Front Cover
 Concept: What was the basic technical...
 Design: How was this basic technical...
 Implementation: How was the project...
 Evaluation: How was the project's...
 Institutionalization: How did the...
 Project description sheet

Group Title: Case Study - A.I.D. Farming Systems Research and Extension Projects ; No. 4
Title: Malawi agricultural research project (612-0202)
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00073353/00001
 Material Information
Title: Malawi agricultural research project (612-0202)
Series Title: CDIE working paper
Alternate Title: Case studies of A.I.D. Farming Systems Research & Extension (FSRE) Projects, Case Study no. 4
Physical Description: 20 leaves : ; 28 cm.
Language: English
Creator: Byrnes, Kerry J., 1945-
Publisher: Center for Development Information and Evaluation, Agency for International Development
Place of Publication: Washington DC
Publication Date: 1984?
Subject: Agricultural extension work -- Malawi   ( lcsh )
Agriculture -- Technology transfer -- Malawi   ( lcsh )
Genre: government publication (state, provincial, terriorial, dependent)   ( marcgt )
non-fiction   ( marcgt )
Spatial Coverage: Malawi
Bibliography: Includes bibliographical references (leaf 17).
Statement of Responsibility: Kerry J. Byrnes.
General Note: Caption title.
 Record Information
Bibliographic ID: UF00073353
Volume ID: VID00001
Source Institution: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 80566686

Table of Contents
    Front Cover
        Front Cover
    Concept: What was the basic technical idea underlying the project?
        Page 1
    Design: How was this basic technical idea translated into a project?
        Page 2
        Page 3
    Implementation: How was the project managed by the host-country implementing agency, the TA team, and USAID?
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
    Evaluation: How was the project's performance measured or assessed?
        Page 12
        Page 13
        Page 14
    Institutionalization: How did the project provide for the implementing agency to develop a sustainable capability to continue to perform the types of activities supported by the project?
        Page 15
        Page 16
        Page 17
    Project description sheet
        Page 18
        Page 19
        Page 20
Full Text



Case Studies of
A.I.D. Farming Systems Research & Extension (FSR/E) Projects

Case Study 'To. 4

Malawi Agricultural Research Project (612-0202)I


Kerry J. Byrnes2

Center for Development Information and Evaluation
Agency for International Development
Washington, DC 20523

1This CDIE Working Paper is one of the case studies prepared
for a cross-cutting analysis of A.I.D. FSR/E projects, A Review of
A.I.D. ExDerience with Farming Systems Research and Extension
Projects (A.I.D. Evaluation Special Study, forthcoming). The 12
FSR/E projects reviewed in this series are:

Botswana Agricultural Technology Improvement (633-0221)
Gambia Mixed Farming and Resource Management (635-0203)
Lesotho Farming Systems Research (632-0065)
Malawi Agricultural Research (612-0202)
Senegal Agricultural Research and Planning (685-0223)
Tanzania Farming Systems Research (621-0156)
Zambia Agricultural Development Research & Extension (611-0201)
Nepal Agricultural Research and Production (367-0149)
Philippines Farming Systems Development-Eastern Visayas (492-0356)
Guatemala Food Productivity and Nutritional Improvement (520-0232)
Honduras Agricultural Research (522-0139)
ROCAP Small Farm Production Systems (596-0083)

Information on how to order any of the CDIE Working Papers this
series is provided on the last page of this report.

2Senior Social Science Analyst, Progran and Policy Evaluation
Division, CDIE. This case study, prepared under a CDIE contract
with Labat-Anderson Incorporated, is based on a review of project
evaluation documentation. Interpretation of the data reported is
that of the author and should not be attributed to A.I.D. or Labat-
Anderson Incorporated.

Malawi Agricultural Research Project (612-0202)

The Malawi Agricultural Pesearch Project (MARP) was author-
ized, as a five year project, August 1979, for $9,000,000. The
Project Grant Agreement with the Government of Malawi (GOM) was
signed in August 1979. Technical assistance (TA) to the project
was provided by the University of Florida (UF). The TA contract,
signed in May 1980, provided for seven TA positions. Members of
the TA team began arriving in county in June 1980 and the full
team was on board by September 1981. The TA positions included
research coordinator and chief of party, agronomist, horticul-
turalist, plant breeder, agricultural economist, farming systems
analyst, and animal scientist.

