Journal of Social Development in Africa (1997), 12,1,53-66 
Farmer Participation in Research: 
Implications for Agricultural 
Development + 
ERASMUS D MONU * 
ABSTRACT 
Recently, there has been a great deal of attention devoted to the notion of farmer 
participation in agricultural research. In this connection, there has been a challenge 
to the conventional research model which conceives of the fanner as a passive 
partner so far as the identification of the problem, the conduct of the research and 
the development of the solution is concerned. 
This paper reviews a number of models suggested for agricultural research, 
noting the strengths and weaknesses of the models. Based on the author's 
experiences and other case studies, the implications of farmer participation in the 
development and dissemination of agricultural technologies are examined. 
Introduction 
It could be difficult nowadays to find an agricultural researcher who would argue 
against the involvement of the farmer in the research process. However, it should 
be remembered that it has only been since the 1970s that an explicit recognition of 
the role of the fanner in the research process has been made. In fact, despite this 
recognition, most of the agricultural researchers in Africa, at best, pay only lip 
service to active farmer involvement in the research process. 
Those who have strongly advocated for greater involvement of the fanner in the 
research process argue that the failure of the 'Green Revolution' to improve the 
living conditions of the small-scale farmers could be attributed to the incompatibility 
of the Green Revolution technology with the conditions of the fanners. The 
argument further points out that had the small-scale farmers been involved in the 
development of the technology, the scientists would have been made aware of the 
group specificity of the technology. 
A revised version of a paper prepared for the 1995 Annual Meetings of the Canadian 
Sociology and Anthropology Association, Montreal, Canada, June 4-7. 
Professor of Sociology, Department of Sociology, Brandon University, Brandon, 
Manitoba. R7A 6A9, Canada.
54 Erasmus Monu 
It is not surprising therefore that since the 1970s different models/approaches have 
been put forward to improve the involvement of the farmer, especially the smallscale 
farmer in the research process. In this paper an attempt is made to review these 
approaches and to determine how and when does a particular approach help us to 
involve the small-scale farmer in the research process. The implications of greater 
involvement of the small-scale farmer in the research process with respect to 
technology development and dissemination are examined. 
The Consultative Group on International Agricultural Research (CGIAR) has 
developed the following typology of research: 
1. Basic Research - a research designed to generate new understanding. 
2. Strategic Research - research aimed at solving specific research problems. 
3. Applied Research - research designed to create new technology. 
4. Adaptive Research - research designed to adjust technology to the specific 
needs of a particular set of environmental conditions (ISNAR, 1984). 
Although fanner involvement in all the four types of research is desirable, we 
suggest that farmer involvement in applied and adaptive research is critical. 
Models in Agricultural Research 
There are different ways in which the suggestions for the involvement of farmers 
in the research process have been categorised. While some writers have categorised 
the approaches according to authors (Farrington & Martin, 1988), others have 
grouped the suggestions under models (Why te, 1981; Frankenberger, 1992; Acker, 
1992; Monu, 1993). In this paper our discussion will be based on the categorisation 
of models. 
The Research, Development and Diffusion Model (RDD) 
This model has been labelled differently as the "Horizontal Model of Research and 
Development" (Whyte, 1981) and the "Transfer of Technology (TOT) ModeF 
(Chambers & Jiggins, 1987). The Research, Development and Diffusion model 
looks at the process of agricultural research and development from the point of 
view of the originator of an innovation who bases his innovation on a presumed 
receiver's needs. The initiative of identifying problem areas is therefore taken by 
the researcher of the innovation who focuses on the design and development of a 
potential solution. This is followed by the dissemination of the solution to the 
receivers. Thus, in this model, the receiver is a passive partner so far as the 
identification of the problem or the design and development of the solution to the
Farmer Participation in Agricultural Research 55 
problem is concerned. In practice, what actually occurs is a chain of activities 
through which fanning ideas and practices are developed and tested in the research 
plots and then channelled to extension services for dissemination to farmers 
(Monu, 1982). 
