transfer procedure was straightforward. Researchers would work out the kinks on the research station, develop a prototype, then pass the prospective innovation on to the extension worker. The extension worker would shoulder the technical message, carry it to farmers, offer encouraging words, establish demonstration plots, breathe sighs of relief when the technology worked, and take the heat when the technology was inappropriate. Caught between a rock and a hard place when the technology was slow to be adopted, extension workers became adept at the "good news and bad news" game: The good news is farmers are definitely interested in the technology; the bad news is they cannot afford to use it. The cost of time, labor, and energy frequently rekink the technology-transfer process.
No one likes to hear bad news. We take solace in the general validity of learning and adoption curves. We have become accustomed to qualifying our recommendations for adopting innovative technologies with a proviso that farmers use "best farm-management" practices. In the context of subsistencetype farming systems, where sustainability is an objective, the best farmmanagement practices mayinclude more than just agronomic factors; economic, social, religious, political, historic, and even aesthetic factors can make or break technology adoption. Extension workers are well-placed to understand the importance of these nonagronomic factors in farm-level production, even without a detailed farm survey, by virtue of their access and proximity to farmers. When it works as designed, however, FSRE also is equipped to consider the implications of factors and constraints to adoption besides those of a strictly agronomic nature.
As on-farm adaptive research continues to increase, the technical message carrier, as such, will not be needed as before. Instead, we hope that farmers will be brought to test sites to discuss on-farm research with cooperating farmers. Research results are, more frequently, in farmers' fields, where they can be seen and judged on the spot. In this respect, the technical message cra
Journal for Farming Systems Research-Extension

as traditionally known may be on the wane. Nevertheless, some FSRE systems develop technician-type roles for extension workers, who compile logbooks on the progress of research plots, act as translators in meetings with farmers, collect household data, replicate experiments in farmers' fields unfrequented by researchers, set up demonstration plots, and so forth. These duties are a part of what may be called the Old World methods of extension.
The extension worker as a vehicle for carrying technical messages between researchers and farmers is an idea with which we have long felt at ease. From some researchers' point of view, this is a convenient role for extension workers, who stay out of the way until the crucial work of interacting with farmers on a broad scale in order to popularize a new technology begins. But this is a role that makes vassals of extension workers, at least when it comes to the design and control ofresearch. New thinking in FSRE seeks modifications in this Old World view of extension.
Even among the most reknown FSRE practioners, the issue of extension has been wrestled with on-the-ground and in the literature. Yet, in many forms, the Old World view of extension persists. A comprehensive document on farming systems research (Shaner et al., 1982) describes extension's role as diffusion and testing of research results and technologies. Extension also is invited to participate in research design by contributing knowledge of farmer practices, selecting farmers for trials, surveying farmers, and providing feedback between farmers and researchers. Twenty-six activities are identified for the participation of extension personnel in farming systems research, from selecting target areas to conducting on-farm research and analysis to establishing pilot production programs. These prescribed roles represent a forthright attempt to "let extension in" on the FSRE process. Still detectable in the literature, however, is an underlying sense that extension workers should not be leaders in FSRE, possibly because they cannot compete with researchers in the domain of scientific, technological knowledge. Extension workers thus are trained to take their cues for action from researchers, who decide which technologies are ready for farming populations, even if the technologies are developed and refined off the farms.
In a recent work, Spring (1988; citing McKee, 1984) reviews the FSRE process as a series of five stages: prediagnostic, diagnostic, technology design,
Vol. 2, No. 2, 1991

testing, and final extension. Although Spring (1988) focuses on the problems of gender issues in FSRE, the author agrees with the notion of extension as the final stage in the FSRE process-a vestige idea from the Old World view of extension. By this Old World view, extension is very much akin to an afterthought, rather than an educational activity in which a learning facilitator assumes a leadership (as opposed to "followship") role from Day One.
IWe cannot assume that everyone in FSRE takes extension for granted as an educational activity. Hildebrand and Poey (1985) suggested that on-farm research can be a hands-on learning experience for extension personnel. Their r~sum6 of the nature of farmer-researcher relationships, however, does not address the role of the extension worker relative to farmer-researcher relationships. They discuss farmers' evaluation of on-farm research, perhaps the most important aspect of technology transfer, but not the role of extension workers in such evaluation. Moreover, they neglect the option of beginning evaluation during the design phase of on-farm research, a time when farmers' insights may serve as guides to appropriate research. Herein lies a potential new FSRE role for extension: that is, to serve farmers from the beginning of research design, acting as their advocate of appropriate research by constructively criticizing research objectives before the first plot is laid out. This is the basis of 'an educational leadership role that could redefine the identity of extension in the context of FSRE.
The identity of extension workers as technical message- carriers makes them easy targets when research results are met with indifference by farmers. Alt hough extension is not solely to blame for failures in technology transfer, extension workers do effectively shield researchers from farmers' criticism for such failures. Placement of the onus to succeed on extension workers is symptomatic of an ill-defined role for extension in the FSRE process. Although it is possible that the training of extension personnel is at fault when technology transfer does not proceed as planned, the real weakness in this process more likely is the technical message itself. Thus, an objective of extension training programs should be to teach extension workers to recognize when researchers are on the right track in the design of research for developing technology appropriate to local farming systems.
Development of appropriate technology is the purpose of FSRE, but surveys alone do not necessarily assure that appropriate research and technology will result. Researchers and extension workers who think beyond quantitative characterizations of villages and farming systems as the best way
Journal for Farming Systems Research -Extension

to understand a farmer's view of the world may be of better service to farmers. It is time that we make a quantum shift in FSRE, from studying farmers to informing farmers through educational activities that focus on discussion. Because of their periodic direct contact With farmers, extension workers can conveniently assume the role of educational leaders through discussion with farmers of issues associated with sustainable agricultural production.
EXTENSION IN THE NEW AGE With the FSRE diagnostic survey mechanism, farmers' nondiscrete thoughts and perceptions can be rationalized into categorical tabular results. An understanding of these perceptions by researchers and extension educators is complete only after follow-up discussions of survey results With respondents. In some cases, discussions have preceded surveys with good results. In fact, it usually is difficult to know what to ask in surveys without prior discussions with members of the target population. In any case, surveys and nonformal discussions have complementary roles in FSR.E. My experience in West Africa indicates that conversation and discussion count most with respect to sharing information with farmers. Thus, in defining extension as an educational process, we ought to place greater emphasis on discussion and less on items such as survey-based, detailed farmer and household typologies. These surveys may have a purpose, but they may not be totally indispensable to the design of appropriate research, technology development, and effective extension education.
Recently, in the Fuuta Jalon highlands of Guinea, 32 male and 32 female farmers were asked, "What makes farming difficult for you?" and "What would make farming easier for you?" Although the first question appears to lead responses by presuming that farming is, indeed, difficult, in this case it is appropriate because the researchers themselves had picked up hand hoes and tried farming every few days for several hours. This experience, together With several months of residence in the villages, helped the researchers formulate their questions. Nevertheless, the predominant responses to these questions did not address the physical pain required to produce a crop. They addressed labor issues, such as return for effort expended, supplemental labor supply, and equipment to substitute for labor. These and other open-ended, conversational questions, as well as observations and discussions with other farmers in the region, provided ample occasion to cross-check hypotheses that
Vol. 2, No. 2, 1991

labor was a leading issue in the minds of farmers. It is appropriate in these villages to research any technologies that address issues associated with labor.
A survey might or might not have led to similar conclusions. Nevertheless, FSRE proponents in extension education should consider what constitutes the best leadership tools for educational purposes. Then, in light of limited training and operating resources, we could decide how best to expend our educational efforts. For example, should extension educators shoulder technical messages as their primary mandate? Or can prospective, on-farmtested technologies find their own way from farmer to farmer on the strength of their suitability to local farming systems and farmer visits to test sites? There are many such questions of emphasis and priority for FSR.E.
As noted earlier, agronomic factors are not the only set of decision criteria used by farmers in adopting or adapting technologies to their farm operations. There are economic, social, political, religious, historic, and even aesthetic factors to consider. This is particularly true in nations such as Niger, Burkina Faso, Mali, and Senegal, where natural resources are scarce and where FSRE has a foothold. In fact, ESRE does attempt to address nonagronomic factors of farm-level decision -making. But because of a preoccupation among FSRE researchers and extension personnel for collecting quantifiable data, certain facets of the resource systems within which farmers live are liable to be overlooked in research design.
Therefore, in the design stage of FSRE on-farm research, to complement survey work, extension educators should lead discussions in a public forum by questioning researchers and cooperating farmers as to how a particular agronomic trial will yield results appropriate to the local farming system. "Appropriate" results respond to issues such as human-energy inputs to production, health and nutritional conditions, the quality of the water supply, money-saving measures for farmers, time-saving measures that allow farmers to perform other tasks, the creation of solidarity with the community at large, or tempering the forest spirits, even when these issues do not appear in survey results.
FSRE genuinely tries to address these issues, but a successful effort requires cooperation among all of the FSRE actors. McDermott (1987) believes that the farming systems research model may not provide an effective means by which to close the gap between research and dissemination of tested technology because of poor leadership in defining the mission of extension. McDermott suggests that extension "must develop capacity in technology that will
Journal for Farming Systems Research -Extension

enable it to deal effectively With research." This can be realized by the addition of a "technical liaison and support function" to extension's repertoire. McDermott assigns three major responsibilities to this function, which, with liberty, may be termed: (1) technological literacy on the part of extension personnel, (2) technological marketing by extension through linkages With input suppliers, and (3) technological support for extension workers via appropriate training.
McDermott's vision with regard to technical capacity is correct, and extension training should continue in these directions. But there is a fourth dimension to the technical liaison and support troika that is needed, technological criticism. This fourth responsibility offers extension educators an opportunity to create for themselves a distinct identity-apart from researchers-and to develop a sense of leadership for the purpose of better service to farmers. Moreover, this responsibility for leadership by extension lends itself to cooperative action between researchers and extension educators outside the traditional subservient, technical -message -carrier role.
The term technological criticism does not imply that extension educators should be trained to become cynics of new and innovative technologies evaluated in the FSRE process. On the contrary, technological criticism can be an educational tool for encouraging farmers to discriminate between appropriate and inappropriate technologies before on-farm or researchstation trials are put in place. This would provide a counterbalance in FSRE trial designs. Benor (1987) touches briefly upon technological criticism in discussing the "field -and- farmer orientation" needed by extension workers. Benor's incontestable premise is that extension workers must be in contact with farmers to be effective. Benor rightly observes that one basic question should be kept in mind by extension workers: How will any proposed [research] activity most readily benefit farmers? This question should be absolutely foremost in the conscience of extension workers. Extensiontraining programs oriented to teach prospective extension workers to constructively criticize agronomic research, as opposed to becoming researchers' message carriers, are, therefore, in order. Such a training orientation may point us toward emphasizing and developing an educational mandate in extension rather than maintaining the traditional status quo, the technicalmessage-carrier function.
Waugh et al. (1989) briefly address the need for a reorientation of thinking about the role for extension workers in FSRE. They identified eight more or less traditional activities for extension in FSRE, including characterization and
Vol. 2, No. 2, 1991

analysis of farming regions, supervision of farmer-managed trials, feedback of information from on-farm trials to researchers, and on-the-job learning about new technologies. They clearly demonstrate a trust in extension workers to contribute constructively to FSRE. They also seem to downplay the technicalmessage-carrier role. Possibly Waugh et al. (1988) feel that such duties can be counterproductive when the technical message or the technology related to the message is not suited to the local farming system. They may understand, too, that the adoption/adaptation of technology will occur naturally within and among farming communities, provided the technology is truly on-target, by virtue of farmers making their own rational deductions based upon their observations and discussions with other farmers.
This succes or failure of demonstration plots in this scenario for extension is no longer solely the concern of extension workers. Demonstration plots should be (1) solicited, without coercion, by farmers and (2) established by researchers who take full responsibility for the success or failure of the plots. When extension workers are found alone, next to the corpse of a failed demonstration crop, it does little for their credibility with farmers. Researchers (or their technicians), on the other hand, are trained and paid to know how to grow crops successfully in field plots for purposes of technology evaluation or technology demonstration. Although extension workers should assist in the establishment and monitoring of such plots, there generally are not enough resources to train extension educators to mimic or compete with researchers. Where specific training is necessary, such as for "on-farm proven" animal traction or fertilizer use, in-service training of extension educators and farmer leaders can occur. Otherwise, extension personnel can learn from onthe-job, shoulder- to -shoulder interaction with researchers, provided extension workers are taught to ask the right questions.
Because researchers usually have greater diploma power and bigger operating budgets than do extension workers, their cooperation in this kind of training is necessary. In order to be successful, researchers must avoid the researcher-chauvinist' attitudes that often dominate the relationship between extension educators and research scientists. These postures are especially acute where extension services have not had the historical abd financial support of local and regional communities or of national institutions such as
3 By this I mean the perception that, when all is said and done, agricultural researchers
arc the ultimate authority on questions of what is appropriate agricultural technology
and what is not.
Journal for Farming Systems Research-Extension

universities. In these cases, farmers are optimally served by neither research nor extension and are thus cheated in their role as the clients of FSRE. However, With modifications in the function and structure of extension services, FSRE may be ready for a new era.
In the New Age of FSRE, functional and structural aspects of extension services can be modified to give extension a new lease on life. An initial task for FSRE extension educators is to break the tech ni cal -message -carrier mold used more or less ineffectively over the past 30 years to define and orient extension education. Alternative approaches require that extension educators not aspire to compete with agricultural researchers in terms of scientific expertise. Rather, training programs for extension personnel can focus on the educational aspect of FSRE, which presently is weak in the FSRE wholesystems approach to agricultural development.
Prospective extension educators certainly ought to be trained in fundamen tal agronomic principles and techniques, but it may be as important to teach them to think critically about technology and to ask the right questions in order to train themselves on-the-job, alongside researchers and farmers. In the case of West African nations, thanks to FSRE, we already know a great deal about the major categories of constraints to technology transfer at the farm level: labor and land. These constraints and allied issues are natural points of departure for the development of critical thinking skills through formal training and hands-on experience.
With critical thinking skills, extension educators will have a basic tool With which to glean technical information from researchers, a most useful resource in discussions with farmers. As critical thinkers equipped with fundamental information, extension educators can learn from farmers and researchers. As their knowledge of traditional and nontraditional agronomy increases, so will their ability to serve farmers by acting as public advocates of appropriate research. In the early stages of ESRE, extension educators will then be able to lead constructive discussion and criticism of current and proposed on-farm and research -station technology evaluation. Extension educators might discuss with farmers how the proposed research simulates or differs from local farm management and household realities so that, prior to testing, farmers become aware of the implications of adopting or adapting the technology.
Vol. 2, No. 2, 1991