MARP was evaluated two times: a mid-tern evaluation in 1981
(Thorne, 1981) when most of the TA team members were arriving at
post; and a second evaluation in 1983 (Baker, et al., 1983).
This case study is based primarily, unless noted otherwise, on
the second evaluation (Baker, et al., 1983).

MARP was not a farming systems project per se. Indeed, the
second evaluation team reported: "One of the things which this
evaluation team became aware of at the beginning of the evalua-
tion is that this project is a Malawi Agricultural Research
Project, not a Malawi F3R project" (Baker, et al., 1983:36).
Thus, the team evaluated the project's farming systems analysis
(FSA) section "as any other section of the project would be
evaluated and not as if it were the dominant thrust or philosophy
of the entire project" (Baker, et al., 198.:36). Nevertheless,
this case study does focus primarily on the experience of the
project's FSA section, noting information about other project
components where relevant.

Concept What was the basic technical idea underlying the

MARP was conceived as the first phase of a longer term
effort to improve agricultural research in Malawi; and in the
short to medium term as a support element to the National Rural
Development Program (NRDP), a nationwide integrated rural
development activity. The goal of MARP, as stated in the PP, was
"to increase agricultural production and real incomes of small-
holders." Malawian smallholders comprises 85% of the country's
population. The purpose of MARP was to strengthen the capability
of the Department of Agricultural Research (DAR) within the
Ministry of Agriculture (MOA) "to provide socially acceptable and
economically sound research for smallholder needs in satisfactory
quality and quantity and in a form usable by the extension

The overall strategy to achieve this purpose and goal was to
address the critical shortage of trained research scientists by
funding graduate studies, and to alleviate the lack of physical
facilities and equipment through a major construction and
procurement effort.

Design How was this basic technical idea translated into a

MARP was designed using AID's "collaborative assistance"
mode which permits a Title XII university to be competitively
selected to participate in the final design of the project and
promptly commence its implementation when AID approves funding
for the project. However, delays in executing the TA contract
resulted in the UF losing several of its intended TA team mem-
bers. This, in the view of the evaluation team, had a

negative impact on project implementation. It appears that
the project designers wrote the project job descriptions
with fairly specific individuals in mind but could only
field part of that team by the time the contract for tech-
nical assistance was finally signed. As a result, the
particular skills mix of the team actually fielded has not
been as comprehensive as what seems to have been intended at
the time the project was designed. For example, the PP
called for a crops agronomist who was expected to work on a
variety of food and forage crops. The individual fielded
was primarily a forage crops agronomist and as a result,
food crop agrcnomic research was somewhat neglected during
the early years of the project (Baker, et al., 1983:20).

The project design did not fully anticipate the effect that
sending half of the DAR's best staff on long-term training would
have on ongoing research programs. The TA team found

it extremely difficult to fill in for the 23 Malawian
researchers...in training. Most of the UF researchers have
become heavily involved in the management of their respec-
tive research sections as well as certain management func-
tions associated with the project (managing participant
training, commodity procurement, reporting, etc.) In addi-
tion, they are often called upon to perform...tasks for the
DAR...which were not anticipated at the time of the design
of the project (Baker, et al., 1983:20).

In the face of this workload, some of the functions that the
design ha'. planned to be performed by the TA team received less
aitentioi than intended. As a result, the TA team tended

to place less priority on following closely the various
research programs in progress and working closely with
research staff and extension personnel to develop research
recommendations and conduct...on-farm trials. Rather, they
...tended to give higher priority to research program plan-
ning and management, developing longer term research
priorities (Baker, et al., 1983:20).

Further, the PP had emphasized the importance of a multi-
disciplinary team approach. The evaluation team found that the
TA skill mix contained in the PP would lead one

to believe that most of the expatriate researchers were to
advise on several crop programs. For example, the plant
breeder would assist in plant breeding for maize, wheat,
sorghum, and other grain legumes.... The agronomist would
also assist various commodity programs. Furthermore, the
-ndividual expatriate researchers would work as...[a]
team[,] each making some contribution to improving the
technical quality or relevance of the various research
programs to the smallholder farmer (Baker, et al., 1983:21).