The Research, Development and Diffusion model has been credited with the 
miracles of the 'Green Revolution.' However, it is this very 'success' that 
generated the criticisms against the RDD model. While the 'Green Revolution' 
technologies led to dramatic increases in crop yields, most of the beneficiaries were 
large-scale resource-rich farmers. The small scale resource-poor farmers fell 
behind further. In addition, these technologies were not sensitive to the environment 
It has been argued that the results of the adoption of the 'Green Revolution' 
technologies reveal the inadequacy of the RDD model especially as it applies to the 
resource-poor farm families. 
Although there have been several criticisms of the RDD model (Roling et al, 
1976; Roling, 1985; Sands, 1986; Monu, 1982), perhaps the most thorough and 
comprehensive review of this model has come from Chambers & Ghildyal (1985) 
and Chambers & Jiggins (1987). 
First, unlike the practice within industry, the RDD model does not differentiate 
between output-oriented science and client-oriented technology. In industry scientists 
are trained in market research and user participation in research. In addition, 
the scientists are encouraged to use methods which would enhance responsiveness 
to user concerns. On the other hand the RDD model is output-oriented and leads 
to a situation where scientists develop the technology and then expect the extension 
service to sell it to the farmers. Thus, in most cases scientists and pressure groups 
determine what is to be researched. The scientists then design and conduct the 
experiments under controlled conditions on experimental farms and laboratories 
and pass on the results to the extension service for transfer to farmers. Since 
resource-poor farmers are not part of the pressure groups which influence the 
research agenda, most of what is researched and the recommendations emanating 
from the research are often inapplicable to their situation. 
Agricultural scientists who operate within the RDD model are committed to the 
model because of the education and training they receive, funding and influence 
of government and commercial organisations, the research methodology associated 
with the RDD and the professional and personal rewards and incentives the 
scientists receive. 
The training is modelled on "learning from above and teaching to below" and 
this is reinforced by the curriculum which is concerned with scientific method and 
detail. The training contains very little about technology development and how to 
learn from fanners.
56 Erasmus Monu 
"By the time they leave universities, scientists have been deeply conditioned 
to believe that they know more thanfarmers, that their knowledge is 
superior, and that they should be the people who determine what research 
should be done and how it should be conducted" (Chambers & Jiggins, 
1987:39-40). 
Other sources of reinforcement for the RDD model are government and commercial 
funding and influences. Government and commercial organisations are more 
likely to give support to export cash crops and/or commercial food crops. This 
emphasis on commercial production directs attention to regions that have the 
natural resources (irrigation, rainfall, good soil) to support increased yields and 
focus on resource-rich farmers who could more readily take advantage of the 
technologies developed. 
The research methodology of the RDD model supports the commercial orientation 
of the resource-rich farmers. The RDD research methodology studies only 
a few variables which affect the farming system at a time. This type of methodology 
fits more the simplified cropping patterns of the resource-rich farmers (eg, 
monocropping). This type of methodology is unlikely to yield a comprehensive 
knowledge of the resource-poor fanning systems (eg, mixed cropping). The 
scientists' support for the RDD model is also reinforced by the personal and 
professional rewards received by the scientists. Rewards are based on output and 
not service. These are calculated in terms of publications in academic journals or 
how useful the findings are for commercial organisations. 
In addition to the above, Chambers & Jiggins (1987) suggest that a comparison 
of the physical, social and economic conditions which exist on research stations 
with those of the resource-rich and resource-poor farmers would indicate that the 
technologies produced by the RDD model are likely to be inappropriate for 
resource-poor farmers. In most cases the research efforts have been concentrated 
on the individual (mostly the male head of the family) as the unit of analysis. This 
ignores the fact that in Africa, women contribute a greater portion of the labour 
required in agriculture. Indeed, in certain areas women are completely in control 
of food production. 
