WORKSHOP ON AGRICULTURAL RESEARCH POLICY AND ORGANIZATION IN SMALL COUNTRIES
Wageningen, Setember 11-14,1984



/ v"


RESEARCHER/FARMER LINKAGE












Peter E. Hildebrand
















Prepared for presentation at the "Workshop on
Agricultural Research Policy and Organization in
Small Countries". Wageningen, The Netherlands

September 12-14, 1984


Secretariat
Sponsored by


: Ir. J. Leeuwangh / P.O.Box 59 / 6700 AB WAGENINGEN / Tel. 08370-19145.
: Agricultural University Wageningen / Technical Centre for Agricultural and Rural Cooperation (CTA) /
Directorate for Agricultural Research (PLO, NetherJands.&MiisLty. oLt.Agriculture & Fisheries) / International
Service for National Agricultural Research (ISNAR, Den Haag).















RESEARCHER/FARMER LINKAGES


Peter E. Hildebrand*





Researcher/farmer linkages are basic to effective agricultural

research policy and organization. This paper is organized into three

sections. The first discusses how, where, when, why, by whom and with

what methods researcher/farmer contact is carried out. Emphasized is

researcher/farmer linkage for purposes of technology development,

including dissemination. This purpose also serves other applied

research involving policy and infrastructure implications. The second

section deals with these policy and infrastructure implications.

Finally, the third section discusses the implications of

researcher/farmer linkage for research policy and organization.



RESEARCHER/FARMER LINKAGE

How

In a word, researcher/farmer linkages must be based on confidence.

Farmers historically are suspicious of any representative of

government. In many developing countries, the fear is that government




*Professor, Food and Resource Economics Dept., Univ. of Florida,

Gainesville, Florida 32611, USA.









knowledge may be utilized in some form for tax purposes or in other

means detrimental to farmers' best interests. In developed countries,

the concern may be with time spent with the government official at no

apparent benefit, if not a detriment to the farmer. Confidence can be

gained when farmers are convinced that researchers are going to be

working in a partnership with them to help solve problems which are

important to the farmers and have been articulated by them to

sympathetic representatives of government.

Technology development research in what is now called the Farming

Systems approach provides an entree for creating the confidence

necessary for an effective and efficient researcher/farmer link.

Research conducted on farms in partnership with farmers and on

problems of direct and immediate concern to farmers, particularly when

conducted on a realistic basis, provides farmers with confidence that

the researchers are attempting to help improve their lot.

Where

To be effective, technology development research must be

conducted in such a way that a clientele can be clearly identified.

Farming systems practitioners use the term Recommendation Domain to

identify homogeneous groups of farmers. Research conducted with

farmers who are representative of a specific domain provides the basis

for extrapolation to all farmers in the domain. Research efficiency

is improved because locations for on-farm research are selected for

specific characteristics. Resources are not wasted on obtaining

research results for conditions which do not apply to the

recommendation domain. Extension efficiency is improved both because

the clientele can be clearly identified and because the technology







3

developed and being disseminated precisely fits the agro-socioeconomic

conditions of the clientele.



When

On-far research in partnership with farmers is to be contrasted

with researcher managed and controlled experiments on farmers' fields

but conducted under the conditions of an experiment station. The

purpose of the two types of trials is distinct. Research conducted in

partnership with farmers and under real farm conditions is designed to

evaluate the effect of alternative technologies under the conditions

in which they would be put into use if they were to be adopted by

farmers in the recommendation domain. This means that each location,

or farm, is different in many ways from the other locations or farms

and that research techniques therefore need to be different from

experiment station practices where locational differences are

minimized by following prescribed experimental procedures.

Farmers are by nature experimental. However, many cannot accept

undue experimental risk. Technology should be evaluated under their

conditions and in partnership with them only when researchers have a

fairly high degree of confidence that the technology will be effective

when used under real farm conditions. This means that the

technologies usually will have been evaluated on local experiment

stations and perhaps in rented fields on a few farms prior to being

placed in farm trials in partnership with farmers. By extrapolation,

this implies that experiment station research, at least in part, is

oriented toward the solution of farmers' problems as defined by the

researcher/farmer linkage.









Why

In the process of technology development, there are at least

three critical reasons for this researcher/farmer linkage or

partnership. The first relates to quantity of resources, the second

to quality of resources and the third to the all-important factor of

management of scarce resources, including management time.

A standard production function is an estimation of the response

of the output of a production process to a variable input when other

inputs into the production process are held at a fixed or constant

level. The level at which these fixed inputs are held influences the

shape and/or level of the production function. Yield gap or

constraint analyses have amply shown that responses on farms differ

significantly from responses under controlled conditions such as those

used in usual experimental procedures. Conclusions as to the

significance of responses and/or their profitability based on higher

levels or better quality of fixed resources than are available to

farmers can lead to faulty recommendations. If other farmers try the

technology, they can be disillusioned at best or subject to a loss of

profit, cash invested or family sustenance at worst. Such a

situation can be avoided if technology is evaluated under the

conditions in which it would be used by farmers if and when adopted by

them.