However, each TA team member tended to specialize in particular
crops. This, the evaluation pointed out, deemphasized the
importance of a multidisciplinary approach. Also, TA personnel
tended to work independently and not in accordance with any
defined work plan. Indeed, the evaluation reported that TA team
members had only recently begun to write formal work plans. The
evaluation also noted "that the assumption of the existence of
national research priorities was a major omission in the project
design" (Baker, et al., 1983:67).

These factors suggest that a major problem in MARP was that
the project's design was overly ambitious but poorly defined. On
the one hand, the Outputs section of the PP (p. 15) stated that

The Title XII team will have introduced new research
programs in farming systems analysis, production economics
and smallholder appropriate technology and will have
improved the capability of the [DAR] in research coordina-
tion for the selection and implementation of research
benefitting smallholders and in research/extension liaison.

Yet the project continually found itself trying to do too
much with too little. The following examples are illustrative:

Regarding the quality of TA personnel, "more experience
in agricultural research in developing countries,
particularly in Africa, would have been beneficial"
(Baker, et al., 1983:21).

Regarding short-term TA, "most...consultants were in-
country for a duration of two weeks or less and that
few...consultants...made repeated trips to Malawi. The
team would have preferred to see fewer...consultants,
longer durations of the consultancies and key...tech-
nical expertise returning periodically to assist
the...the DAR (Baker, et al., 1983:21).

Regarding the DAR's maize research program that was
charged with identifying smallholder constraints to
maize production and conducting low cost input studies,
"these are not adequately covered in the current
research program due to lack of resources and person-
nel" (Baker, et al., 1983:47).

S Regarding the pastures research program, "there seems
very little effort to identify problems and constraints
of the smallholder dairy" (Baker, et al., 1983:49).

Regarding interaction between the TA team and Malawian
project staff, the TA team members "almost unanimously
...regret...that they are not providing more on-the-job
training and supervision for junior research staff"
(Baker, et al., 1983:20).

Implementation How was the project managed by the host-country
implementing agency, the TA team, and USAID?

Overall implementation of MARP was impeded by delays in
procurement, construction, contractual arrangements for provision
of TA, and identifying candidates for graduate training. There
were also delays in fielding the TA team and "a failure to pro-
vide qualified individuals for the positions designated in the
Project Paper" (p. 8 of PES for Baker, et al., 1983). Project
implementation also suffered from inadequate support of research
by the GOM, specifically in the assignment of research and
support personnel to the project.

The evaluation found the PP's description of the anticipated
project management arrangements to be "extremely general." Also,
the evaluation found that the system that had evolved since the
project's inception was "diffuse and...rather vague" (Baker, et
al., 1983:60). Indeed, the evaluation concluded that the
"present management arrangement does not provide for clear and
coordinated project direction" (Baker, et al., 1983:64). The
project's financial management arrangements were also found to be
"diffused." While the designed and established approval
procedures were apparently adequate to protect against misuse of
project funds,

the financial management system for the project does not
permit the UF, USAID and the DAR to jointly and periodically
meet and plan future actions and budgetary implications of
desired courses of action.... ...because of the number of
actors involved..., it is impossible to assess the project's
financial position with any degree of accuracy. With
little project direction and limited opportunities for joint
planning and budgeting, the project lacks flexibility to
make adjustments. Probably most critical, there
appears to be no single management unit or individual who
has responsibility for, control over, and a thorough know-
ledge of the project's resources and the totality of the
...activities of the project (Baker, et al., 1983:61, 64).

Compounding the management problem, as the evaluation team
also found, was the fact that the Chief of Party (COP) was "re-
sponsible for a myriad of administrative tasks that are seriously
distracting him from the role he was expected to play" (Baker, et
al., 1983:65). The evaluation recommended that the project hire
an administrative assistant to free the COP to concentrate on
making a more substantive input to the project, and that such
administrative assistance support should be included in the
design of any follow-on activity.

But the more interesting and relevant area in considering
problems affecting the project's implementation was that of the
project's Farming Systems Analysis (FSA) section. The FSA
section of the DAR was one of two new sections of the DAR
instituted under the project, the other being the Agricultural
Economics section. One of the TA team's technical assistance
positions for the initial two years of the project was a farming
systems analyst to establish and serve as acting head the FSA
section. UF provided an anthropologist to fill this position.
The acting head of the FSA section had outlined nine objectives
in the FSA section work plan. These were:

1. Establish a farming systems research (FSR) program.

Successful on-farm research is based on commitment to farm-
level field work. The opportunity for the FSA section head to be
assured that all five of his Malawian FSA staff were prepared to
make this commitment was precluded when higher level MOA officers
appointed three Malawians without the FSA section head's input.