Furthermore, using the farm family as the unit of analysis (rather than individuals) 
allows us to examine the distinct roles and multiple goals of individuals within 
the farm family, in addition to the recognition that farming is only one of several 
strategies within the farm family economy. It also enables us to analyse how the 
farm family adjusts to different demands in order to satisfy its m ul tiple goals which 
may compete with one another at times. 
Finally, the RDD model fits well with the principles of bureaucratic organisations 
which characterise most of the agricultural research institutes/centres in
Farmer Participation in Agricultural Research 57 
Africa. Such organisations are normally characterised by close adherence to 
formal procedures, a centrally-administered control system with a hierarchical 
structure within which one's authority is a function of his/her position and 
organisational rank. Organisational procedures are specifically designed to limit 
varied responses to individual or group demands since such variations are seen as 
interfering with the rational functioning of the organisation. 
Farming Systems Research and Extension Model 
The Fanning Systems Research and Extension approach was largely developed in 
reaction to the RDD model. 
"The concept FSRE explicitly recognises the value of the farmers' experience 
and their traditional experimentation as inputs into strategies for 
improving the productivity of existing farming systems" (Gilbert, et al, 
1980:14). 
In this approach, there are four successive stages in the research process - 
description, design, testing and extension. The description or diagnostic stage is 
undertaken to understand the constraints, flexibility and needs within the fanning 
systems. This information is then fed into designing, testing and extending 
improved technologies and strategies. Norman & Gilbert (1982) have identified 
five attributes of the FSRE approach: 
1. The objectives of the farm families are incorporated into the research process. 
This involvement of the farmers ensures the use of evaluation criteria that will 
be relevant to them. 
2. Efforts are made to take into account community and societal needs. For 
example satisfying the short-run needs of individual farm families could lead, 
in the long-run, to the degradation of natural resources and increased inequalities 
in welfare distribution. In addition, by involving fanners, the 
approach draws on the pool of knowledge in the group and thereby the 
researcher could start with the strengths of the system, thus minimising the time 
spent on 'rediscovering the wheel.' 
3. The strategy recognises the locational specificity of the technical and human 
element. This means that for research purposes, the farm population is divided 
into homogeneous sub-groups. This allows the researcher and indeed the 
extensionist to deal with a group of farmers with similar farming activities, 
social customs, access to support systems, comparable marketing opportunities 
and resource endowment
58 Erasmus Monu 
4. The process used is dynamic and interactive with links in both directions among 
farmers, researchers and extensionists. 
5. Finally, unlike the Research, Development and Diffusion approach, the FSRE 
approach is concerned with the entire farming system. Thus, it is able to deal 
with technical and non-technical or institutional issues. 
The FSRE approach is thus more holistic than the RDD model. Through a system 
analysis three subsystems are delineated. These are the Research Subsystem (the 
information-technology generating subsystem), the Extension Subsystem and the 
User Subsystem (farmers). These subsystems are envisaged to be in interaction 
with each other throughout all the stages - from description or diagnosis stage to 
the extension stage. 
Clearly the FSRE model has a number of advantages over the RDD model. As 
Collison suggests: 
"The farming systems perspective (FSP) identifies farmers' most (pressing) 
problems and best expansion opportunities and the appropriate 
technology to solve those problems and better exploit those opportunities. 
By this process it focuses attention on to recommendations most likely to be 
rapidly absorbed by local farmers, enhancing the cost effectiveness of 
research and extension efforts" (Acker & Sungusia, 1986:172). 
However, the available evidence indicates that the practitioners of the FSRE model 
have fallen short of their theoretical model. Although scientists have succeeded in 
working with farmers in diagnosis and testing, the: 
"...information is extractedfrom the farmers and their farms and analysed 
by scientists to diagnose and prescribe for thefarmers...The key decisions 
about what to try and what to do remain with the scientists" (Chambers & 
Jiggins,1987:45). 