Discussions such as this are important because, during reconnaissance or diagnostic surveys, researchers, in their bid to obtain data, sometimes forget that farmers exchange information primarily through discussion and conversation, communication modes quite alien to quantitative survey methods. Thus we can expect certain, perhaps crucial, information to slip through the cross-cultural cracks built into formal surveys.
Throughout this paper, I have discussed a functional modification of the traditional, technical- message -carrier role of extension as one way to overcome the propensity of research to consider extension an afterthought in the FSRE4 process and to make better use ofextension personnel. We would now suggest a structural modification in extension likely to be met with skepticism. The structural modification is this: move extension education services from ministries of agriculture (where agricultural research normally is housed in most FSRE-friendly nations) and formally link it with national university faculties. When in the same ministry, the academic and economic power of research scientists allows them to overshadow extension educators in matters of agricultural research policy and action. Given that extension is an educational service, it should be in an educational ministry. I do not believe that the linkage between research and extension would necessarily suffer if such a realignment were made. On the contrary, the relationship may well improve as researchers work to maintain a professional spirit of interagency cooperation.
As university staff members, extension educators will be accorded a formal status associated with knowledge, information, and learning. In addition, this status will identify extension educators as people whose service function is education, as opposed to publicity and sales. A structural overhaul of this nature would require the development of university- level curricula geared to producing four-year degree graduates, the New Age extension educators. Current extension personnel already in the field, whatever their age, would be accorded a two-year leave of absence to complete a degree, toward which two years of on-the-job experience would be credited as experimental learning. Some experienced extension educators, along with regular faculty members in the social sciences, might also be used to train four-year students. The presence of university- linked personnel in the rural areas might also elevate the
4It is noteworthy that, after years of using FSR to designate "farming systems research," we have now tacked on E, as if to say, "Oh yeah, and extension, too."
Journal for Farming Systems Research-Extension

status of agriculture in the eyes of the nation.
This educational approach to New Age extension in FSRE is one to consider as we move into the next century of agricultural development worldwide. By the year 2000, I hope that a new profile of extension is drawn into the landscape of FSRE and sustainable agriculture, one that will distinguish clearly the technological development role of research, the educational role of extension, and the farmers we serve.
Benor, D. 1987. Training and visit extension: Back to basics. In W.M. Rivera and S.G.
Schram, eds., Agricultural extension worldwide: Issues, practices and emerging priorities. London: Croom Helm.
Hildebrand, P.E. and F. Poey. 1985. On-farm agronomic trials in farming systems research and extension. Boulder: Lynne Rienner Publishers.
McDermott, J.K. 1987. Making extension effective: The role of extension/research linkages. In W.M. Rivera and S.G. Schram, eds., Agricultural extension worldwide:
Issues, practices and emerging priorities. London: Croom Helm.
McKee, C. 1984. Methodological challenges in analyzing the household in farming systems research: Intra-household resource allocation. In C. Flora, ed., Proceedings
of the Kansas State University's 1983 Farming Systems Research Symposium.
Shaner, W.W., P.F. Philipp, and W.R. Schmehl. 1982. Farming systems research and development: Guidelines for developing countries. Boulder: Westview Press.
Spring,A. 1988. Using male research and extension personnel to target women farmers.
In S.V. Poats, M. Schmink, and A. Spring, eds., Gender issues in farming systems
research and extension. Boulder: Westview Press.
Waugh, R.K., P.E. Hildebrand, and C.O. Andrew. 1989. Farming systems research and
extension. In J.L. Compton, ed., The transformation of international agricultural
research and development. Boulder: Lynne Rienner Publishers.
Vol. 2, No. 2, 1991

Adjusting and Transferring Agricultural Technologies: Three Examples from Peru'*
J.I. Matal
Transferring new or improved agricultural technologies to farmers is one of the greatest challenges to increasing agricultural productivity and improving farmers' physical and economic standards of living. To achieve this goal, some approaches emphasize technology generation by strengthening research programs. Other approaches seek to identify methodologies and strategies to more effectively transfer research-generated technologies to farmers. Often, both approaches address only one side of the problem.
Technology transfer in agriculture is an integrated process of generation and transfer that must begin and end with farmers. Research must originate from farmers' needs. The success or failure of technologies and practices must be evaluated based upon how farmers have adopted and used them and on the results they have obtained through implementing them.
Experiences of the Communication for Technology Transfer in Agriculture (CTTA) Project in the Callej6n de Huaylas, Ancash, Per&, clearly demonstrate the importance of adjusting technologies to meet farmer needs and provide examples of how communication approaches and channels can facilitate and enhance technology adjustment and adoption.
Deciding What to Transfer
C'PTA's first step in choosing what technologies to attempt to transfer to farmers in the Calleion de Huaylas was to identify with the farmers the main problems faced by their crops and the traditional methods with which they combatted these problems. The analysis helped to identify farmers' demands for technologies relevant to their perceived problems.
lAcademy for Educational Development.

The next step was to learn what technologies were available from sources of technology generation, in this case the local agricultural experiment station. Technology availability is relevant for solving immediate problems, such as insect attacks and plant diseases, and also for offering new practices (e.g., planting distances) or inputs (e.g., improved seeds) that may increase crop productivity.
CTLA compared farmer demand for technologies with the technologies that were available from the experiment station. Staff members paired the problems identified by farmers with solutions that were currently available from the experiment station. Next, they noted experiment-station technologies that did not directly correspond with farmer-identified problems, but that were related to them and might increase productivity. In addition to identifying the technologies that farmers were interested in using, CTTA also carefully studied farmers' knowledge and practices and recorded their expectations and attitudes toward adopting new patterns and behaviors and their feelings about technical assistance.
All af this information-agricultural technologies, farmer knowledge and practices, farmer attitudes toward adopting new practices, and farmer feelings about technical assistance-was used to select the first set of potentially successful technologies and to plan a strategy for transferring them.
C'TTA divided its intervention into two stages, both of which corresponded to one complete agricultural cycle of the potato, maize, and wheat crops that predominate in the pilot area.
In stage 1, experiment-station technologies were introduced to farmers and an effort was made to adjust the recommended practices to match farmers' ability to use and adopt them successfully. Stage 1 had two objectives: (1) to learn how best to translate new technologies into a series of behaviors that could be easily followed by farmers and (2) to identify the most effective tools (communication channels, media, materials, etc.) for transferring new practices to farmers. With this information, CTTA began informing farmers about the new technologies, motivating farmers to try them, and training farmers to ensure that the new practices would be applied correctly and with good results.
The best way to adjust technologies was to give them to farmers and then to watch how they used them. The project needed to identify problems with implementation, farmers wilIlingness to adapt the practices to meet immediate
Journal for Farming Systems R esearch -Extension

needs, and farmers' ability to solve problems using their own resources. This analysis could only be done by observing the users' traditional practices and how they managed the introduced technologies.
Also in stage 1, CTTA tested methods for transferring the technologiestraditional and new communication approaches, media, and channels-to learn which combination would most cffcctivcly reach and be used by farmers.
In the pilot area, farmers first work together in community fields and then individually in their own plots. Group cultivation of community fields offered CTTA staff an invaluable opportunity to work closely with farmers to learn traditional practices and to help introduce new technologies. Working with all the farmers at the same time in community plots provided a chance to learn their reactions and opinions about both old and new practices and to encourage them to adopt the new ones.
These first communication efforts delivered the adjusted information to farmers so that they might apply them in their own plots. Farmers opinions about the new technologies helped communicators know what to include in the first radio programs and flyers that were designed to introduce the technologies to a broader audience.
As each new technology or practice was introduced, its reception and adoption by farmers was evaluated in the field. Constraints to adoption and correct application were carefully discussed with farmers. Findings from this analysis guided stage 2, in which a final technology set and a communication strategy for its transfer were developed and implemented.
Based on experience gained in stage 1, CTTA constructed a communication strategy in stage 2 that could be implemented within local institutional capacities and resources. The strategy was designed to reach across long distances and difficult terrain and, thus, integrated the use of radio programs and graphic materials with group training and demonstration plots. This strategy allowed the information to reach many farmers through several channels, but at low personnel and input costs.
Graphic materials, which were widely distributed throughout the pilot area, broke down each technology into a sequence of simple steps. The illustrations were designed to be understood by farmers who could not read. The radio programs, in a combination of Quecbua (the local language) and Spanish, explained each illustration in the sequence included in the graphics.
Vol. 2, No. 2,1991

This multichannel approach helped farmers to study and learn about the technologies in their own homes before going to till their individual plots.
Group training helped refine extension workers' techniques for teaching new practices and provided a forum for discussing problems and learning about farmer concerns. These opportunities for direct communication built a system of continuous feedback that helped communicators, extension workers, and researchers to continually adjust technology recommendations to meet specific circumstances. Information collected during group training was immediately responded to and relayed in radio programs. Demonstration plots were managed using the technologies that were being promoted. They provided an opportunity to put into practice each new technology and to demonstrate the results of their implementation.
Following are three examples of how specific technologies were adjusted and recommended to farmers.
Controlling Papa-Kuru
Farmers said the potato worm (Premnotrypes sp.), called "papa-kuru" in Quechua and "gorgojo de losAndes" in Spanish, was one of the most damaging insect pests of potato in the Callej6n de Huaylas. Papa-kuru larvae bore through potato tubers, thus reducing quality and quantity of the harvest. The adult weevils appear after certain rains, causing many farmers to believe that papa-kuru comes down with the hail.
Farmers try to control papa-kuru with foliar insecticides such as Gusation, but they often apply it too late.
Experiment station recommendation. The potato specialist at the experiment station suggested using Curater powder, 5 g/10-m rows, applied at seeding, to control papa-kuru. The powder should not contact the potato seed. The specialist also suggested (for farmers with greater economic resources) applying Curater powder in the rows at seeding and granulated Volaton around the base of plants at hilling. The experiment station recommendation had an important implication for farm labor.
Farmers in the Callej6n de Huaylas break ground and sow the same day. Generally they use two crosses of an oxen-pulled plow to break up the soil and then plow the furrows. A man plows a furrow, followed by a woman who drops seed into it. Another woman follows behind and puts a handful of fertilizer
Journal for Farming Systems Research-Extension

near each seed. The plow closes one furrow as it opens the next.
Applying the experiment station recommendation requires making two more crosses with the plow and using one more person to apply the Curater. The application sequence is as follows: (1) open the furrow, (2) plant the seed and apply the fertilizer, (3) cover the seed with the plow, (4) apply the recommended dose of Curater, and (5) plow another pass to cover the Crater. The sequence was recommended to farmers and some tried it. They rejected it because it was expensive and because Curater was hard to apply in the correct fashion and dose.
Compromise recommendation. Discussing these constraints with farmers and experiment-station specialists produced an adjusted method that builds upon and reinforces traditional methods for controlling papa-kuru.
Method 1: Crop rotation helps prevent papa-kuru infestation. Farmers should rotate potato with crops of maize or other cereals. Papa-kuru dies during the rotation crop.
Method 2: Fields should be plowed at least three times to break up and turn the soil and then be left for several days before planting. This exposes the larvae to the sun and predators, such as birds, and reduces their numbers.
Method 3: After plowing, while waiting for the sun and predators to act, the potato seed should be moistened and place in the shade until it develops a greenish color (verdeo). This makes the seed bitter and less palatable to the papa-kuru.
Method 4: High hills should be built around the base of plants, which helps prevent papa-kuru larvae from reaching the tubers.
Method 5: Farmers who can buy insecticides should apply granulated Volaton around the base of the plants when hilling and Gusation to their foliage. Both insecticides are easily available in the market at costs affordable to most farmers.
Results. The first formative evaluation, in December 1987, after six months of intervention, showed that 70 percent of the sample farmers had heard the recommendations. Of these, 80 percent (55 percent of the sample) followed the recommendations, and 70 percent of them (49 percent of the sample) used them correctly.
Plant Spacing
This second example also illustrates the adjustment of a recommendation that was chosen to respond to a problem identified by farmers. The technology was developed through experiment station research where it successfully
Vol. 2, No. 2, 1991

increased maize production by more than 15 percent.
Most farmers in the Callej6n de Huaylas sow maize "a cola de bucy." A woman follows behind the plow and drops a continuous line of seeds into the furrow, at no controlled interval. The maize then germinates and grows at irregular spacing, preventing irrigation and rain water from circulating freely and encourages soil erosion on hillsides.
Experiment station recommendation. To achieve efficient plant spacing, experiment station specialists recommended that farmers plant three maize seeds each in a pocket spaced at 60-cm intervals along the sides of furrows spaced 80 cm apart. Such spacing would encourage development of strong root systems.
However, to achieve the plant spacing recommended by the experiment station, a farmer must walk behind the plow, make pockets in the side of the furrows with a spade at one-step intervals, and drop three seeds into each pocket. Applying this recommendation would require using a spade, which is not commonly used in the area. Planting one hectare would require digging about 21,000 pockets, each time lifting the spade to the shoulder. This represents considerable effort at 9,000 ft elevation. The recommendation thus presented two major difficulties: sowing would take much more time and effort and require a new tool and women, who would not have the strength required to use the spade, would be excluded from a task that they traditionally have performed because farmers of the culture believe that women's and earth's fertility are profoundly linked. The recommendation therefore required not only adopting a new tool and expending greater effort, but also a shift in cultural practices.
Compromise recommendation. The problems of time, effort, economics, and culture represented by the recommendation were discussed with the specialists and a compromise recommendation was made that followed the same principles as the initial recommendation-even spacing between furrows and pockets, with three seeds per pocket.
It was recommended that seeds be dropped into the furrows at regular spacing; 60 cm was translated to mean "three hands" and 80 cm to mean "four hands." Farmers were urged to cut light staffs "three hands" long for women to use as measures between seed drops and to place a "four hands" stick behind the plow to mark the appropriate distance between fulrrows.
One flyer and several radio programs promoted this recommendation during a one-month period when farmers were sowing maize.
Journal for Farming Systems Research -Extension

Results. Five months after promoting the new technology, formative evaluation showed that 56 percent of the sample farmers had sowed their crops at the recommended furrow spacing with the appropriate number of seeds per drop; 43 percent of the respondents had followed the recommendation for spacing within furrows.
Controlling Utush (Maize Ear Worm) As described above, some research-generated technologies require adjustments in farming practice; others require adjustment within the local cultural context. The technology for controlling maize ear worm (Elliothiszea), which seriously reduces maize harvest in the Callej6n de Huaylas, required adjustments in recommended practice.
Experiment station recommendation. Specialists from the experiment station recommended spraying Sevin on the maize silk when utush eggs appeared and then repeating the spray application at eight-day intervals until no more eggs were evident. Most farmers who tried to control the pest used the recommended insecticide. Many, however, did not use the correct dosage, application procedure, or timing. Usually, they applied the insecticide to the plant foliage, and only after damage occurred. Farmers who did not own or have access to a sprayer did not use any control.
Utush attacks were so serious in the project area that insecticide application was the only effective control measure. Under the experiment station recommendation, applying the insecticide required having access to and using a sprayer, being able to detect the presence of utush eggs, and applying the insecticide before the larvae hatched and began destroying the harvest.
Compromise recommendation. The specialists were asked to develop simpler ways to apply Sevin than by sprayer, so that farmers could use the insecticide. The specialists found that brushing the insecticide on the maize silk with a paint brush was an effective alternative to spraying and guaranteed that the insecticide was applied directly to the developing ears of maize. Specialists also revised their recommendation for mixing the insecticide to correspond with the new application procedure.
Farmers also needed to learn how to check their crops properly for presence of utush eggs. Specialists and extension workers showed farmers how to identify the eggs, how to distinguish them from similar-looking items, such as dew drops, and how to walk in pairs on each side ofa row of plants and check each ear for eggs. Farmers also were told that Sevin should be applied when at least one of every ten plants had egg deposits.
Vol. 2, No. 2, 1991