Quality of resources can have an impact very similar to that of

quantity and at times the two are difficult to separate. Soil

quality, basic animal nutrition and reliability of irrigation water

deliveries are bio-physical examples. The socio-economic conditions

which farmers face, as distinct from physical, biological and

climatological conditions are also qualitative and quantitative and







5

have an important impact on the adoptability of technology. Farmers

are the ultimate decision makers regarding adoption. Prior evaluation

and understanding by researchers increases the probability that

technologies are acceptable to the clientele, but the farmers' own

evaluation is the final link in the chain. Technologies or goods and

services created in the absence of a close clientele/researcher

linkage often are rejected or utilized only after significant

modification. Absence of researcher/farmer linkage can only decrease

the efficiency of the technology development or research process. One

need look no further than the inefficiencies created in centrally

planned economies where decisions are made by bureaucracy with little

consideration of the needs, desires and conditions of the user. This

is also the reason why extension in many areas has become an agency

trying to sell poorly adapted products rather than one oriented toward

solving farmers' problems.

Three important functions of farmer management are evaluating

alternative technologies, adapting them and learning to use those

which are being adopted. Wake (1984) describes two activities in the

learning process. One is the activity of learning from secondary

information, either oral or published in one form or another. The

second is hands-on learning. The shape of the learning curve is,

perhaps, debatable, but if one considers a learning process beginning

at a level of no-knowledge, an S shaped curve could be envisioned. In

a highly developed economy with a sophisticated farm clientele,

secondary sources can easily be conceived as allowing movement along

the learning curve to the point where initial hands-on learning

results in rapid gains. In a very poorly developed economy, with

little availability of secondary information, early hands-on learning









by individual farmers probably is a tedious process with only slow

gains during early attempts.

Early adopters provide a community learning experience which

augments the amount of secondary information available to later

adopters. Early adopters also modify or adapt a technology to local

conditions, so that the technology is more suitable to a specific

community. However, those who are better able to take the risk of

early adoption usually have a different resource base than later

adopters. Their results differ from those of later adopters with an

inferior qualitative or quantitative resource base.

The farming systems approach to technology development is an

organized complement to community adaptation and learning in

agriculture. It provides the additional benefit of being able to

adapt and evaluate alternatives under the conditions of the majority

of farmers in a community or recommendation domain and not just under

the conditions of the "most progressive" farmers. This increases the

efficiency of the technology development and adoption process and

effectively combines research and extension activities.



By whom

The term researcher is used here in a broad context. Researchers

at different levels in the institutional hierarchy will have varying

intensities of contact with the clientele. Those researchers who

comprise the on-farm research teams will have the most continuous

contact with farmers. In many small or poor countries, these teams

may be comprised largely of sub-professional personnel. Professional

level personnel in many countries may have to support two or more

on-farm teams. Their contact with the clientele will necessarily be









less but it is still critical. If support personnel do not work on

farms with the teamSas often as possible, they will not be able to

communicate with their own team members who would be speaking with a

much better understanding of reality. Even national level commodity

team scientists should maintain researcher/farmer linkage. This

linkage not only provides the researchers a better understanding of

the farmers' situation. It aso creates confidence and a sense of

accomplishment that is too often lacking among research personnel.

The farmers involved in researcher/farmer linkage are those who

are representative of a specified recommendation domain. A

recommendation domain is comprised of a group of farmers, homogeneous

with respect to specific activities on the farm. An individual farm

can be in more than one recommendation domain at any one time and can

change recommendation domains if the technology used is changed.

Individual farm members can also belong to different recommendation

domains. The women may be in one recommendation domain with their

crops while the men are in another with crops which are predominantly

managed by them. Commercial crops on a farm may be part of a

different recommendation domain than the subsistence crops on the same

farm.



Methods to provide researcher/farmer linkage

A farm is a complex organization comprised of many facets. Most

farms are comprised of one or more household units with a complex set

of functions designed to provide for the welfare of family members.

Many products are required and several means are used to achieve

desired ends. In order to approach an understanding of the farm, a

team of researchers from a number of disciplines is essential. It is









not sufficient for members of a number of disciplines to work

individually in a given area or on a given problem. Rather it is

critical that different disciplines work together in the technology

development process (Hildebrand, 1981). Technology development in

recent years has been oriented primarily toward biological

interventions. It was therefore reasonable that heavy emphasis be

placed on the biological sciences. However, it is also critical that

the social and economic sciences be included in the multidisciplinary

teams involved in researcher/farmer linkage.