Secondly, the FSRE model assumes a multi-disciplinary collaboration. This 
cooperation between social scientists and agricultural scientists is hard to come by. 
As Rhoades & Booth (1983:2) have observed: 
"Differences in perception and role definitions between biological and 
social scientists result in a mutual respect that is miserably low....the 
upstartoflhisdisciplinarytribalism(is)thatsocialandbiologicalscientists 
tend to line up on opposite sides of the fence and throw spears."
Farmer Participation in Agricultural Research 59 
Jerhaps this 'tribal warfare' could be partly attributed to the fact that the convenional 
methods of social investigation have not produced the relevant, useful and 
imely information required. In the desire of the social scientist to have a 
comprehensive database, the agricultural scientist could be frustrated with the 
tndless process of socioeconomic data collection. 
It is further argued that like the RDD, the FSRE lack "explicit focus on resourcewor 
fanners" since resource-rich farmers are seen as better informants and better 
Collaborators. These resource-rich farmers who have the resources to experiment 
vith the technology being developed are also seen as the most effective group for 
(iffusion. This means that in most of the on-farm trials that are undertaken most 
if the collaborators are resource-rich farmers whose conditions are very different 
rom the resource-poor farmers, who form the majority of African farmers. 
The Farmer-First and Last Model 
:hambers & Ghildyal (1985) and Chambers & Jiggins (1987) argue that in order 
o increase farmers' participation in the research process and to make the research 
nore relevant to resource-poor farmers (RPF), a change in the formulation of the 
'SRE model is required. Their model is referred to as the Farmer-First and Last 
FFL) model. The model starts: 
"...not with scientists and their perceptions and priorities, but with RPF 
families and theirs'. It begins with a systematic process of scientists 
learning from and understanding RPF families, their resource needs and 
problems. The main locus of research and learning is the resource-poor 
farm, rather than the research station and the laboratory. Research 
problems and priorities are identified by the needs and opportunities of the 
farm family rather than by the professional preferences of the scientist. The 
research station and the laboratory have a referral and consultancy role, 
secondary to, and serving, the RPF family. The criteria of excellence is not 
the rigour of on-station or in-laboratory research, or yields in research 
station or resource-rich farmer conditions, but the more rigorous test of 
whether new practices spread among the resource poor" (Chambers & 
Ghildyal, 1985). 
According to Chambers & Jiggins (1987), the ecological and social complexity and 
liversity of resource-poor fanners' farming systems demand two simplifications, 
lamely, large-scale surveys and massive multi-dimensional data analysis and 
eduction of dependence on multi-disciplinary teams. In order to effect these 
implications the scientist should "... directlyencourage andenableRPFfamilies 
fiemselves to identify priority research issues" (Chambers & Jiggins, 1987:112).
60 Erasmus Monu 
The eight major elements of the model which together lead to what Chambers 
and Ghildyal describe as reversals of explanation, learning and location are: 
1. Research priorities and conduct are determined mainly by the needs, problems, 
perceptions and environment of farmers. 
2. The crucial learning that takes place is the scientists learning from farmers. 
3. The role of the farmer is that of a client and a professional colleague at the same 
time. 
4. The role of the scientist is that of a consultant and a collaborator. 
5. The main research and development location is the farmer's fields and 
conditions. 
6. The physical features of research and development are mainly determined by 
fanners' needs and preferences. 
7. The explanation of non-adoption of innovation is sought in farm-level 
resources, failure of scientists to learn from farmers and research station 
constraints. 
8. The evaluation of the innovation is done through adoption by farmers. 
The FFL model has been criticised to be "extremefarmer-centric" (Farrington & 
Martin, 1988:21). It is argued that scientists and the scientific method have a more 
important role to play in the technology development process than suggested by the 
FFL model. However, others would argue that this "farmer-centric" view is 
justified, if only to emphasise the importance of the involvement of resource-poor 
farmers in the research process. 