Results. Messages explaining how to check for and control utush were broadcast via radio programs and spots and distributed in a flyer for one month during utush season. No quantitative data were collected because the recommendations were disseminated before the formative evaluation. However, visits to the communities verified that utush control had significantly improved in the pilot area. Visits also confirmed that the prospect of harvesting utush-free maize encouraged farmers to plant larger areas of the crop.
Agricultural communication and technology-transfer processes often are seen as ways to disseminate information about research-generated technologies and practices. Practitioners of these processes often expect farmers to adapt themselves to new methods, rather than for the methods to be adjusted to farmers' needs.
CITA Project experience in Perfi shows that this approach itself requires adjustment. Farmers certainly need to adapt their traditional methods to include new technologies and practices, but researchers and technologytransfer specialists also must adjust their recommendations to build upon farmers' traditional practices and reflect farmers' physical, cultural, and economic realities.
Developing technologies that are well-suited to farmers' demands and circumstances requires a new focus in agricultural research and technologytransfer programs. Technology transfer should not be an activity that begins after a research program generates a new or improved technology or practice. The technology-transfer process must be part of the technology-generation process and must focus on the demands, expectations, and needs of the intended users. Technology generation must be directed by practical adjustment to produce pragmatic and efficient practices.
To achieve these aims requires that agricultural and technology-transfer specialists thoroughly know and understand the farming practices and cultural values of the potential users. This knowledge may be learned through information collected in developmental investigation. The importance of observing and, recording farmers' behavior and seeking to discern their possible constraints to adopting new practices cannot be overemphasized. The emphasis of agricultural technology generation and transfer should be on building upon and improving, not dismissing, traditional practices.
JournaI for Farming Systems Research-Extension

Technologies that are developed without being tested within the circumstances in which they are to be implemented must undergo postgeneration adjustment if they are to be useful to and meet the requirements of farmers. This phase should produce technologies that are applicable by farmers and serve their needs and also should reflect the knowledge and research of agricultural specialists.
A technology may require different kinds of adjustment. Some researchgenerated technologies require adjustments in technology-for example, a change in dosage or chemical. Other technologies may require changes in the tools required to use them-for example, using a brush instead of a sprayer. Still other technologies may require adjustment to reflect the local cultural context-for example, to allow women to fulfill their traditional role.
It is also necessary to consider the cost of a technology, which may be not only a monetary cost, but a labor and social cost as well. To respond to all of these issues, it is essential to observe carefully the behaviors and listen to the demands of farmers. In the Callej6n de Huaylas, CTTA found that economics was the greatest barrier to farmer adoption of new technologies. The project found that a technology presented as a sequence of behaviors to be followed throughout the cycle of a crop had little chance of being accepted and consistently applied. This did not mean that farmers refused to adopt technology packages. It simply meant that technologies had to be presented to farmers as individual practices with specific and immediate goals.
A communication strategy for transferring agricultural technology must focus not only on disseminating information and motivating adoption, but also on training farmers to use a technology correctly. This requires that communication strategies must be incorporated into an educational methodology with a very strong pedagogical component.
The examples describe above demonstrate that if new methods are not adjusted to meet farmer needs, every communication strategy is useless, costinefficient, and time-consuming. It is unwise to approach technology generation and technology transfer as separate entities. To be successfull, these must be complementary and intricately related.
Vol. 2, No. 2, 1991

Looking Beyond the Farm for Gender Issues in FSRE'
Patricia Ladipo2
The history from 1969 to 1989 ofa maize project, which involved ten Yoruba villages near Ile-Ife in western Nigeria, is examined in order to highlight gender issues in the maize system. The villages, which are involved in the University of Ife's Rural Development program were primarily cocoa growing villages, similar to those described by Galletti et al. (1956) and Guyer (1980).
Findings from several studies conducted during the agricultural development process are combined to demonstrate the following points: (1) program interventions at any point in a food production system may influence all other points, (2) program interventions can raise gender issues, (3) gender issues affecting postharvest processes can influence farm production, and (4) an ecosystems approach involving client participation in research and extension can help to predict and accommodate gender issues.
The 1969 base-line study3 of the villages indicated that men dominated farm production, not only of cocoa but also of the small amounts of food that were grown. Men also dominated the cocoa trade. However, women in these villages had traditionally processed and marketed food crops, including any of the local, low-yielding white maize that might be in excess of subsistence needs.
Figure 1 shows the flow of maize and money from maize prior to extension intervention. Most of the maize, whether stored or not, went toward family consumption after being processed into either pap or pottage (a mash ofmaize
1 Paper presented at the Tenth Annual Farming Systems Research-Extension Symposium, Michigan State University, East Lansing, October 14-17, 1990.
2 Ladcon, Iwo, Nigeria.
This study was conducted by the Department of Extension Education and Rural Sociology, University of Ife, under the direction of Dr. S.K.T. Williams.

40 LA DI P0
and beans). Some of it was sold and bought as grain, and from there some went into small-scale commercial production of pap or pottage. Most of the proceeds of these commercial uses of maize went to support the care and education of children.
Figure 1 also indicates if men or women were usually responsible for each process. This determination involved two considerations: (1) who actually did the work and (2) who was in control of the process. Because these factors changed over time, it is important to note that initially men both controlled and performed farming and storage, while women controlled and carried out processing and marketing. Both parents contributed maize, maize products, and maize profits to the family pool. The local white maize was a low-input crop and hardly any of the income from it was put back into the farm. Another thing to note is that shelling was done by women (who pounded or rubbed the ears), but it was performed mainly as an obligation to their husbands and was thus a process controlled by men. The extent and importance of men's control over shelling was initially underestimated by the development agents.
Pottage Making!
Manual ainyc
FaringShelling Development
I I Pap Making
* Marketing I
Maize Men do, men control
--Money Women do, men control
........................ Women do, women control
Figure 1. Local White Maize in the Community, 1969
Jo urnai for Farming Systems Research -Extension

In 1972, in order to help farmers increase their incomes and diversify their income sources, the project staff introduced a high-yielding yellow maize variety as a cash crop. Because this variety was not at first acceptable for food but was demanded by the animal-feed industry, it radically changed the way maize and money flowed in the community (Figure 2). First, with the introduction of the new variety, there was no locally grown maize available for women's maize-based activities. Second, shelling became a major problem because: (1) the yield and the acreage were much higher than those of the local white variety and (2) instead of being stored and released gradually, almost all of the crop was rushed to the market in order to get the highest possible prices. Because these prices dropped at the end of the rainy season, the crop was dried in a wood-fueled batch drier, and some women participated in carrying maize and wood to the drier, which was operated by men. Maize prices were dictated by the industrial consumers, because they were few in number and there was no demand for the yellow variety through the local food markets.
Pottage Making
' ~ ~ Family q
Farming Manual* $Maintenance& t
Shelling 0 Development I
I4 Pap Making
Development Local
I Drying
I -- -rKey.
....... ...... Maize Men do, men control
Industrial .4- Money ---Women do, men control
Marketing f ......Men do, industry controls
.................... No one does
Figure 2. Improved Yellow Maize in the Community, 1972
Vol. 2, No. 2, 1991

Yellow maize was a high-input crop and in order to acquire credit, seeds, information, and other inputs, the men formed farmers' cooperatives. The existence of these cooperatives and bulk purchasing outlets made yellow maize comparable to cocoa, and men took over the marketing of maize. Finally, it should be noted that because of the high input requirements, the debt incurred, and the excitement about the crop, most of the proceeds went into the cooperatives and the farms. Very little maize-based resource flowed into the family pot.
This situation came to the attention of project staff during a child nutrition program in 1976. Realizing that the project had benefited men and disadvantaged women, the latter group requested a program that would make their processing work easier and would give them their own source of capital. The project responded by helping the women to develop their own cooperative and to purchase a mechanical maize sheller (for a detailed project description, see Ladipo, 1981).
As shown in Figure 3, this project replenished women's resources and benefited the family, although money, not maize, circulated within the community. It is important to note that with the introduction of the mechanical sheller, control of the shelling process switched from men to the women who actually performed the shelling.
Storage Pottage Making
i I F Il
Farming Mechanicl Cooperative ,-. Maintenance &
Shelling Development I Development .
Me' -- -----Men's I II ----Cooperative I PMn
Development Pap Making
-Food --
Marketing A -I
I Key:
? Maize Men do, men control
Industrial I - Money -- Women do, women control
Market ............ Men do, industry controls
- No one does
Figure 3. Yellow Maize in the Community, 1976
Journal for Farming Systems Research-Extension

Although male community leaders approved of the establishment of the women's cooperative and welcomed the idea of the Sheller as a way of quickening the pace of marketing, the men reacted negatively to the machine when it began operating. Specifically, they argued about the price for shelling, withheld their maize from the machine, and demanded free shelling, and, finally, some even seized the machine.
There are many possible reasons for the men's negative reaction to the women's Sheller, but the main reasons become plain from an examination of Figures 2 and 3: (1) men had always controlled the shelling process in the past,
(2) all new technology had previously been targeted to men, and (3) whoever controlled the shelling controlled the timing of all subsequent processes and, thus, the price received for maize.
With the seizure of the sheller, women realized that the issue was one that could threaten the stability of their homes. They decided to sell the machine, not to their husbands but to another community. Having done so, however, they did very little manual shelling and the flow of maize almost ceased. Everyone realized that something had to be done.
Fortunately, at that time high-yielding white varieties of maize were becoming available, and both men and women expressed interest in them. After all the problems with fodder maize, staff and villagers were eager to adopt a highly acceptable and profitable variety.
Project participants took part in identifying the salient characteristics affecting acceptability. These characteristics were quantified for the various types available and the economic implications of the acceptability factors were explored with maize handlers. For each available variety, the numerical values assigned to each characteristic were combined into an acceptability score and this score was then multiplied by the yield potential as reported from research stations. This approach adjusted yield figures, which were of major concern to men, by acceptability factors, which were of major concern to women. The variety with the highest adjusted yield value was adopted by the farmers in 1979 (for details, see Ladipo and Nout, 1981).
As shown in Figure 4, the introduction of the improved white variety reconnected maize and money with all parts of the system in the community, except for industrial aspects. Women returned to their normal pursuits: making and selling maize foods and shelling maize by hand, although the shelling was occasional and the amounts were moderated by storage.
Meanwhile, the women worked to develop their own cooperative, which was built on the proceeds raised through trade in maize and other items.
Vol. 2, No. 2, 1991

StoragePottage Making
Women's Faml
Farming Manua Cooperative ~- Maintenance a
Shelling Deelopment J Development
Men'sPap Making
q Drying L I------- --- -- -- -- --q Industrial Ky
Marketing *Maize -Men do, men control
- ---- ----j- -- Money Women domen control
- - - - -j...... Women do, women control
.- - No one does
Figure 4. Improved White Maize in the Community, 1979
Eventually, the women's cooperative grew and subdivided into large, viable groups, while the men's groups became increasingly burdened by debt. Even so, in the area where the maize sheller had been introduced, there was no direct connection between any women's group per se and the production and handling of the crop.
The direct connection between women's groups and crop production and handling in the 1990s, which is occurring now, will probably have a major impact on control of the whole system, right back to the procurement of farm inputs. Before examining this latest rearrangement, it is useful to review the ways in which client participation and asystems approach could have been, and at times were, used to guide programming in these villages.
Journal for Farming Systems Research -Extension

Tracing the flow of inputs and outputs from one process to another can provide information not only about what happens and who is involved, but also about people's expectations. For example, had flow diagrams been used from the beginning of the project, the effects of introducing yellow maize and the central importance of shelling might have been anticipated by tracing and two of the failures might have been avoided.
Within a longer time frame, tracing inputs and outputs was usefully applied to women's enterprises. It was found that women usually engage in a series of occupations, starting with whatever amount of capital they have and progressing through a series of increasingly capital-intensive and decreasingly laborintensive jobs (Figure 5). With the women's help, it was possible to understand their views, where farming and trade featured in them, and, particularly, the
0 Farming
. Distant produce buying
0Local produce buying
9 Urban petty trade
o 0 Rural petty trade
g Selling cooked food
* Making cold pap
*Gathering wood or leaves
*Paid labor
O Very Hard 0, Easy
Level of Physical Exertion
Figure 5. Women's Occupational Ladder
Vol. 2, No. 2, 1991

importance of the cooperative in saving them from sliding down the occupational ladder.
Food flow was used to identify male farmers, female marketers, roadside cooks, and housewives as people to consult about maize acceptability. Not only were these the people to be affected by any changes, they were also the local experts on the subject.
Focusing on particular processes
The systems approach also facilitates a process-oriented approach. For example, after tracing maize to the relevant participants and processes, the participants in each process supplied detailed information about inputs, outputs, and transformations. Once the course of events was determined, some aspects of the process were removed from their natural settings and examined and quantified in the laboratory. The quantitative data were then used to predict how new inputs would affect processes and outputs. This type of analysis can only be performed through client cooperation.
Participation is also important for identifying the problematic aspects of a system. For example, in the case of the maize sheller, project staff focused on the processing bottleneck and the physical labor output ofwomen. With more open communication from participants, it might also have been possible to focus on the family maintenance and development process, to see that shelling was an important aspect of the exchange between husband and wife, one that could not be altered without prior ideological change and a different economic balance between the sexes (Figure 6).
Considering the environment
Finally, an ecosystem view encourages consideration of the local environment. Figure 4, above, presented a situation very similar to the initial one. A high-yielding variety had replaced the local maize type, but, except for the addition of cooperative development, there was very little change in activities or resource control.
Rcent changes in the political and economic environments, stimulated by structural adjustment, are bringing radical changes to maize production in the study villages, and seem about to upset former gender-based patterns of resource control in the maize system.
Journal for Farming Systems Research-Extension