That linkage usually begins with an initial characterization of

an area, frequently using a sondeo or rapid reconnaissance survey for

the purpose of identifying tentative recommendation domains,

evaluating constraints to the farming systems within those domains and

determining possible interventions for the improvement of those

systems.

Characterization, evaluation and refinement of recommendation

domains is a continuous process. The multidisciplinary team uses

several means including on-farm records, on-farm trials, directed or

verification surveys, and frequent researcher/farmer contact to

achieve a better understanding of the clientele and to initiate

evaluation of technological alternatives.

Biologic researchers must keep in mind that the major objective of

conducting on farm trials is to evaluate the potential response of

technological alternatives under the real and varied conditions to be

found on the farms in a specific recommendation domain. Disciplinary

training has convinced most biological researchers that it is

necessary to reduce sources of variation from non-studied variables to

a minimum in order to effectively determine significant differences









among levels of treatment or treatment variables. To gain the most

benefit from on-farm research, biologic researchers must comprehend

the value of working with variability among farms and not attempt to

follow disciplinary mandates that dictate controlling this

variability. One statistical procedure which shows great promise in

helping biological researchers to evaluate technological alternatives

when subjected to the variability of individual farmer management is

modified stability analysis (Hildebrand, 1984a). This analysis

utilizes the environment within which a product is produced as an

independent variable reflecting soils, climate, and socioeconomic
/lkCdi ^l i
conditions management. Against this independent variable, results can

be measured by any of the revelant evaluation criteria including yield

per 4o, production per unit of cash input, yield per unit of labor in

a critical period, or any other criterion which is relevant to farmers

in a recommendation domain. The procedure also provides a method by

which recommendation domains can be refined or partitioned.



Guatemala -- one indication of success

Perhaps the best example of a national research institute which

has followed the farming systems approach, is that of the Guatemalan

Institute of Agricultural ScienceV and Technology (ICTA). In the

early 1970's Guatemala embarked on a daring endeavor to significantly

modify the impact of its investment in agricultural research and

technology development. In 1973, when the new institute was

established, Guatemala was importing large amounts of the basic

grains needed as food. The primary goal of the Institute was to

achieve self sufficiency in the production of these basic grains

utilizing primarily the small farmers who produced approximately 65%









of the grain in the country. The methodology developed by ICTA has

been well documented and the results achieved within one decade are

gratifying if not startling. Imports have been reduced to a minimum,

and yields have been increased up to 70% even in the face of large

increases in area of production, (Table 1) which normally results in

reduced yields. Seldom, if ever, has a country achieved self

sufficiency in four basic food products simultaneously over such a

brief period of time.



POLICY AND INFRASTRUCTURE RESEARCH

The farming systems approach to technology development with its

strong researcher/farmer linkage, is directly amenable to augmenting

infrastructure and policy research. The multidisciplinary teams

involved at the farm level can provide direct information to policy

makers and infrastructure managers and can incorporate information

received from them in the development of alternative technologies

(Hildebrand, 1984b). Economists and social scientists, in particular,

can help provide policy makers and infrastructure managers with much

more realistic information on probable responses to policy and

infrastructure stimuli than is possible from studies of models based

on data from standard statistical survey techniques. Directed surveys

are occasionally conducted by the multidisciplinary teams for purposes

of answering specific questions regarding technology development. The

same types of directed surveys, if not conducted so frequently they

interfere with on-going work, could provide specific responses to

policy makers and infrastructure managers.

Because of resource constraints, many multidisciplinary teams

have a minimum number-- often only one --of economic and social









scientists. If research for policy and infrastructure is to be an

important component of the work of multidisciplinary teams with

researcher/farmer linkage, the proportion of social scientists and

economists necessarily must increase. This increase should not be to

the detriment of efforts in the biological sciences. Rather, the

teams should probably be made larger by one or two individuals. These

individuals would not have sole responsibility for policy and

infrastructure research but should be completely integrated into the

multidisciplinary teams so they have a thorough understanding of the

agro-socioeconomic conditions of the clientele.



IMPLICATIONS FOR RESEARCH POLICY AND
ORGANIZATION IN SMALL COUNTRIES



Poor countries in general, and poor, small countries in

particular have little justification for conducting anything other

than applied, problem-solving research. Larger, wealthier and better

developed countries, as well as the international agricultural

research center network, must carry most of the burden of the more

basic research activities required as input into the applied research

of poor countries. Fortunately, experience has shown that applied

agricultural research can be effective in solving both micro and macro

level problems in small countries and effective researcher/farmer

linkage is key to efficient applied research. It must therefore

receive top priority in research policy in small countries.