Despite the fact that the FFL model has provided a number of excellent 
suggestions to improve farmer participation in the research process, all the 
recommendations cannot be adopted in a wholesale manner (Baker, 1991). 
According to Frankenberger (1992), farmer-articulated demands tend to relate to 
short-term priorities. Thus an exclusive focus on fanner priorities could lead to 
over-concentration of research on issues related to short-term benefits, to the 
neglect of those issues that would deal with long-term benefits, hence sustainable 
option systems. 
Secondly, it should be recognised that even among resource-poor farm families 
there are inter- and intra-differences in household priorities in terms of gender 
roles, geographical location of villages, etc. These differences must be taken into 
account in deciding which and how farmers would be involved in the research 
process. Furthermore, although farmer-conducted and controlled research may 
provide us with useful results, we can not run away from the fact that in order to 
convince policy makers and extension workers, 'quantitative' measures are often 
needed. In other words, some form of 'acceptable scientific method' is required.
Farmer Participation in Agricultural Research 61 
Finally, the attempt by the FFL proponents to give greater control of the research 
process to fanners is likely to be resisted by most national agricultural systems 
which are used to the RDD model. Thus, an educational programme is needed to 
convince the researchers in these institutions. In some cases compromises may be 
needed to gain acceptance of greater farmer involvement in the research process. 
Implications for Technology Development and Utilisation 
The review of the models for farmer participation in agricultural research presented 
above indicates that there is a growing recognition of the feet that fanners can 
contribute to the research process. Below an attempt is made to outline some of the 
implications the participation of fanners in the research process has for technology 
development and utilisation. 
Reduction in Research Time and Cost 
Farmer in volvement in the research process could save scientists time and cost The 
development and introduction of improved cotton in Northern Nigeria illustrates 
this point. The scientists, concerned with low yield of the local variety of cotton, 
set out to develop an improved cotton variety. The outcome of the research was a 
very successful one. The improved cotton was highly productive with demonstrated 
yield increase of 100 percent in the fanners' fields. 
The improved cotton was to be planted during the months of June-July. In 
addition, the improved cotton was recommended for sole cropping. The improved 
cotton also required spraying and the recommended spraying technology was a 
water-based method with a hand pump. It was found out that, even at a reduced rate 
of 135 litres per hectare, 800 kilograms of water per hectare were needed (Norman 
etal, 1974). 
Despite the dramatic increase in yield, farmers rejected the improved cotton. An 
evaluation (Norman, et al, 1974) indicated a number of reasons for the rejection. 
First, the improved cotton was to be planted during the months of June-July, 
exactly during the period when labour requirements for food crops are high. In the 
traditional system, this constraint is avoided by planting cotton after the food crops 
have been planted and partially weeded, a clear sign of the family's priority for 
maximisation of food crop production over cash crops. 
Secondly, the improved cotton was recommended for sole cropping while the 
predominant fanning system in the area is mixed cropping. In addition, the 
adoption of the improved cotton costs a significant amount of money. The fanner 
must not only use fertiliser but must also spray the cotton. The small-scale farmer 
is not likely to have the resources to finance such a project from his/her meagre
62 Erasmus Monu 
earnings, especially when the financial demand occurs at a time when the farmer's 
cash resources are lowest, that is during the rainy season before any crops are 
harvested. Above all, the average net return from cotton using recommended 
practices was only 13 percent better than cotton grown in crop mixtures. 
The fanners' rejection of the technology forced the scientists to re-examine it 
and to introduce changes. First, a later planting date was accepted to avoid 
competition with food crops. The scientists were able to develop a new package 
with equivalent yield performance. A new spraying technique using an oil-based 
insecticide and an ultra-low volume sprayer operated with a battery-powered 
spinning disk was also introduced. Despite these changes most of the fanners 
rejected the improved cotton because the cost was more than they could afford and 
the sole spray cotton did not fit their farming system. The cost and time devoted to 
the development of the sole spray cotton would have been significantly reduced if 
the fanners had been involved in the research process from the beginning, which 
would have enabled the scientists to gain an understanding of the traditional 
fanning system. It is also argued that where research budgets are severely limited 
a transfer of more responsibility for technology design, testing and adaptation 
would be of great assistance to researchers (Ashby, 1991). 