Heavy Labor
Food StaplesWOE
Housing Clothing
Education CHLRNEducation
MEN aborCooked Meals
House Keeping
Figure 6. Reciprocity of Contributions Among Family Members
Devaluation has led to a rapid rise in agricultural input costs with a decrease in maize production and on-farm maize storage. In fact, increasingly higher proportions of maize are sold green to bridge the widening seasonal food gap and to help farmers repay debts incurred on maize production. Policies encouraging export production cause farmers spend more time on cocoa and less on food crops. Meanwhile, import restrictions and other policies aimed at increasing home industry have multiplied the industrial uses of maize, making the maize trade very competitive. These same restrictions have reduced the availability of many imported manufactured items, pushing more traders into dealing with agricultural products.
Other important changes include the creation or reconstruction of rural feeder roads; bank reforms encouraging small-scale, rural-oriented loans, especially loans to women's groups; and a program (Better Life for Rural Women) that has promoted the idea that women are fit and proper people to handle new technology.
All of these economic and social stimuli are occurring at the same time as public support for health care and education is being cut back, making most families unable to ensure their children's futures. In this situation of extreme pressure and unusual opportunities, change--even change in gender rolescan be expected.
In 1989, one of the women's cooperatives prepared an application for a rather large loan for the purpose of creating more maize storage .4 Part of the maize they plan to buy and store will be their husbands'. Vol. 2, No. 2, 1991

For the first time, women's cooperative money will be feeding into the maize farms as men accept forward-buying arrangements in order to pay for inputs that they could not finance otherwise. The loan project calls for construction of the Food and Agriculture Organization/Danish International Development Agency/International Institute of Tropical Agriculture (FAO/DANIDA/IITA) maize cribs, which will be owned and run by women. These women will control the flow of maize out of the crib, and thus the rate of shelling. The women will shell manually for the first year while they consider whether or not to invest in a sheller.
As always, women will control the food marketing of maize, but because all of the varieties grown are now acceptable as food, women can also negotiate the prices for industrial sales. Finally, it is envisaged that in terms of maize and maize income, women will be making the major contributions to family welfare.
Figure 7 illustrates that within the structural adjustment environment, the loan is likely to change control of the maize system so that women will have influence in production decisions and will have dominant control in all other areas. The effects of these changes should be closely monitored through frequent communication with all family members. In this way, program support could be made available if needed to assist families in making constructive adaptations.
The author is grateful to the Association for Farming Systems ResearchExtension for inviting and supporting her participation; the Isoya Rural Development Project staff and other colleagues at the University of Ife/ Obafemi Awolowo University for their contributions to the action and research projects described in this paper; and the rural and urban people of Ife for their willingness to experiment and share.
4 Before the application was processed, the government became involved in maize storage and prices dropped so low that production has since been suppressed. The women postponed action on the loan until the government abandons the project and production and prices return to normal.
Journal for Farming Systems Researeh-Extension

I- I
I Storage Pottage Making
I I L -r -J I I .. _.
+ Ii
Farming Sellhinig Cooperative Maintenance & .
S Shelling Development Development
-Men's -I 'Cooperative i -~ _Pap Making
Development Local L. ----..
. .Marketing I
S Drying L --------- J
SIndustrial I Key:
Marketing -- Maize Men do, men control
- Money -- Women do, women control
........... Women do, industry controls
...... Men do, women control
- - No one does
Figure 7. Women in the Maize System, 1990s
Galletti, R., K.D.S. Baldwin, and I.O. Dina. 1956. Nigerian cocoa farmers. Oxford. Guyer, JI. 1980. Food, cocoa, and the division of labor by sex in two west African
societies. Comparative Studies in Society and History 22(3):355-373.
Ladipo, Patricia. 1981. Developing women's cooperatives: An experiment in rural
Nigeria. Journal ofDevelopment Studies 17(3):123-136.
Ladipo, Patricia, and R. Nout. 1981. Maize acceptability: A problem in extension
programming. Nigerian Journal of Agricultural Extension 1(1): 15-23.
Vol. 2, No. 2, 1991

Integrating Women into Farming Systems Research:
A Homestead Vegetable Production Project in Coastal West Bengal
S. Chakraborty, J.E. Gleason, B. Mandal, and C.S. Das
This paper is based on an action research project integrating women into farming systems research (FSR). The research focus is on women's role in vegetable cultivation in homestead gardens in coastal West Bengal, where vegetable production occupies a major place in the agricultural production system. The study demonstrates that integration of women into FSR can be achieved by addressing their traditional roles and needs.
In coastal West Bengal, as in other parts of the country, women play a major role in rice production as well as in vegetable production, the other major crop in the area. But women by and large have been bypassed in agricultural development and extension systems and generally have not received the benefits of improved technologies. Although not ignored completely, women's activities are treated as "traditional," and their skills viewed either as innate to women or as easily acquired through practice without any technological or research support. Social attitudes are also a barrier to women's access to technology. Women are perceived to be confined to their homes, to have low literacy levels, and, therefore, to be unable to absorb scientific information and skills.
Realizing that successful agricultural development for resource-poor farm families depends on the full integration of all human resources, the FSR team at the Ramakrishna Mission, Lokasiksha Parishad, Narendrapur, conducted research on the traditionally recognized role of women in agriculture, i.e., homestead vegetable production. The program was conceptualized as a mechanism for integrating women into the mainstream of FSR.
The homestead garden program ofRamakrishna Mission is an integral part of the overall research project in the region because it develops the knowledge

and skills of women in vegetable production, a primary economic activity of many villages in coastal West Bengal. Specific objectives of the kitchen garden program are to develop an understanding of the value of women's contributions to the farm economy and to enhance their contribution by upgrading their skills and putting available resources to maximum use. The program also aims to help women understand the dietary importance of vegetables and the nutritional needs of family members and to ensure a steady supply of vegetables for consumption and sale throughout the year. The program also promotes contact between women vegetable producers and FSR scientists.
The FSR scheme of the Ramakrishna Mission is being implemented in nine villages in three coastal districts of West Bengal. Nearly 80 percent of the population lives in rural areas and is mainly dependent on agriculture. The average land holding is very small and many people are landless, working as agricultural laborers on other peoples' farms. A household survey of 250 families selected by stratified random sampling showed that the highest percentage of families has land holdings from 0.11 to '1.00 acre (43.2%), followed by families with land holdings from 1.01 to 2.50 acres (22.4%), and families with land holdings of 2.51 to 5.0 acres (11.6%). Landless laborers comprise 22.8 percent of the total population. Owing to the different environment in this part of West Bengal, these resource-poor farm families have not benefitted from the Green Revolution that occurred in other parts of India.
The main crops grown in the area are rice and vegetables; multiple cropping is constrained by lack of irrigation. Rice, followed by vegetables, is the main cash crop for most farm families. A study of 20 families selected at random from one village (Jelerhat) showed that most of the families obtained a cash income from vegetable cultivation and wage labor. Only 25 percent sold rice. The production of vegetables thus occupies a very prominent place in the economy of this village and women, as equal partners with men, play a significant role in vegetable production.
Genesis of the homestead vegetable production program
In coastal West Bengal, women's contribution to vegetable production normally is recognized as an extended function of their domestic responsibilities. The word "garden" has a different connotation than "field production"
Journal for Farming Systems Research-,Extension

because it implies production of vegetables and flowers on homestead land. Vegetable or flower gardening around the house is treated as women's work in which men extend a helping hand. In field production the situation is reversed, and men are regarded as producers and women as helpers. As a result, the agricultural extension system does not address women as commercial producers of vegetables, although they have an important role to play.
FSRscientists working in the field have discovered two major barriers to the full integration of women into sustainable development programs: (1) men believe that women are too shy to receive any information from scientists, especially males; and (2) women feel uncomfortable meeting jointly with men in FSR projects. Scientists recognized that these cultural barriers to women's participation could only be overcome by targeting production systems in which women were involved extensively and in which this involvement was deemed culturally appropriate. Because women's involvement in home garden vegetable production was accepted as a traditional female role, scientists felt that research focusing on vegetable production would help break down the barriers and integrate women into mainstream research.
Meetings were organized to discuss home garden vegetable production with village women. Discussions in the meetings focused on types of vegetables produced, on sources of supply of seeds, on manures and fertilizers used, on sources of irrigation, and on consumption and marketing of produce. Based on the information gathered in these meetings, a scheme of vegetable production in home gardens was designed and was first implemented in the rainy season of 1987. Two hundred women in nine villages participated in the program. They were given vegetable seed packets, worth about Rs. 5,1 containing seeds of bottlegourd, okra, snakegourd, bittergourd, french beans, and amaranthus. The seeds were thought to be sufficient for growing up to 200 kg of vegetables on 3 to 5 cents (100 cents= I acre) of land. Women were instructed in the sowing application of manures/fertilizers and in the utility of systematizing vegetable production for better nutrition and sale. Followup visits were made and progress was reviewed in periodic meetings.
Although success in the first season of the program varied considerably, it generated enough interest that the number of women participating increased
1 US$0.30; US$1 = Rs. 16.
Vol. 2, No. 2, 1991

to 220 in the next season. Feedback from the women guided the scientists in improving the program. Seeds that did not do well were eliminated and a better mix of vegetables was developed. Distribution of seeds (carrots, peas, radishes, spinach) and of seedlings of tomato and knolkhol was preceded by training on five specific topics: (1) use of organic manures, (2) preparation of land, (3) soil management, (4) raising of seedlings, and (5) nutritional aspects of kitchen gardening. Individual records of vegetable production were kept. As a result of the improved training and monitoring, vegetable production increased significantly that winter. Table 1 shows the vegetable production record of women in two villages.
In Tulsiberia, average production was greater and the production range was smaller than in Jelerhat. In addition to increased consumption, many of the women in the village occasionally sold vegetables. Price data for these vegetables were not collected, so the value of production was not known. However, it can be concluded with confidence that none of the participating families could have purchased the same quantity of vegetables that they grew. A comparative study was made of the earnings from vegetable production Table 1. Production of Vegetables:
Jelerhat and Tulsiberia, Winter 1988
Vegetable Average Production (kg) Minimum (kg) Maximum (kg) JELEHAT (40 Women)
Radish 9.9 6.0 15.0
Spinach 12.2 8.0 16.0
Tomato 38.5 8.0 75.0
Carrot 2.0 0.5 4.5
Pea 0.7 0.0 2.5
Knolkhol 80.3 0.0 120.0
TOTAL 187.9 25.0 213.0
TULSIBERIA (25 women)
Radish 10.0 8.0 15.0
Spinach 11.5 8.0 16.0
Tomato 54.7 47.0 65.0
Carrot 4.8 3.0 8.0
Pea 1.5 1.0 2.5
Knolkhol 150.5 140.0 162.0
TOTAL 232.8 216.1 255.2
Source: Gleason, J.E., and S. Chakraborty, 1988.
Journal for Farming Systems Research-Extension

among the participants and among nonparticipants in the kitchen gardening program in the village of Jelerhat. The data in Table 2 show that the families participating in the kitchen gardening program during May and September 1988 had greater average earnings from vegetable production than did nonparticipants. The difference in earnings was Rs. 149.34 per 10 acres, or Rs. 192 per family. Home garden production complemented the family's cash income from vegetable production.
Regular meetings and field visits with the participating women gave adequate feedback to the scientists about the need for further systematizing the program. More attention was paid to topics such as selection of vegetable seeds, timely delivery of information, preparation of land, and marketing information. In addition to production records, records were kept of individual performance of various vegetables in different plots.
During village meetings it was found that decisions on vegetable production were made only on the basis of market demand. There was no systematic planning of vegetable production in home gardens from a nutritional point of view. Seeds were sown according to the individual grower's interest. It was found that in spite of easy production of leafy and other green vegetables in gardens, families consumed more potatoes than any other vegetable. Nutritional information on the production and consumption of vegetables, therefore, was included in the next package of technological messages. FSR scientists developed a plan for rotation of vegetables that would ensure both maximum utilization of available land and enhanced production. The criteria for the selection of land in terms of location, soil type, etc., were explained. Participants were given a design for land shaping and rotation of vegetables.
Table 2. Comparison of Earnings From Vegetable Production of Kitchen Garden
Families and Nonkitchen Garden Families, May-September 1988
Kitchen garden families Nonkitchen garden families (Rs./O.10 acre) (Rs./0.10 acre)
May 351.60 219.65
June 116.11 111.27
July 75.00 65.75
August 74.00 76.00
September 86.75 81.95
TOTAL 704.26 554.92
Note: Based on exchange rate: US$1 = Rs. 16 IC.
Source: Gleason, J.E., and S. Chakraborty, 1988.
Vol. 2, No. 2, 1991

The design included digging a compost pit in one corner of the land. Principles of making the pit were explained to the participants. They also were advised to store the ashes normally thrown away after cleaning ovens in order to add potash to the land. Field staff have been briefed for necessary field follow up.
In the current season the program will focus more on means of attaining stability in vegetable production and on improved family nutrition.
In cases of small and marginal land holdings, an important step toward stabilizing the production system is to maximize the use of resources available within the command of the farm family. Thus, the question of the development and use of women's resources is very significant for the stability of the agricultural production system.
As a preliminary step to integrating women in farming systems research and extension, the Ramakrishna Mission FSR scientists identified women's involvement in home garden vegetable production as an entry point. This was an important target because vegetable production is a major source of cash income in coastal West Bengal and thus occupies a very significant place in the primary sector of the economy. Home garden vegetables are consumed primrily by family members and are a contribution to the improvement of their nutritional status, which represents an economic gain in the long term. FSR scientists, therefore, give vegetable production the same priority as crop production.
The kitchen garden program was identified after a series of meetings with village women. They showed interest in the program and demanded good seeds, which were supplied by the scientists. The program was closely monitored through regular meetings and field visits. Selection and proper treatment of seeds, raising of seedlings, preparation and use of farmyard manure to improve soil fertility, and nutritional aspects of home garden vegetable production were included in the training package. Women also were advised on the methods of selection and preservation of seeds. Scientists visiting the field identified the best fruits in the gardens from which seeds could be collected. Women showed interest in production of their own seeds.
The program has shown some definite results. Village women are becoming increasingly interested in producing and consuming more and new types of vegetables. In addition, there is greater rapport between the scientists and
Journal for Farming Systems Research -Extension

rural women gardeners and freer interaction between the two. Women turn up in large numbers at village meetings to discuss vegetable productionwith the scientists and to collect seed packets. They report on increased consumption of traditional and new vegetables and on their modes of preparation. For example, carrots, which were not grown locally before the kitchen garden program, now arc commonly consumed. The economic benefits of the program also are important, as participants can save the money formerly used to purchase vegetables.
The project of homestead vegetable production was designed to initiate the process of integrating women into mainstream agricultural research and development. The project has demonstrated that the involvement of women in specific agricultural programs can have strong positive benefits. Owing to regular interaction with the scientists through this specific program, women now are much more articulate and can express their needs and interest freely. This has accelerated progress toward an integrated and more productive use of farm household resources. Scientists have a better understanding of the role of women in agriculture and of their needs as producers.
A number of significant findings have emerged from this ongoing project. Despite perceived cultural assumptions, women are not necessarily too shy to interact with outsiders (even males) about their specific ideas and needs. They are keen to seek relevant information from scientists to improve their garden production.
Training needs to be an intrinsic component of the home garden program. One message cannot be applicable for all times and in all locations. Training is necessary to meet the problems and challenges posed by ecological differences, variation in soil conditions, and natural calamities.
To ensure the stability of the agricultural production system, a supplementary economic support system is needed that will use women's labor to increase family income. Women have expressed an interest in starting some homebased enterprises, such as poultry raising or other marketable activities. Possibilities are being explored to start up household enterprises in the project area.
As has been observed, women play a vital role not only in homestead vegetable production but also in commercial production. By participating in this project, women realized the value of their work in vegetable production
Vol. 2, No. 2, 1991

to the household economy and their need for better scientific information. Through their participation they have been able to improve their small-scale vegetable production both for home consumption and for sale. The home garden program ensures the use of the total resource endowment of the family in terms of the labor of women and children, available land, organic manure, and other resources. Home garden vegetable production can counter the challenges of food scarcity and nutritional imbalance.
Gleason, J.E., and S. Chakraborty. 1988. Using traditional roles to integrate women into
agricultural projects. Farming Systems Research Newsletter 3(1).
Journal for Farming Systems Research-Extension