Effective applied research with strong researcher/farmer linkage

requires an investment in field personnel and the transportation and

logistical structure for them to be efficient in the field.







12

Incentives are required to attract quality personnel to isolated areas

in the interior of many countries. Administrative structures must

provide the flexibility required to operate efficiently without being

bogged down by bureaucratic paper work at central offices in place of

action in the field. Research policy must support this type of

structure and program.

The need for transportation, field logistics and incentives for

quality field personnel is often construed as unreasonably increasing

the cost of a nation's agricultural research program. However, if

emphasis is placed on efficient applied research with a strong

researcher/farmer linkage, the traditional large investment in elegant

experiment stations, offices and laboratories can be minimized.

Research policy should consider the investment in a strong field

research program as an alternative to and not an addition to an

expensive centrally located research facility. This is not to say

that support research is not needed. Rather support research can be

conducted with more modest facilities in-country and with more use of

the international agricultural research network.

Agricultural development will not occur as a result only of the

development of appropriate technology. Provision must be made for

required infrastructure to provide a constant and reasonably priced

supply of the technology or its components and market infrastructure

must provide efficient commercialization channels. National

agricultural research policy therefore must link technology and

infrastructure development. This linkage can be provided via

multidisciplinary teams with strong research/farmer linkage.

Development also cannot occur unless technology is made available

to farmers. Extension services are traditionally the purveyor of this







13

service. Effective multidisciplinary teams working in well identified

recommendation domains can work with 50 to 100 farmers each year. It

is well known, if not well documented, that good technology travels

rapidly from farmer to farmer and widespread adoption occurs even in

the absence of organized extension efforts. Strong researcher/farmer

linkage is an effective extension procedure and should be considered

by policy makers as such. Some specialized extension services can be

utilized by farm level multidisciplinary teams for providing

pamphlets, audiovisual materials and other equipment to improve

presentations at field days and less formal gatherings. Integration

of extension personnel into these multidisciplinary teams should also

be part of agricultural policy. This integration will involve a

small proportion of the total extension personnel of a country. The

remaining personnel can be freed for the many other duties which are

always placed with extension.

National agricultural research policy should also attempt to link

university level research with the applied research organization. On

the one hand, this implies that universities will not be directly

involved in the kind of farm level development research which has

been discussed. The nature of most university research does not make

it amenable to providing responsibility for development research.

However, this does not mean that research conducted at the university

level cannot be applied research and provide input into the main

agricultural research organization. A close university linkage helps

orient that research and provides students with valuable applied

research experience. It also provides the research organization an

opportunity to evaluate graduating students to help them in their own

personnel selection process.









Finally, national agricultural planning research activities can

certainly benefit from a strong linkage with the farm level

multidisciplinary teams. However, care must be exercised by national

planning groups not to usurp the time of the farm level teams. The

primary product of these teams must be the development of technology.

However, it has been seen that they can provide valuable information

for policy makers. If these teams are augmented with social

scientists or economists they can be particularly useful to national

planning activities.

In summary, strong researcher/farmer linkage is possible. It

increases efficiency of technology development, can serve as the focal

point for extension, university, policy, infrastructure, and

international agricultural research network linkages. A strong

researcher/farmer linkage, then, can easily be envisioned as being the

key ingredient in national agricultural plans for development.









Table 1.


Comparison of Production, Yield and Importation
of Basic Grains in Guatemala
1973-1983


Unit


Maize


Beans


Rice


1973
Production
Yield
Area
Imports

1983
Production
Yield
Area
Imports

1973-1983
Increase in:
Production
Area
Yield


metric tons
kg/ha
ha
metric tons


metric tons
kg/ha
ha
metric tons


659,530
1,180
558,920
72,050


1,031,260
1,636
630,350
2,440*


58,460
636
91,920
390


102,800
986
104,260


19,370
1,600
12,110
210


42,320
2,850
14,850
140**


60,840
1,360
44,740
150


99,040
2,080
47,600
160**


118
23
78


Source: Adapted from Banco de Guatemala
* Animal feed
** Improved seed only, not for consumption.


Sorahum


Maize Beans Rice Sorahum







16



REFERENCES


Hildebrand, Peter E. 1981. Motivating small farmers, scientists and
technicians to accept change. Agricultural Administration 8 (1980-81)
375-383.

1984a. Modified stability analysis of farmer managed,
on-farm trials. Agr. J. Vol 76, March-April, pp 271-274.

1984b. Summary of FSR/E participant, activities, products
and time frame. Farming Systems Support Project Newsletter. Vol.2
No.l. First Quarter. University of Florida, Gainesville.

Wake, J.L. 1984. The cost of learning by doing effect on technology
adoption. M.S. thesis, University of Florida.