Selection of Relevant Problems 
One of the reasons for advocating fanner involvement in the research process is 
that fanners are more aware of their problems than outsiders and hence are in a 
better position to identify the issues to be researched. Although this could be a 
debatable point, the fact remains that the involvement of farmers would increase 
the relevance of research outcomes in the field. A supporting evidence comes from 
the International Potato Centre in Peru (Gamser, 1988). The Centre scientists, on 
the assumption that some potatoes commonly were stored over a long period before 
marketing, and since it was known that post-harvest losses occurred, considered 
that a declining quality of potatoes through storage would be a problem. The Centre 
scientists therefore devoted much effort to developing potato strains that would 
endure long periods of storage, but without consulting farmers. 
However, when they did bother to consult with the local farmers, they were 
surprised to find that what to researchers had seemed a critical issue was of little 
importance to the fanners. The farmers said that their big problem was sprouting 
during storage and the new strains developed by the scientists were just as bad as 
the traditional ones in this respect. Sprouting necessitated tedious, time-consuming 
pruning work before potatoes could be sold or used for re-seeding, and the fanners 
wondered whether anything could be done about this problem.
Farmer Participation in Agricultural Research 63 
Farmer Experimental Skills 
Fanner involvement in the research process is likely to improve farmers' own 
capacity for carrying out experiments. Although it has taken researchers a long 
time to accept the fact that fanners do carry out their own experiments, there is now 
agreat deal of evidence to support the existence of farmer-designed and conducted 
experiments. An example from the agro-forestry project in the Eastern Region of 
Ghana is illustrative of this point (Monu, 1994). 
The technicians of the project advised the farmers to establish their hedgerows 
four metres apart However, one farmer after establishing four hedgerows realised 
that the distance between the hedgerows was too short considering the type of crop 
grown in the area. Cassava, the main crop grown in the area, could have tubers 
beyond two metres in length. The farmer felt that the short distance between the 
hedgerows would lead to a situation where the cassava tubers could be entangled 
with the roots of the hedgerow trees. He therefore decided to experiment with three 
different distances between the hedgerows, the 4 metres suggested, 6 metres and 
8 metres. After harvesting the crops he concluded that the 6 metres was the 
appropriate distance. While the 4 metres did not provide enough space for the 
cassava tubers, the 8 metres was too wide and this resulted in inefficient use of the 
land. 
The evidence from the Farmer Innovation and Technology Testing Programme 
in Gambia has shown that farmers with primary school education can master the 
principles of experimentation (Ashby, 1991). 
Use of Indigenous Knowledge 
The integration of farmers' perspectives and knowledge into the research process 
could enhance the relevance and acceptability of the technology developed. We are 
aware that fanners' knowledge, especially technical knowledge, has certain 
limitations but the evidence available clearly indicates that in many cases scientists 
could have improved their research results and make the technologies recommended 
more acceptable if farmers' knowledge is seriously incorporated into the 
research process. 
According to Irvine (1987), farmers among the Runa of San Jose obtain their 
food through gardening, hunting and fishing. These gardens are cultivated under 
the slash and burn techniques. However, rather than just abandoning the field 
during the fallow period, the Runa engage in what Irvine refers to as "resource 
enhancement." This enhancement includes weeding naturally occurring new 
species, protecting and occasionally transplanting desirable fruit trees and other 
species such as coffee and cacao. The fallows tend to serve as game attractants to 
enhance hunting success since the fruit trees in the fallow serve as food sources for 
animals. Thus, in comparison with unmanaged fallow, the managed fallow has
64 Erasmus Monu 
greater diversity of species and greater economic subsistence value. Despite this 
knowledge the agricultural research in the area is based on a monocropping system. 