Farm Management Information Systems: A Role for Literacy Training in FSRE Programs'
K.L. Mintz2
The role of farming systems research-extension (FSRE) in relation to farmers and sustainable agriculture is a relevant topic for proponents of farmer-based research. This paper focuses on the improvement of on-farm information management with a methodology based on the incorporation of farmer training into FSRE programs. Expansion of the farm family's role in information management improves community resource development and strengthens FSRE.3
. FSRE programs incorporate the commodities of information and knowledge. The programs should facilitate the participation of farmers, farm families, and farm communities within what is broadly referred to as the agricultural knowledge information system (AKIS). AKIS is defined as a set of agricultural organizations and/or persons and the links and interactions between them (R61ing, 1990). Participants engage in processes such as the generation, transformation, transmission, storage, retrieval, integration, diffusion, and utilization of knowledge and information with the purpose of working synergistically to support decision-making, problem-solving, and innovation in a given country's agriculture or a domain thereof.
I Paper presented at the Tenth Annual Association for Farming Systems ResearchExtension, Michigan State University, East Lansing, October 14-17, 1990. International Extension Training Program, Michigan State University Cooperative Extension Service.
3 The point of view evident in this paper is a hybrid one based on experinece with Farming Systems Research Programs and with Non-Governmental Organization (NGO) Comminity Development Programs. The reality of the geographic setting, West Africa, is that the majority of resident farm family members are illiterate.

Within the AKIS, FSRE is a mechanism that links research-station scientists to farming communities. In developing countries these links are often weak because there are few structural mechanisms to make researchers accountable to farmer interests. In these situations, FSRE is a participatory methodology that is able to focus research on relevant problems and that provides timely feedback relating to on-going research.
Farming systems research (FSR) survey work focuses on collecting information from farm families. This information is then transported and analyzed within the structure of the FSR program. The analysis step produces the "knowledge," which is circulated to agricultural research institutions and used to improve current farming practices. The resulting innovations are then tested with farmers, often under the auspices of the farming systems program. It is at this point or later that extension involvement may become significant; widespread diffusion of the technology concerned usually is the responsibility of extension.
In FSRE programs, the farmer is both the source and the end target of information. In developing countries, where small-scale farmers are predominantly illiterate, farmers have a minimal input in the recording of FSR information. Instead, they answer questions and wait to see what studies will involve them in the future. Much of the information collected in FSR surveys concerns observations of farmer behavior.
Benefits of FSRE Programs for Farmers
FSRE programs enjoy some popularity in farming communities. One of the most immediate and obvious benefits of FSRE programs is that a convenient transportation link is provided from the participating village to a major city. No matter how disciplined FSRE supervisors may be, older people, medical evacuees, and important messages often will find transport in the projectliaison vehicle. Village-based enumerators can usually be pressured into using motorcycles or mobilettes when they are available (and operational) to travel to regional capitals in the event of a local crisis.
Participation in FSRE programs has some economic benefit for the concerned villages as well. A salaried position is created in the village in the form of a project enumerator. Several times a year a project tour will lead VIPs to the site. Festive celebrations usually are supported by project funds. Local government officials may become more responsive to their constituents, who
Journal for Farming Systems Research-Extension

now have regular contact with these often internationally financed project administrators.
Role of FSRE for Farmers and Communities Clearly, these are all incentives for farm families to participate in FSRE programs. But how are household agricultural practices affected by FSRE on a sustained basis? Generally, there are program-sponsored on-farm trials that are of significant interest to the family. It is not obvious, however, that the farmer learns to manage experimental activities. The trials provoke much village discussion, but within ESRE the farmer usually is accorded neither the role nor title of researcher. The measurements go to the program office and results may be presented back to a village group.
What role does FSRE play in local resource management? In addition to participation in meetings organized by the FSRE program, community groups should be encouraged to plan and promote independent activities that use aggregated farm data as a take-off point for exploring ways to improve production methods. Instead of promoting this pro-active type of approach, FSRE conditions the farmer to wait for presentation of new technologies. The formal research sector is important to the farmers as a liaison to new technologies, but groups concurrently could research adaptations of known techniques. Overreliance by farmers on standardized recommendations from external sources can lead to adoption of technologies that threaten farmers' survival. Management capacity to process information, economic decisionmaking, and policy analysis are fundamental to development of on-farm human resources (Mding, 1986).
Farmer-first Methodologies
Within the realm of farmer- based or fhrmer-first research, several participatory methodologies of information generation and recording are being used and refined. These include the formal use of meetings and the production of maps and diagrams (Conway, 1989; Lightfoot et al., 1989). Researchers interested in identifying and tapping indigenous knowledge systems find these techniques useful. But are these techniques tools for sustained improvement in farm management? Do they help farm families integrate information into production decisions? Close examination of sample diagrams shows that most would be of limited use to illiterate farmers.
Vol. 2, No. 2, 1991

It is assumed that FSRE programs are designed to make the AKIS increasingly responsive to farmer needs. Yet it is not clear that FSRE programs have expanded the farmers' role within the AKIS. Training people to transform knowledge within their part of the AKIS is fundamental to creating a more effective AKIS (Rbling, 1990). What training have farm families received? The FSR literature is replete with training manuals for all parties involved in AKIS, except for farm families. FSR generally leaves training and education up to extension. In developing countries, however, the role of extension is often limited to delivering training in "new and improved" technology packages.
Transformation, the movement and translation of knowledge generated in one part of the AKIS and used in another part (R61ing, 1990), is a process in which farmers currently have a very limited role. In contrast, FSRE programs have been structured to play a dominant role not only in transformation but also in the transmission, storage, and retrieval processes. In filling such a dominant role, is FSRE serving farmers or reducing their participation within the AKIS?
Increasing Farmer Participation in FSRE
In order for farm families to have an active role in FSRE, their role in the analysis of information needs to be expanded. This point of view is based on the recognition that a farmer is the manager of a complex production system. How can the role of manager be supported? How can FSRE be more responsive to farmers' immediate needs? By integrating appropriate farmer training into FSRE programs, farm families may more directly benefit from them. Participating farm communities may then become more active within the AKIS.
Record keeping: A key method for improving farm management is the activity of record keeping. Record keeping is a first step in the management of information and the control of access to it. The willingness to keep records is based on recognition of their ultimate value to the farm manager. This theory is widely accepted within the U.S. agricultural system. As early as the 1930s, extension services in several states joined together with the Farm Credit Association to offer training to farmers in record keeping (Rasmussen, 1989). The Cooperative Extension Service Telfarm system is a program that facilitates and promotes farm information management. Farmers input data into an automated system and then are provided with reports that they use for
Journal for Farming Systems Research-Extension

farm management. In this symbiotic relationship the Extension Service gains access to a large aggregate pool of data that can be used to serve more general needs of the state and national AXIS.
Literacy programs: The ability to use records as a management tool generally is dependent on some level of functional literacy. Where FSRE programs work with illiterate farmers, functional literacy training programs to serve the community will be necessary. Ideally, farmer training needs should be assessed at the outset of the FSRE program. This might be done while the community is gathered to learn about the objectives and operations of FSRE and to explore the possibilities for farmer participation. In this situation farm families would be offered a tangible benefit, a local education program, in return for the time they would be asked to devote to FSRE activities.
The literacy curriculum might be designed to dovetail with the varying information needs of farming systems research. At the beginning of an FSR program, group activities could focus on village histories, demographic surveys, market studies, and other baseline data. As skills become more developed, the techniques for filling in field production forms and marketing questionnaires might be worked into lesson plans.
Literacy programs could certainly be integrated midstream into FSRE projects. During classes participating family members would learn exactly what the enumerators had been noting on the forms and notebooks that they tote around the village. In FSRE, gathering additional data about the village and zone is a continual process (Shaner et al., 1982). On-going classes could serve as a resource for this sort of sustained research.
One criticism ofFSRE is the ratio of costs to benefits. It has been suggested that benefits need to be increased and distributed more widely (Hart, 1990). What are the benefits of offering FSRE-linked literacy training to the farm family? Someone in the family, although not necessarily the head of the household, would have access to community-based education. This education would be designed around themes of local interest and skill development relevant to improving agricultural production. Although differences in farm management practices could conceivably be noticeable in the short term, skills development would continue to be developed as technology and natural resources change. In this sense, improving farm management is a long-term investment in promoting the sustainability of an agrarian way of life.
On a community level, the resource information collected would be gathered, transformed, and stored locally, and would be retrievable by the community. Control over access to community knowledge resources would
Vol. 2, No. 2, 1991

empower the local population. No one research group or program would be able to control access to the collective knowledge of the community, nor would outsiders be able to control distribution of this resource. The ability to store and transfer information confers the power to determine what information is available to whom and the power to influence other groups (Garforthe, 1986).
What are the costs associated with creating a companion literacy program for FSRE programs? Who would bear them? Obviously a literacy trainer, materials, and supervision of the program would involve cost. Within FSRB, training generally is the domain of extension, but literacy training does not generally fall within the accepted realm of extension education. The major basic skills of reading, writing, and arithmetic, however, make significant contributions to agricultural growth by facilitating the transmission of knowledge and the improvement of production skills (Wharton, 1963). But extension programs in developing countries generally do not have the skills and financial resources to support literacy programs.
NGOs and literacy programs: Nongovernmental organizations (NGOs) tend to be weak in agricultural expertise and "inexpert" at making links with formal agricultural research (Chambers, 1989). Literacy training, however, is one of their particularly strong suits. A common problem with village-based literacy classes is a lack of reading materials; one of the major problems with literacy education overall, in fact, is the loss of basic skills owing to nonuse (Wharton, 1963). By integrating literacy training with information gathering and the recording of knowledge, a valuable village document collection would be created. It is sometimes difficult for NGOs to convince donors that literacy programs lead to improved diets for "hungry people" in developing countries. Collaborative FSRE-linked literacy programs geared toward functional uses, such as improving household agricultural production and community organization, would certainly receive NGO funding.
Experience shows that functional literacy and numeracy campaigns were well received by the local populations once a defined need for them existed. In the World Bank Mali Sud project, these education programs not only benefited the farmer organizations involved but also facilitated the adaptive field-crop trials designed to increase cotton production. It also should be noted that by developing the literacy skills of farmers, it becomes possible to reduce the FSRE personnel needed to supervise activities. In the second phase of the Mali Sud project the ratio of farmer groups served by extension staff was more than doubled (Russell, 1986).
Journal for Farming Systems Research -Extension

What are the other costs of literacy training and increasing the level of farmer participation in FSRE programs? Researchers are likely to lose some of the autocratic control they now exercise over the design of survey instruments and the dissemination of research results. As farm families become more involved in production and analysis of farm records, they are likely to become more opinionated about what information is worth collecting and recording. Researchers will have to present logical proposals for each study, including justification for data requirements.
From the farming systems researcher's point of view, there is another problem that has not been addressed. Study sites often are selected based on representativeness in terms of villages in the same agroclimatic zone. What would a participatory, action-research type of plan such as the one outlined here do to the researchers' credibility in using the data to construct regional models? That question would be open for debate within FSR circles. From the farmer perspective, it is probably a nonissue.
Benefits of Literacy Programs
FSRE programs stand to benefit significantly from working with more literate farm families. Responsibility for the bulk of record keeping could be transferred from village-based enumerators to farmers. This would allow more time for enumerators to conduct special studies or, as mentioned above, to work with a greater number of farmers.
The quality of data arguably would be improved for several reasons. The farmer would be able to record information on a more frequent basis than the enumerator. With the increased possibility for protecting confidentiality, the farmer may provide more accurate information. On a more fundamental note, if the farmer has a personal use for the information and understands its value, he or she will be more interested in accuracy.
FSRE programs will also gain long-term benefits from facilitating farmertraining activities. Wharton (1963) asserts that extension education programs in developing countries have focused primarily on teaching knowledge about new inputs and new techniques in production. Knowledge about the economic aspects of farming, such as the analysis of production and marketing functions, has not been given enough attention. Clearly FSR staffs could be instrumental in providing the expertise for development of these sorts of curricula in conjunction with extension and NGOs.
As farm families and participating communities begin to reap the benefits of improved farm management and resource development, they will become
Vol 2, No. 2, 1991

more interested in supporting local FSRE programs and will develop into real constituent groups. Admittedly, this means they will have a stronger voice and a larger role in FSRE, but it also means that they will be grass-roots advocates for the programs. In thinking about sustainability, FSRE proponents would do well to think about the need to develop such constituencies.
Chambers, R., A. Pacey, and L.A. Thrupp, eds. 1989. Farmer first: Farmer innovation
and agricultural research. New York: The Bootstrap Press.
Conway, G.R. 1989. Diagrams for farmers. In R. Chambers, A. Pacey, and L.A. Thrupp,
eds., Farmer first: Farmer innovation and agricultural research. New York: The
Bootstrap Press.
Garforth, C. 1986. Mass media and communications technology. In G.E. Jones, ed.,
Investing in rural extension: Strategies and goals. Essex: Elsevier Applied Science
Publishers, Ltd.
Hart, R. 1990. Implementing the R & D process. In The proceedings of the Sustainable
International Development Workshop, Sept. 10-15, 1989 (DRAFT). East Lansing:
Michigan State University.
Kaimowitz, D., ed. 1990. Making the link: Agricultural research and technology transfer
in developing countries. Boulder: Westview Press.
Lightfoot, C., O. De Guia, A. Alman, and F. Ocado. 1989. Systems diagrams to help
farmers decide in on-farm research. In R. Chambers, A. Pacey, and L.A. Thrupp, eds., Farmer first: Farmer innovation and agricultural research. New York: The Bootstrap
Raman, K.V., and T. Balaguru. 1990. Human resources development as a key to the
success of farming systems. Journal ofFarming Systems Research-Extension 1(1): 153164.
Rasmussen, W.D. 1989. Taking the university to thepeople. Ames: Iowa State University
R6bling, N. 1986. Extension and the development ofhuman resources. In G.E. Jones, ed.,
Investing in rural extension: Strategies and goals. Essex: Elsevier Applied Science
Publishers, Ltd.
R61ling, N. 1990. The agricultural research-technology transfer interface: A knowledge
systems perspective. In D. Kaimowitz, ed., Making the link:Agricultural research and
technology transfer in developing countries. Boulder: Westview Press.
Russell, J.F. 1986. Extension strategies involving local groups. In G.E. Jones, ed.,
Investing in rural extension: Strategies and goals. Essex: Elsevier Applied Science
Publishers, Ltd.
Shaner, W.W., P.F. Phillipp, and W.R. Schmehl. 1982. Farming systems research and
development. Boulder: Westview Press, Inc.
Wharton, C.R., Jr. 1963. The role offarmer education in agriculturalgrowth. New York:
The Agricultural Development Council, Inc.
Journalfor Farming Systems Research-Extension