Little attention is paid to the resource-enhancement strategies of the farmers in the 
area. 
On the other hand, the integration of fanners' perspectives and knowledge 
assisted ICRISAT to more clearly define research priorities (Matlon, et al, 1994). 
In Burkina Faso scientists designed experiments to maximise productivity of a 
cowpea-sorghum inter-cropped system through increased planting densities of 
cowpea. Through participation in the research, the farmers concluded that the 
increase in productivity and the possible higher financial returns would not 
compensate for the changes that the technology wouldbring in the farming system. 
It was realised that greater cowpea densities increased the risk of animal damage 
to crops, labour requirements for weeding increased substantially, animal traction 
for weeding and ridging could not be used and the reduction in yields of sorghum 
(the staple food crop) was unacceptable. As a result of the above, ICRISAT 
abandoned the cowpea-sorghum research and concentrated on research to intensify 
production in sorghum-groundnut systems, an area of more interest to farmers. 
Conclusion 
The need to involve fanners in the research process should no longer be contested. 
The issues at hand should be when and how should farmers be involved. The paper 
has reviewed a number of ways suggested for farmer involvement in the research 
process. An attempt has also been made to delineate the possible ways in which 
farmer involvement in the research process could enhance the relevance and 
acceptability of technologies in the farmers' fields. 
The effectiveness of the models reviewed earlier could be evaluated against the 
participation categories developed by Farrington & Martin (1988). Farrington & 
Martin (1988) identified four types of farmer involvement in the research process. 
The first is "contacts." In this case the farmers' land and services are used to 
provide more agro-ecologically diverse conditions for verification of technologies 
developed at research stations. The second type of participation is "consultative." 
This is like the doctor-patient relationship: the researchers consult the intended 
users but they are the ones who make the bulk of decisions regarding content and 
methodology. "Collaborative" participation is the third category. This involves a 
continuous interaction between farmers and researchers, including how to go about 
cost-effective village level research. Inthelastcategory, "collegiate," thescientist 
not only actively consults fanners on specific technologies/ideas or methods of 
experimentation, but also actively works to strengthen the local capacity to conduct 
informal research and development at the individual and community levels.
Farmer Participation in Agricultural Research 65 
We suggest that if we are to develop technologies that would be relevant and 
adaptable to small-scale farmers, farmer involvement in the research process must 
move closer to the "collegiate" category described above. A meaningful involvement 
of farmers in the research process must include their involvement in the 
definition of the problem, in the conduct and evaluation of the research and in the 
dissemination of research results. 
References 
Acker, D G & Sungusia, D (1986) "A Study of the Role of Extension in Tanzania," 
pp 65-175 in C B Flora & M Tomececk Selected Proceedings of Kansas State 
University's 1986 Farming System Research Symposium, Fanning Systems 
Research Paper Series No. 13, Kansas State University, Manhattan, Kansas. 
Ashby, J A (1987) "The Effects of Different Types of Fanner Participation of On- 
Farm Trials," in Agricultural Administration and Extension, 24,pp 235-252. 
Chambers, R & Ghildyal, B (1985) "Agricultural Research for the Resource-Poor 
Fanners: The Farmer-First-and-Last Model," in Agricultural Administration 
and Extension, 20, pp 1-30. 
Chambers, R & Jiggins, J (1987) "Agricultural Research for Resource-Poor 
Farmers, Part 1: Transfer-of-Technology and Farming Systems Research," in 
Agricultural Administration and Extension, 27. pp 35-52. 
Chambers, R & Jiggins, J (1987) "Agricultural Research for Resource-Poor 
Farmers Part 11: A Parsimonious Paradigm," in Agricultural Administration 
and Extension, 27, pp 109-128. 
Chambers, R & Jiggins, J (1986) "Agricultural Research for Resource-Poor 
Fanners: A Parsimonious Paradigm," in IDS Discussion Paper 220, IDS, 
University of Sussex. 
Farrington, J & Martin, A M (1988) Farmer Participation in Agricultural 
Research: A Review of Concepts and Practices, Agricultural Administration 
Unit Occasional Paper 9, Overseas Development Institute, London. 
Frankenberger, T R (1992) "Farmer/Consumer Participation in Research and 
Development," pp 195-209, in "Collaborative Research Support Programs," 
Proceedings of the Workshop on Social Science Research and the CRSPS, 
INTSORMIL Publication Number 93-3, Lexington, Kentucky. 
Gamser, M. (1988) "Innovation, Technical Assistance and Development The 
Importance of Technology Users," in World Development, 16, pp 711-721. 
Gilbert, E H; Norman, D W & Winch, F E (1980) Farming Systems Research: 
A Critical Appraisal, MSU Rural Development Paper No 6, Michigan State 
University.
66 Erasmus Monu 
Irvine, D (1987) "Resource Management by the Runa Indians of the Ecuadorian 
Amazon," Unpublished PhD Dissertation, Stanford University. 
ISNAR (1984) Considerations for the Development of National Agricultural 
Research Capacities in Support of Agricultural Development, The Hague. 
Matlon, P; Cantrell, R; King, D & Benoit-Cattin, H (eds) (1994) Coming Full 
Circle: Farmers' Participation in the Development of Technology, IDRC, 
Ottawa. 
Monu, E D (1994) "Agro-Forestry in the Mampong Valley: An Attempt to 
Operationalize the Learning Process Approach," a paper presented at the 1994 
Annual Meeting of the Rural Sociological Society, Portland, Oregon, August 
10-14. 
Monu, E D (1993) "Research: The Missing Element in Rural Reconstruction," in 
Martha McKeehn (ed) Y C James Yen's Thoughts on Mass Education and 
Rural Reconstruction China and Beyond, IIRR, New York. 
Monu, E D (1982) "The Contribution of Adoption and Diffusion of Innovation 
Research to Agricultural Development in Africa," in African Studies Review, 
XX V.pp 117-126. 
Norman, D & Gilbert, E (1982) "A General Overview of Farming Systems 
Research," pp 17-30 in W W Shanner; P F Phillips & W R Schmehl (eds) 
Readings in Farming Systems Research and Development, West View 
Press, Boulder, Colorado. 
Norman, D E; Hay wards, J A & Hallam, H R (1974) "An Assessment of Cotton 
Growing Recommendations as Applied by Nigerian Farmers," in Cotton 
Growing Review, 51, pp 266-280. 
Rhoades, R E & Booth, R H (1983) "Interdisciplinary Teams in Agricultural 
Research and Development," in Culture and Agriculture, 20, pp 1-7. 
Rhoades, R E & Booth, R H (1982) "Farmer-Back-To-Farmer: A Model for 
Regenerating Acceptable Agricultural Technology," in Agricultural 
Administration 11, pp 127-137. 
Roling, N (1985) "Extension Science: Increasingly Pre-Occupied with Knowledge 
System," in Sociologia Ruralis, 25, pp 269-289. 
Roling, N G; Ascroft, J & Chege, F (1976) "Innovations and the Issues of Equity 
in Rural Development," in Communication Research, 3, pp 155-170. 
Sands, D M (1986) The Technology Application Gap: Overcoming 
Constraints to Small-Farm Development, FAO Research and Technology 
Paper 1, Food and Agricultural Organisation, Rome. 
Taylor-Powell, E & Von Kaufmann, R (1986) "Producer Participation in 
Livestock Systems Research: Experience with On-Farm Research Among 
Settled Fulani Agro-Pastoralists in Central Nigeria," pp 257-276 in C B Flora 
& M Tomececk (eds) Selected Proceedings of Kansas State University's 
1986 Systems Research Symposium, Farming Systems Research Paper Series 
No 13, Kansas State University, Kansas.