Whyte, W.F. 1983. Toward new systems of agricultural research and development. In
W.F. Whyte and D. Boynton, eds., Higher yielding human systems for agriculture.
Ithaca: Cornell University Press.
Whyte,W.F., and D. Boynton, eds. 1983. Higheryielding human systemsforagriculture.
Ithaca: Cornell University Press.
Vol. 2, No. 2, 1991

Informal Research With Farmers: The Practice and Prospects in the Hills of Nepal
S.P. Chand and B.D. Gurung2
Nepal, in the Himalayan mountains of South Asia, has a very diverse topography. Approximately 83 percent of the total area of the country lies in the hills and mountains, where about 57 percent of the total population lives. Agricultural research and development in the hilly areas are constrained by inaccessibility of inputs, fragility of the mountain area, marginality of cultivated land, and diversity of the environment. For these reasons, agricultural research and development activities are mostly urban-based and urban-biased. It is rare to find those who believe farmers have something important to say to scientists or that the research methods employed by farmers are complementary to the formal research employed by scientists. Formal research methods, such as on-farm and on-station research and government minikit distribution, usually have been confined to accessible areas and to the fields of a few influential farmers.
This article describes the background and objectives of Informal Research. A detailed methodology followed by Pakhribas Agricultural Centre (PAC) for the implementation of Informal Research is described. The article also examines the complementary roles of Informal and Formal Research, how "Informal Research differs from conventional, nationally adopted minikit programs, and problems and prospects for the incorporation of Informal Research into the National Agricultural Research Systems.
1 Paper presented at the Tenth Annual Farming Systems Research-Extension
2Symposium, Michigan State University, East Lansing, October 14-17, 1990. Chief Agronomist and Senior Agronomist, Pakhribas Agricultural Centre,
Dhankuta, Nepal.

Pakhribas Agricultural Centre, a multidisciplinary research and resource center, started its research and training activities in agriculture, forestry, and livestock in 1975. In 1982, work was expanded to include four hilly districts of Koshi zone, an area of more than 7,000 km2 with 100,000 farm families. Following the establishment of the National Agricultural Research Center (NARC), which coordinates all crop and livestock research in Nepal, PAC has become an important part of the national agricultural research network. Since July 1990, the center has had a mandate to cover the eleven districts in eastern Nepal, an area of 21,267 km2 with 252,436 farm families.
Based on more than 10 years of on-station and on-farm research, PAC developed many recommendations suitable to farmers in the hilly regions. In order to bring appropriate technology and varieties to the farmer at the right time, in 1985 PAC took responsibility for supplying minikits3 to farmers in four Koshi hill districts, according to the target fixed by the Government Commodity Research Program (NCRP) on major crops (e.g., rice, wheat, maize, and legumes). These minikits greatly helped to disseminate crop varieties suitable for hill conditions and discouraged the practice of importing varieties to the hills that were selected for Terai (the plains area) and other agroclimatic regions.
Since the involvement of PAC in the minikit distribution program, the effective feedback of minikits has increased by more than 50 percent (the national average is less than 20 percent). During PAC's two-year involvement with minikit distribution (1985-1987), however, it became apparent that distribution of minikits was confined to certains area and certain groups of influential farmers. According to the district agricultural extension program, minikits reached fewAgriculture Services Centers (ASC) and there were many remote areas where minikits were not distributed. Limited distribution was attributed to poor infrastructure (most areas are accessible only by foot) and to the tendency to favor resource-rich farmers and influential political leaders.
In order to reach all groups of farmers, PAC initiated the Informal Research Package (IRP) distribution program through its own on-farm, site-based Junior Technical Assistants (JTAs) and other research staff visiting on-farm sites. Informal Research Packages are small packets of seeds (usually 100 to
Minikits are packages of seed of one or two promising varieties (sometimes with fertilizers) sufficient for planting in one ropani (500 m 2) of land (Green, 1987).
Journal for Farming Systems Research-Extension

1,000 g) packed in a cloth bag with a brief description of the varietal names included in the package, the person who distributed the package, the amount of seed, distribution date, and altitude (high, mid, low). The Informal Research Package also is a tool of research-extension to reach more farmers at low cost.
When the IRP program was initiated, the main objective was to disseminate technologies to as wide an area of the hills as possible. It was believed that evaluation of the spread of technologies by farmers could be made after two to three years. Later, it became necessary to collect feedback as soon as possible, enabling researchers to modify technology in a shorter period of time.
PAC launched the IRP program with the following objectives: to test proven and promising crop varieties (local or improved) with a wider range of farmers covering different castes and gender, to introduce new crops found suitable to the different environments of the country, to involve farmers in evaluating varieties for themselves under whatever management conditions they may consider appropriate, to get feedback from farmers to further improve/modify technology, and to develop direct links between all groups of farmers and the researcher in wider areas of the hills.
In formal Research with farmers has a very flexible methodology in comparison to any conventional, formal research methodology. Following are the main steps usually followed in carrying out Informal Research with farmers.
Distribution of Informal Research Packages
No single person is responsible for distributing IRPs to farmers. It could be any staff person visiting farmers' fields or trekking in the hills, a group of staff people performing Samuhik Bhraman (a joint trek done by multidisciplinary groups; Chand and Gibbon, 1990; Chand and Thapa, 1990), trainees who come to PAC, or farmers who visit the center or research sites. IRPs also are distributed to farmers by field-based JTAs at HatBazaar (the local weekly market) or during regular supervision of on-farm trials. Interested farmers sometimes come to the JTA's house to get IRPs for a given crop.
PAC distributes about 500 Informal Research Packages annually. They contain various crops, such as maize, wheat, rice, lentil, chickpea, vegetables, potato, millet, and peas. Three-year results of Informal Research on various
Vol. 2, No. 2, 1991

crops clearly showed that the feedback on some new crops such as lentil, chickpea, and wheat, were higher (64 to 81%) than for relatively traditional crops, such as rice, millet, potato, and peas, which ranged from 27 to 56 percent. It was observed that farmers are curious to test new crops. Small quantities of seed supplied in IRPs greatly help to reduce the risk of crop failure.
PAC staff involved in the distribution of IRPs provide minimum information about the crops, such as variety name and the altitude at which the crop can be grown. All decisions on the management aspects of a particular crop, (e.g., when and where to plant and acceptance or rejection of crop varieties) are left completely to the farmer.
A complete record of IRP distribution is kept in the Agronomy Section. The record contains the names of farmers, districts, villages, seasons, and the name and quantity of crop/variety distributed.
Supervision and Feedback Process
One of the important objectives of Informal Research is to get feedback for further improvement/modification of technologies. However, there is no hard-and-fast procedure laid out for the supervision process. As there are no specific staff designated for supervising IRP programs, staff people visit Informal Research sites during the regular supervision of on-farm research sites, and the on-farm JTAs visit Informal Research sites when they are not busy with their on-farm research activities. Field staff also visit Informal Research collaborators enroute to their regular on-farm sites. They discuss crop performance with farmers and, if possible, observe the site. Some JTAs plan their time to see the crop at the maturing stage and others at harvest time. Some collaborator-farmers call on JTAs if there is a problem with the crop (e.g., disease, insect attack, or hail damage) or if the crop is performing well and is distinctly superior to local varieties. If the crop is very new to the farmers (e.g., chickpea, lentil), some may seek advice at regular intervals. If JTAS are unable to visit the fields of farmers who live very far away, they try to get feedback at Hat Bazaar, village fairs, health posts, and the Village Cooperative Office. Neighbors or relatives of collaborating farmers sometimes give feedback on crops to the JTAs, too.
Reporting Mechanism
Field-based JTAs submit weekly and monthly reports of all on-farm research in their areas. In these reports they also describe the performance and
Journal for Farming Systems Research-Extension

problems of Informal Research being carried out in farmers' fields. The information is analyzed and compiled by an agronomist at the center. Immediately after crop harvest or at a mutually convenient time, farmers and field JTAs meet to fill in the response card and send it to PAC for further processing. Questions asked in the card are very simple and qualitative and are based on socioeconomic and agronomic criteria, such as taste, market, byproducts, yield, and so on. (See Appendix A for a translated version of the evaluation form.)
Reporting and documentation of Informal Research results are vital for further improvement of the Informal Research program. Once the results are obtained from the field, they are compiled and tabulated and site-specific and overall reports are prepared. The reports are presented during the regional and national crop seminars.
A Comparison of Informal and Formal Research
Informal Research carried out by PAC is complementary to, and by no means a substitute for, Formal Research. Most of the varieties included in the IRP program either arc tested and found superior during Formal Research or are relatively new to the farmer but have been found suitable elsewhere in similar environments. In other words, varieties found superior during onstation and/or on-farm trials are selected for Informal Research to test in very wide areas that no conventional research and/or extension system could reach. Other varieties included in Informal Research are those generated through natural selection or found most promising and stable in certain areas of the country. For these reasons, Informal Research carried out by PAC differs from the Informal Research and Development ofIndia and Bangladesh reported by Biggs (1980).
The links among farmer, Informal Research, and Formal Research are complex and strong. There are procedural differences between Informal and Formal Research too, in that Informal Research is conducted strictly in farmers' fields, whereas Formal Research is carried out both on-station and on-farm. Informal Research is completely designed, managed, and implemented by farmers; Formal Research usually is designed and managed by researchers and implemented by farmers.
In Nepal, complex farming systems exist under heterogeneous ecological and socioeconomic conditions, making it impossible to develop appropriate
Vol. 2, No. 2,1991

technologies suitable to all areas. In addition, available resources are not sufficient to generate technologies for those areas through Formal Research alone. Informal Research requires less resource investment and, thus, could be a reliable, complementary research approach. Formal Research on-farm trials cannot be transfered to other sites to receive farmers' reactions, for example, but the same technologies/varieties/crops can be distributed to a wide area of similar environments through low-cost Informal Research.
Scientists are now realizing that farmers are willing to try new technologies. They evaluate technology in terms of biological, physical, and socioeconomic feasibility. Farmers are more enthusiastic about experimentation if they are given a free hand to decide when, how, and where to plant. This could be one of the reasons that farmers have shown more interest in Informal Research than in Formal Research, which often is run under the guidance of researchers.
The Case of Lentils
After many years of on-station research, PAC identified trial varieties of lentil. Although a new crop in the hills of Eastern Nepal, lentil is a traditional crop of the Terai, where it usually is relayed with, or planted as a sequential crop after, rice. Research done at PAC and in nearby areas showed that lentils can successfully be grown after rice or relayed under rice in the irrigated low land (khet) in mid-altitude areas (1,100-1,700 m). During 1988-1990, 198 lentil IRPs were distributed in seven districts of Koshi and Mechi zones. The field staff reported excellent performance of lentil crops throughout the hills.
Many farmers of Madimulkharka (Sankhuwasabha District) have been growing lentil at high altitude (over 1,750 m), a situation in which lentil was never tested by researchers. This is because of the preconception that lentil does well only in the Terai and in mid-elevation hills, where the climate is somewhat warmer.
A farmer who received 500 g of lentil in an IRP planted one-third of the seed on khet land and paddy bund in July. He thought it would grow well in paddy bund, as do such other legumes as soybean and black gram (Phaseolus mungo). Unfortunately, the farmer discovered that lentil cannot be grown during the summer; his crop was heavily infested with summer weeds and his plants did not grow well in heavy summer rain. The farmer then intercropped
Journal for Farming Systems Research-Extension

one-third of the seed with potato during January in a high-altitude potatomaize system. The crop again did not grow well, this time owing to the cold temperature and, at a the later stage, to damage caused by premonsoon rain. The farmer continued his experimentation and planted the remaining seed during the first week of September, after harvesting potato in a potato-maize cropping pattern. The growth of lentil was good, and the crop utilized residual moisture for its development. The farmer thus was able to harvest lentil successfully during February. In this way the farmer not only discovered the proper planting time for lentil, he also provided feedback to the researchers that lentil can be grown successfully at high altitudes where a potato-maize system is practiced and land is kept fallow during the winter season.
Farmers growing lentil in high- and mid-altitude systems also reported that lentil biomass mixed with khole (kitchen waste and cooked bran) can be fed to animals; these farmers reported a 20 percent increase in milk yield of cattle and buffaloes.
The Case of Early Rice
Rice is a prestigious food in Nepal. Owing to population growth (2.6 percent annually), farmers have been putting great effort into crop intensification. In the lower altitudes (below 1,100 m), farmers grow two to three crops a year. A double rice crop usually is followed by winter wheat ifsufficient irrigation is available. Compared with mid-level and high hills, fertility is high in low-altitude khet land because those areas receive summer floodwater, which often is rich in plant nutrients. In other words, the land can sustain three crops per year, including a winter legume. Most of the local rice varieties, however, take too long to mature to allow farmers to plant three crops per year. In 1988, some IRPs of early rice (khumal-3 variety), which matures about two weeks earlier than local Tauli, were introduced in the lower altitude ofGhoretar (Bhojpur District). Farmers in that area planted khumal- 3 as early rice and after its harvest planted the same variety as main rice, enabling them to take one more crop in winter. Researchers at Pakhribas did not know that khumal-3 could also be planted as main rice, but the farmers demonstrated that it is possible.
Informal Research for Crop Intensification
The rice-fallow cropping system is one of the most important cropping patterns of the eastern hills of Nepal. About 40 percent of the total cultivated
Vol. 2, No. 2, 1991

rain-fed lowland falls under this pattern. Fields usually are left fallow after the rice crop because of the lack of sufficient moisture to grow a subsequent crop. It has been found that even some of the drought-tolerant and forage crops cannot be grown successfully due to grazing pressure. Agronomy and livestock sections of PAC have demonstrated that drought-tolerant legumes, such as chickpea, and some of the forage crops, such as vetch, can be introduced successfully through IRPs with the involvement of a group of farmers. Grazing can be avoided when all surrounding farmers feel responsible for saving the crop. In this way, farmers can obtain an extra crop and quality forage after rice harvest (Chand and Gibbon, 1990; Chand and Thapa, 1990). This shows that Informal Research also can be a strong tool for mobilizing groups of farmers having a common interest.
In the winter of 1987, when the IRP program began, distribution and supervision were conducted by field-based staff during their regular on-farm research activities. Due to the abundance of on-farm activities, field staff could not devote much time to supervision or collection of feedback from Informal Research. This resulted in comparatively less feedback at that time. However, feedback can be obtained after a few years when farmers adopt new varieties in larger areas. Another problem pointed out by the field staff was their inability to visit Informal Research testing sites, especially those in remote areas. This suggests that either separate staff should be employed exclusively for the IRP program or that Informal Research on certain crops should be conducted in a cluster in one area, at least for one season, and moved to another area for the next season. In other words, crop varieties should be given to a particular agroecological zone where the crops are likely to perform well. For example, IRPs of maize, chickpea, and pigeon pea can be distributed in one season to low-altitude, red-soil zones where soil fertility and irrigation are a problem, and maize, potato, and peas IRPs to high-altitude zones in another season. This practice may help staff in collecting feedback more efficiently in a relatively shorter time.
Farmers, especially resource-rich farmers, are familiar with minikits. Small quantities of seed in IRPs without external inputs (such as fertilizers) usually are not well received and often are planted in marginal and unproductive land. Results, therefore, are comparatively poor, which may negatively impress neighboring farmers.
Journal for Farming Systems Research-Extension

Farmers sometimes take IRPs of seeds out of curiosity, but then fail to plant them in the field because of insufficient land or they store the seeds improperly for the next season's planting. This may waste seed.
Altogether, there are 25 different farm-types and stations in Nepal that have certain command areas representing different agroecological zones. The farms and stations run their on-farm and outreach research activities in those command areas. The IRP program may substitute minikit programs supervised by district extension, at least in those command areas of farm and station. Experience has shown that it often is difficult to get effective feedback from minikits, because the minikit distribution and supervision responsibility is taken by district extension workers who do not come under the direct chain of command of researchers. The bureaucratic and single chain of command of the present governmental set-up has created problems in getting efficient feedback. If the responsibility for Informal Research is taken by researchers, district extension workers will have more time to understand farmers' problems and to organize input supply in the district. The linkage and understanding between extension workers and farmers will be strengthened further if they are involved in extending those technologies already found promising in farmers' fields after Informal Research. In addition, this process may drastically reduce governmental investment in minikits. It is estimated that the National Maize Research Program alone spends about Rs 250,000 (US$8,333) annually on minikits. Proven technology from Informal Research can be introduced to hill farmers through farmer-to-farmer seedexchange programs, such as giving some seed to neighboring farmers if the variety performs well. It may greatly help local farmers to get the right type of crop variety at the right time at low cost.
In a country like Nepal, which cannot afford to conduct intensive Formal Research, an Informal Research Program could be an effective, complementary approach to Formal Research.
Experience in the hills has shown that improved technology, through Informal Research, can be taken to wider areas among a wider spectrum of farmers covering different castes and gender with minimum cost. Informal Research gives a free hand to the farmers to decide whether to accept, reject, or modify any technology given to them. All management decisions, such as where and when to test IRPs, are left completely to the farmers. Informal
Vol. 2, No. 2, 1991

Research with farmers has enabled researchers to get direct and firsthand feedback that helps researchers to improve or modify technologies.
The authors would like to thank Mr. C. Borman, former Director ofPAC, for nominating us to participate in the symposium. Special thanks go to Dr. R. James Bingen, Program Co-chair of the Tenth Annual AFSRE Symposium, for arranging a travel grant to the senior author to attend this symposium. Our colleagues at PAC, especially R.J. Khadka, Y.B. Thapa, and Bijaya Bajracharya, are acknowledged for their valuable comments on this report.
Biggs, S. 1980. Informal R and D. Ceres 76. Chand, S., and D. Gibbon. 1990. Samuhik Bhraman: A rapid and appropriate method
of prioritizing and replanning agricultural research in Nepal. Journal of Farming
Systems Research-Extension 1(1): 1-15.
Chand, S., and B. Thapa. 1990. An approach to research and development for hill
farming systems: The experiences of Pakhribas Agricultural Center. Paper presented at International Symposium' on Strategies for Sustainable Mountain Agriculture,
September 10-14, 1990, ICIMOD, Kathmandu, Nepal.
Green, T. 1987. Farmer-to-Farmer seed exchange in the eastern hills of Nepal: The case
of Pokhreli Masino' rice. PAC Working Paper 05/1987. Pakhribas Agricultural
Centre, Dhankuta, Nepal.
Crop ...................... Variety ...................... Distributed am ount of seed .........................
A ltitude ....................... C ropping Pattern ....................................................................
F arm er's N am e ..........................................................................................................
A d d ress .....................................................................................................................
(1) Have you planted the provided seed?
- If not, why?
- If yes, where? In what types of land and why?
(2) Emergence ability of the given seed. Did you find any reasons for bad or good
(3) How do you like the varietal characteristics (plant height, maturity days, etc.)
of this variety?
(4) Have you observed any disease? If yes, explain.
Journal for Farming Systems Research -Extension

(5) Have you observed any insects? If yes, explain.
(6) This variety matured early or later than your local variety. Both, late and early
maturing variety have advantages and disadvantages. Because of early and late
maturity period, what are the major advantages and disadvantages of this
(7) Was there any problem in planting subsequent crop, owing to this crop/variety
(8) Do you think this variety is more responsive to chemical fertilizer/compost?
(9) How was production of this crop in this year, from your seed?
(10) Production from the given amount of seed:
(11) Was there any difference in postharvest activities (threshing, grinding, etc.) to
be done with this variety compared to local? If yes, what are they?
(12) Your view on its straw/stalk yield:
(13) How do you like grain color, taste of this variety?
(14) Have you saved the seed of this variety for the next year? Why? If yes, how
(15) Was this crop/variety grown before on your land? Why?
(16) Have you heard before about this crop/variety?
(17) Have you planted any improved cultivar of any crops prior to this? If yes, which
one and how much?
(18) Do you think your neighbors also are interested in participating in this Informal
Research Program? Why?
(19) We are planning to run this program again in the future. Are you interested in
participating in coming years? Why?
(20) Reactions from neighbors or comments on package:
Vol. 2, No. 2, 1991

Moroccan Experience in Establishing a Local Coordinating Network: Researchers-Farmers-Extension Agents
E. Elmzouri and L. EdwardsI
In Morocco, agricultural research and extension are housed in two separate institutions: research in the "Institut National de la Recherche Agronomique" (INRA) and extension in the "Direction de Vulgarisation et de la Reforme Agraire" (DVRA). The headquarters of both institutions are located in the capital city of Rabat, with regional offices in the different provinces of the country.
Communication between researchers and extension agents is limited. Researchers tend to conduct their trials on research stations rather than on private farms, and the results are communicated to the extension headquarters in Rabat through reports. There is almost no contact between the researchers and the extension agents, who have direct contact with farmers in the field. INRA publishes only a few bulletins, brochures, and leaflets for farmers. DVRA publishes bulletins and uses other forms of media to inform farmers about agricultural events. They also provide leadership for national crop operations, although these activities are not necessarily supported by local research findings.
The absence of direct communication between research and extension personnel means that extension agents often are ignorant of new and developing technologies. The absence of farmer inputs in decision-making means that research and extension programs often are out of touch with the realities faced by the farmer. As a result, farmers feel that neither researchers nor extension agents are working in their interests because neither invites them to contribute in the decision-making processes that establish the research and extension program goals. People at higher levels of government
Agronomists with the Technology Transfer Section of the Aridoculture Center, Settat, Morocco.

often send program orders to the local extension agents without taking into consideration farmers' needs or capabilities.
Our objective in this project was to reduce these problems at the local level by improving the working relationships among researchers, extension agents, and farmers of the Settat province of Morocco. The specific objectives were:
(1) to establish a functioning local coordinating network of researchers, extension agents, and farmers; (2) to develop ways of exchanging ideas that could improve agricultural production at the local level; and (3) to obtain institutional approval of this local coordinating network so that recommendations could be directed to the policy decision-makers in the respective institutions. This local network selected the new technologies for transfer that had been adequately tested in research trials and in coordinated technology transfer efforts. It also helped to train extension agents. The extension agents helped to coordinate programs of technology transfer With other development agencies within the project area. Thswas accomplished through a series of on-farm and in-class training courses. Field days were organized for farm leaders and officials of the development agencies. Farmers also were invited to evaluate the new technologies and were involved in defining research objectives and programs.
The coordinating network was initiated upon our first visit with the local extension agents, known as "CT agents" (Centre de Travaus). The purpose of these visits was to promote linkages among researchers, extension agents, and farmers. There was an exchange of information and an effort was made to coordinate a common plan for technology transfer.
The primary focus of the coordinating network was to include the technology- transfer team from the research side, the CTs from the extension side, and the farmers themselves.
In the fall of 1988, the individuals involved in the coordinating network met to discuss and plan projects for technology transfer. In these discussions, the responsibility for nominating technologies ready for transfer was assigned to the head of the research -and-development service (SRD). The responsibility for identifying the regions appropriate for the technologies was assigned to the head of the extension -and -development service (SMV). The coordination between researchers and extension agents was assigned to the technologyJournal for Farming Systems Research-Extension

transfer team and the extension service (BIA), respectively.
The technology-transfer team had a major role to play in disseminating information and coordinating the activities of researchers, extension agents (and their supervisors, BIA), and farmers. Thus, reciprocal exchanges existed among the three partners concerning the development of new technologies and new research or extension programs. For the first time, farmers were asked to evaluate and give recommendations on the research and extension programs that would affect their livelihoods.
The success of this coordinating network was affected directly or indirectly by a number of other institutions or agencies. Local political leaders (e.g., the Governor), SONACOS (the official seed increase and supply company), and FERTIMA (the official fertilizer supply company) were involved in improving agriculture in the Settat region. These agencies have control of important agricultural technologies, such as availability of new seed varieties and new fertilizer formulations. Successful coordination of activities depended on cooperation between these institutions and the technology-transfer programs. The network was officially sanctioned in the spring of 1989.
On-Farm Trials
One of the major projects of the coordinating network was to develop a series of on-farm trials. The technologies that had been developed and tested by Moroccan researchers were presented to the other members of the network. The coordinating network agreed on the technologies that were appropriate for demonstration, and the experimental design and work programs were defined in detail by the research and extension partners.
Early in the growing season, farmers were contacted and arrangements were made for the on-farm trials. The farmers were selected by the technology-transfer specialist (Research) and the extension agents according to the following criteria: (1) the farmer should be someone who is open, receptive, and cooperative, (2) the farm should be within the project area, (3) the farmer should own the farm, and (4) the farm should be on a main road, if possible. Close collaboration among researchers, extension agents, and farmers was important for the successful planning and establishment of the trials. This required good management and follow-up procedures. Each of the three partners was present at each stage of the on-farm trial.
Vol. 2, No. 2, 1991

The technology-transfer specialist and the extension agents visited each farm to determine which parts of the technology-transfer package were appropriate. If the farmer was more interested in milk and meat production, we decided to demonstrate technologies related to forage production and quality instead of cereal production.
The on-farm trials were quite large (1/4 ha. per treatment for demonstrations with many treatments and 1 ha for those with one treatment). We demonstrated a package of technologies that we thought should accompany each crop variety. Where possible, we demonstrated an integrated system of technologies (cereals, food legumes, and forages) on the same farm. The package of technologies included recommendations on date and density of planting, fertilizer combinations and quantities, soil preparation, weed control, and crop varieties, as listed in Table 1.
A total of 15 on-farm trials involved a total area of 40 .48 ha. Thirty-five ha were planted to cereals (durum, bread wheats, and barley), 5 ha were planted to forages, and 0.48 ha planted to food legumes. The emphasis on cereals reflects the importance of cereals in the region. Two farms had all three technology packages. The cereals package was demonstrated on seven farms.
Training Courses
We held our first seminar with all the extension agents of Settat province early in the growing season. The main objective ofthis meeting was to explain the ideas behind the technology-transfer effort and the methodologies and approaches to be used. A general work plan concerning the collaboration between the technology transfer team and the extension agents was discussed. After this initial meeting, regional meetings were held on the farmers' fields. At this stage we began giving seminars about the new technologies and their use by farmers. We explained our objectives in establishing a local coordinating network of researchers, extension agents, and farmers. Seminars were held during tillering, heading-flowering, and maturing stages of the cereals. All of these field-tour presentations were conducted at the on-farm trials to train the extension agents and farmer-leaders.
Field days: Field days were held three times during the growing season to let farmers and extension agents judge for themselves the demonstrated technologies. This gave us an opportunity to explain the differences between the varieties and to explain the importance of each component of the demonstration package. These were large meetings attended by 35 to 125 farmers, extension agents, and local leaders. These field days were organized
Journal for Farming Systems Resea rch -Extension

Table 1. Details of Technologies Demonstrated in the On-Farm Trials, Their Locations, and the Area Used
Demonstrated Area Location and Technical
technologies (ha) total area practices
10 Durum Wheat 2.5 Od. Said (5 ha) soil preparation,
4 Bread Wheat 1.0 Ain n'zagh (5 ha) fertilization, date
5 Barley 1.5 khmis Ben Rahal (5 ha) and densities of
Od. Afif(5 ha) sowing, weed control,
Ben Guerire (5 ha) and seed-drill use Jamaa shaim (5 ha)
Ben Ahmed (5 ha)
Total Area 35 ha
Oat-vetch mixture 1.0 Od. Said (1 ha)
Khmis G'dana (1 ha)
S. El Aydi (1 ha)
Native medic 1.0 Khmis G'dana (1 ha)
Ley farming 1.0 Ouled Said (1 ha)
Australian medic
Total Area 5.0 ha
Winter chickpeas 0.16 Ben Ahmed (0.16 ha) soil preparation,
4 varieties Od. Afif (0.16 ha) fertilization, row
Od. Said (0.16 ha) spacing, date and densities of sowing
Total Area 0.48 ha and weed control
Total 15 Farmers
40.48 ha
Vol. 2, No. 2, 1991

in cooperation with the local extension agents.
Brochures and bulletins: Large numbers of "Fact Sheets" were prepared for distribution to extension agents, farmers, and other visitors during the field days. Small charts showing the names of varieties and their locations within the trials also were distributed. All leaflets were written both in French, the language used by the officials, and in Arabic, the language used by the farmers.
Invited politicians and mass media: Additional meetings and field tours were organized for the local and national political leaders responsible for agricultural development. Representatives of national and international research and development institutions and sponsoring agencies (USAIDMIAC-INRA) also attended these field tours. Representatives from the mass media (video recording and national newspapers) were invited on several occasions, so that they could report to the nation the value and importance of research on improving food production for Morocco.
Results of On-Farm Trials: Cereals
The on-farm plots were machine-harvested and weighed. These data are presented to illustrate how well the new varieties and cultural practices compared with current farmer practices (Figures 1 through 4). For each variety, including the one used by the farmer, a regression was calculated for the yield of the variety against the average yield of all varieties at the respective locations. This average yield is referred to as the "Environmental Index." This technique allows us to determine whether the new varieties do as well as, or better than, the farmers' varieties in both poor and good environmental (yield) conditions. Some varieties may do well under good yield conditions, but may not do well under poor yield conditions.
Durum Wheat: In the durum wheat trials we included the farmer's variety; an old standard variety (#2777); three recently released varieties (Marzak, Karim, and Cocorite); and seven promising candidate varieties (E15, E21, #1715, #1718, #1727, and #1728). Under poor environmental conditions, the new varieties did not yield better than the farmers' or #2777 varieties (Figures 1 and 2). Under good environmental conditions, the following varieties yielded more than the two standards: Karim, #1718, #1727, #1715, E21, and E15.
Bread wheat: In the bread wheat trials we included the farmer's variety; an old standard variety (Nesma); a recently released variety (Marchouch 8); and two new candidate varieties (Saada and Alkanz). Under poor environmental conditions, Saada outyielded all other varieties (Figure 3). Under good
Journal for Farming Systems Research-Extension

a E15
40- E21
35- 1715
30- 71
-25- vFarmer
020 215
-5 I I II I
2 6 10 14 18 22 26 30 34
Environmental Index (e), (QIls/ha)
Figure 1. Grain Production (Qls/ha) of New Durum Wheat Varieties, 1988-1989, Part 1
3 1727
- 1728
30 Marzak
,- Karim
- 25 Cocorite
v Farmer
15 10
2 6 10 14 18 22 26 30 34
Environmental Index (e), (QIs/ha)
Figure 2. Grain Production (Qls/ha) of New Durum Wheat Varieties, 1988-1989, Part 2
Vol. 2, No. 2, 1991

x Saada
30 Alkanz
30 Marchoch 8
L Nasma
@ 25 + Farmer
- 15
6 8 10 12 14 16 18 20 22 24 26 28 Environmental Index (e)
Figure 3. Grain Yields of New Bread Wheat Varieties
4 Acsad 60
40 + Acsad 176
35 Tamllalt
& Asni
@ 30 x 905
v Farmer
-0 20
15 10
5 i I I i I I
10 14 18 22 26 30 34
Environmental Index (e)
Figure 4. Grain Yields of New Barley Varieties, 1988-1989
Journal for Farming Systems Research-Extension

environmental conditions, both Saada and Marchouch 8 outyielded all other varieties. The candidate variety, Saada, was resistant to Hessian flies, a major pest in Morocco, and was a good performer at all locations. It did have some lodging problems under good environmental conditions.
Barley varieties: In the barley trials, we included the farmer's variety and five recently released varieties (Acsad #60, Acsad #176, Tamllalt, Asni, and #905). Under poor environmental conditions, only Acsad #60 had higher yields than the farmer's variety (Figure 4). Under good environmental conditions, all five varieties yielded better than the farmer's variety.
We developed a local coordinating network of researchers, extension agents, and farmers in Settat Province, Morocco. This network was set up to help coordinate the development and demonstration of new technologies to farmers.
During the first growing season we were able to coordinate a series of onfarm demonstration trials, in which researchers, extension agents, and farmers participated. A series of training courses, field days, and official visits was organized and brochures and bulletins were distributed.
The results of the on-farm trials indicated that under poor environmental conditions none of the new durum wheat varieties yielded better than the farmer varieties, but several yielded better under good environmental conditions. The new bread wheat variety, Saada, which is resistant to Hessian flies, did better than the farmer variety under both poor and good environmental conditions. The recently released bread wheat variety, Marchouch 8, was the best yielder under good environmental conditions. Of the five barley varieties demonstrated, only Acsad #60 yielded better than the farmer variety under poor conditions, but they all yielded better than the farmer variety under good conditions.
Farmer reactions to the trials were very positive. They appreciated the effort to share new ideas, especially after we tried to understand their priorities and needs. They are eager to continue the program and there are many more who arc interested. There are now more than 40 farmers in the program for the next growing season. They are especially eager to cooperate if some of the input costs are covered by the experimental team.
The extension agents also appreciated the experience. After receiving training, they became much more effective in sharing this information with
Vol. 2, No. 2, 1991

the farmers. They seemed to recognize their need for sound research information and were eager to participate in the program. They made more frequent visits to the research center to confer with the technology-transfer specialist and researchers. Aridoculture Center researchers were also interested in seeing how their technologies performed on the farm. They were very helpful in training activities and in explaining technologies on field days.
Technology may be defined as the application ofscience to produce desired outcomes, and technology transfer as the process of duplicating these outcomes elsewhere. However, there is a growing sense that a particular technology is appropriate only in the eye of the beholder. The certainty that farmers can recognize appropriate technologies has generated strong opposition to researcher- designed technology packages and has given rise to arguments that agricultural research should be conducted with farmers on farmers' fields. The creation of a cooperative network such as the one discussed in this paper can be effective in assuring that researchers, extension agents, and, particularly, farmers participate equitably in the technology design and transfer process.
Journal for Farming Systems Research-Extension

The Imnportance of Risk in Technology and Diffusion: The Case of Small Maize Producers in Mexico
R.R. Marsh1
Input-intensive technologies that require purchased inputs often increase production and economic risk for small-scale producers, owing to their impacts on yield and net income variability. Imperfect access to labor, credit, and information creates additional uncertainty. The extent to which increased risk affects technology-adoption decisions depends on producer risk attitudes and abilities to bear risk. Risk-averse producers make agricultural decisions based on expectations of net benefits and variability in benefits. Production and economic risk can be minimized if recommended agricultural practices appropriately reflect on-farm agroclimatic conditions and working capital constraints and if effective extension is available to accelerate the "learning curve."
In this paper, low fertilizer and plant-density adoption rates and apparent allocative inefficiency in small-scale Mexican maize production are associated with the multifaceted uncertainties affecting producers. A general framework for comparing alternative technology generation and diffusion (TGD) approaches is presented, and two TGD programs operating in the research area-one government-directed, the other community-based-are compared as a case study. Maize recommendations and observed farmer practices are contrasted with risk-neutral economic optima to assess the importance of risk considerations in the study area. Producers' risk attitudes are determined, indirectly, by comparing producer adoption levels with risk-neutral solutions. The relationship between this revealed risk distribution and selected socioeconomic and institutional factors is examined econometrically. Finally, the net benefits from adoption of maize fertilizer and plant-density recommendations are calculated, and the effectiveness of alternative programs in generating and diffusing appropriate technology is assessed.
1Agricultural Economist, Research Associate, Center for Economic Policy Research, Stanford University.

TECHNOLOGY GENERATION AND DIFFUSION MODELS Four "prototype" approaches to agricultural technology generation and diffusion (TGD), with different economic criteria for formulating recommendations and treatments of risk, are summarized in Tables 1 and 2. These include:
Traditional programs: Optimal factor-use recommendations are derived by maximizing crop profits using results generated under experiment station conditions. Capital constraints, site-specific agroclimatic variability, and producer risk attitudes are not incorporated.
Modified Traditional programs: Optimal factor-use recommendations are derived from profit maximization, subject to a capital constraint, using mean on-farm trial and experiment station results. Technology generation is tailored to specific "agrosystems." Yield and price variability and producers' risk attitudes usually are not incorporated.
Farming Systems Research (FSR) approach: Technology is generated from results of on-farm experiments conducted in homogeneous "recommendation domains." Optimal factor-use recommendations include risk by considering yield and price variability under minimum returns and sensitivityanalysis frameworks.
Community-Based Rural Development: Agricultural research and extension are incorporated into a broader strategy to improve community welfare. Optimal factor-use recommendations incorporate FSR methodologies and community and farmer participation in on-farm research and adoption feedback. Field-level extension is emphasized.
In traditional technology generation and diffusion programs, the economically efficient input quantity is determined by equating the value of the marginal product to the factor price, assuming risk neutrality. This criterion often is inappropriate under conditions of yield or price uncertainty, especially when decision-makers face severe information and capital constraints. Under these conditions, expected utility maximization is a better approximation of farmer behavior. The mean-variance (EV) specification of expected utility used in this study is an acceptable approach when yield/price distributions are approximately normal.
Journal for Farming Systems Research-Extension

Table 1. Technology Generation and Diffusion Approaches Compared
Traditional Modified traditional
Objectives Maximize yields; increase per Improve farmer incomes and hectare income, increase national food production.
Unit of Experimental plots Farmers' fields
Research Basic and applied research; Basic and applied research; station
experiment station-based; pursuit and on-farm; technology "packages"
of technical optima. per agrosystem; pursuit of technical
and economic optima.
Extension Transfer of technology; focus on Agent responsible for multiple tasks; commercial farms; no field-level demonstration plots; occasional
technical assistance (TA) on small field-level TA; weak researchfarms, extension linkage.
Small Neglected; accept or reject Cooperation of leaders and progresfarmer transferred information and sive farmers for informaiton, onparticipation technology.
Farming systems research Community-based
rural development
Objectives Improve productivity of farmers' Promote social change, improved
resources; raise small farmer welfare, and community empoweradoption rates. ment.
Unit of Farming system Communities and households
Research Applied and adaptive research; Optimization of existing resources
on-farm w/farmer; technology with selective integration and adap"sequencing" per recommen- tation of modem practices; on-farm
dation domain; welfare- w/community; continuum of techimproving farming systems' nical and social options to meet
compatible innovations, varied rural community needs.
Extension Agent participates in farm research Frequent on-farm TA; agent and
validation, recommendation farmer-managed demonstration
formulation, and diffusion of plots; training in technical, social,
innovations; provides feedback and organization skills; emphasis
to researchers from farmers. on learning by-doing, building
trust, and community self-reliance. Small "Average" farmers consulted for Focus on resource poor farmers;
farmer farming systems information, rural communities define problems
participation problem identification, on-farm and research agenda, conduct
research, and ex-post assessment, research, and evaluate results; farmer innovations encouraged.
Vol. 2, No. 2, 1991

Table 2: Treatment of Risk in Technology Generation and Diffusion
Approach Economic criteria for technology recommendations
Farming Economic criteria per recommendation domain such that the
systems recommended treatment has:
(1) the highest net benefit among ordered dominant treatments and (2) a marginal rate of return equal to or greater than (_) the farmers'
minimum acceptable rate of return.
n nn
MRR= (7Y. -Y)P -(YYXi.+1x -XijPx.)
j-1 i-j-1 i=inputs, j=treatments;
(3Y_7Xij+IPx -+XFP"
i-lj-i ij i
MRRmin (cost of capital) xno. mos. +risk premium, s.t. VMPxTPxi(1+c),
c =.5-1.0.
Community Recommended formulation subject to:
based rural (1) on-farm and off-farm labor availability as f(local opportunity costs development of labor and family composition and training);
(2) farmer knowledge of technology and participation in research (6Y 6Kh > 0,Kh = human capital);
(3) agronomic, technological, and economic risks (ie. increases variability and/or probability of loss associated with technological
change; and
(4) community obligations and organizational structure (ie. communal decision-making, common resource use, cultural/religious calendar).
In E,V analysis, the individual's utility curve depends upon his/her marginal rate of substitution (MRSEV) of mean profits (E(s)) for variance in profits (V(n)):
dE(n/X) [6U/6V(t/X)]
(1) MRS= u=u =
EV dV(n/X) [6U/6E(n/X)] u=u,
where output price and quantity are stochastic. Using this definition
(Magnusson, 1969) MRS,.EV is negative, zero, and positive for risk preference, risk neutrality, and risk aversion, respectively. For risk aversion, the greater the value for MRSEV, the greater the willingness of the individual to trade higher expected profits for reduced variation in profits, holding utility constant.
Journal for Farming Systems Research-Extension

Table 2. Treatment of Risk in Technology Generation and Diffusion (continued)
Risk and risk preferences
Approach Agroclimatic Technological
Traditional Experiment station conditions; Assumption of perfect information,
variability reflected in station input availability and quality, and
locations, technology/climate independence.
Modified On-farm research by agro-system; Nontreatment, nonstochastic
traditional variability reflected in spatial and factors controlled in on-farm trials;
temporal distributions partial analysis of technology/
( 8 locations, 3 years). climate interactions; mean outcomes determine economic optima.
Farming On-farms research in flexibly Only treatment factors controlled
systems defined recommendation in on-farm trials; multifactorial
research domains; variability reflected in analysis of technology/climate
spatial and temporal distributions interactions; multiple, specific
(- 20 locations, 3 years); minimum recommendations not to increase
returns risk analysis. average production risk.
Community- On-farm research concentrated Farmers w/promoters manage all
based rural on resource-poor farms across aspects of on-farm trials with development rural communities; variability researchers as consultants; use
reflected in spatial and temporal of traditional risk-reducing
distributions (# of trials vary as techniques; concern for resource
needed) and in farmer risk sustainability.
The Cobb Douglas production function (Y = aXb) is used to estimate production elasticities, marginal value products, and input risk coefficients. For the Cobb Douglas production model, the first order condition for optimal input use under the EV framework is:
PvY~E(Y) pi
(2) Xi 1- (6Vy/6Xi)(MRSEV)
where is the elasticity of production of input X, X. is the input quantity, and p, and Py' are known factor and product prices. The marginal value product of input Xi is equated with its factor price compounded by a riskpremium equal to the marginal yield risk associated with input X weighted by the risk preference or attitude parameter, MRSEV. In this study only yield is stochastic. Prices are given but vary among sample farmers, as explained later in this article.
Vol. 2, No. 2, 1991

Table 2. Treatment of Risk in Technology Generation and Diffusion (continued)
Risk and risk preferences
Approach Economic Farmer preferences
Traditional Assumption of unlimited working Assumption of profit maximizing
capital and zero transaction costs; behavior with unlimited capital.
variable costs > net income common under small farm conditions.
Modified Unlimited and limited capital Assumption of profit maximizing
traditional options; variable costs > net behavior subject to a budget
income in some years and agro- constraint; broad groupings of
systems; zero transaction costs; farmers for TGD.
use of average prices.
Farming Recommendations subject to Farmer risk preferences studied for
systems minimum rate of return; oppor- microregional profile; range of
research tunity and transaction costs technology options reflects
considered; use of average farmer variation in acceptable risk levels
prices; sensitivity analysis for and subsistence needs.
impact of price changes on net
Community- Optimization of local resources Farmers assumed risk averse based rural and limited use of purchased w.r.t. k-using technology; TGD development inputs; reduction of transaction geared toward most risk-averse;
costs with collective action; use lower risk aversion with farmer-toof individual farmer prices, farmer extension, learning-bydoing, and community-level
In this analysis, as with Wolgin (1975), Moscardi and de Janvry (1977), and Antle (1987), individual risk attitudes are measured indirectly by calculating the residual between observed farmer input levels and optimal input use under risk-neutrality. This "revealed" risk measure, called R(s), reflects producers' allocation choices when faced with multifaceted sources of uncertainty (agroclimatic, technological, economic). Substituting R(s) for MRSv in Equation 2 gives:
(3) R(s) 1 [16VY/6Xi PyiE(Y)
Journal for Farming Systems Research -Extension