After a FSA section office had been s-t up, the FSA section
initiated FSR according to a four-step model: (1) description
and diagnosis to identify targets for adaptive research; (2)
design of alternative technologies to modify the existing system
rather than dramatically change it; (3) testing proposed alter-
natives to see how they perform; and (4) delivering to the exten-
sion service any innovation proving to be "a good and acceptable
modification of the system" (cited in Baker, et al., 1983:37).

The FSA section completed a diagnostic survey in each of
four Agricultural Development Districts (ADD) in April and June,
1981. DAR agronomists and ADD r.aff participated in the survey.
Also, as a means of establishirJ, a link with the International
Agricultural Research Centers,'the project invited Michael
Collinson (CIMMYT/Africa outreach economist) to Malawi to
participate in supervising the survey work.

Once each district's survey had been completed, a meeting
was held in the district to evaluate the survey's results and to
agree on treatments for on-farm trials for 1981-82. Trials were
established in three districts; shortages of fuel and staff
precluded working in the fourth district. To illustrate the
nature of the trials, consider the case of the Phalombe area,
where trials were designed for a maize/cowpea/sunflower cropping
system. Thete were four treatments: (1) local maize without
fertilizer, (2) local maize with fertilizer, (3) improved maize
without fertilizer, and (4) improved maize with fertilizer. A
total of 16 trials were established; however, only 14 trials were
harvested, the other two being eliminated because of rainfall.
All trials were farmer-managed with orientation and observation
by research and extension staff.

At the end of the 1981-82 harvest, the FSA section had
progressed from step 1 through step 3 of the section's FSR
methodology. Based one season's data, the FSA section concluded
that variety makes little difference without fertilizer, and that
both varieties respond to fertilizer. Yet, during the 1982-83
growing season, the FSA section did not establish any follow up
trials. The answer to the question that immediately arises ("Why
not?") is considered further below.

2. Ensure Research Personnel at the M.Sc. and Ph.D. Levels

There were delays both in GOM clearance to create the first
position in the FSA section and in hiring the first Malawian for
that position. This resulted in a year's delay before the first
participant left in December 1982 for training in the U.S., while
the second participant had still not been identified. These
delays not only made it impossible for either participant to
overlap with the current FSA section head upon return from study
in 1985 (or 1986) but raised the question of who would head up
the FSA section between the departure of the current expatriate
FSA section head and the return and field orientation of the FSA
section participants studying in the U.S.

3. Improve Support Staff Capability

The FSA section head engaged in various activities to sup-
port staff orientation and training. These were: (a) continuous
in-service training of FSA staff; (b) cosponsoring a multiple-
cropping research conference; (c) sponsoring a socio-economic
research conference; (d) conducting a seminar on design and
analysis of on-farm trials; (e) arranging for six DAR and ADD
staff to attend CIMMYT-sponsored FSR courses in Kenya; and (f)
arranging for three DAR staff to attend a CIMMYT-sponsored FSR
course in Zimbabwe.

4. Establish a Research Coordination System

The FSA section worked at the national level on coordination
between the MOA's research and evaluation units, and explained
the FSR approach at ADD evaluation staff meetings. The section's
survey work was also aimed at refining the survey as a tool for
eliciting information on smallholders' practices, problems, and
conditions. Also, the FSA section head served on the Agricul-
tural Research Committee.

5. Establish Baseline Data and Field Trials

Surveys had been conducted in several ADDs and the data were
being analyzed with the aim of gaining insight into smallholder
practices. While on-farm adaptive trials had been conducted in
villages in three ADDs, the trials covered less than half of the
Extension Project Areas (EPAs). The evaluation team pointed out
that it would be impossible to realize the PP goal of 110-130
EPAs by the project's end. Further the team suggested that the
FSA section and DAR establish a more realistic goal (e.g., local
research teams in two or three pilot ADDS for on-farm trials).

6. Strengthen Research Programs Relevant to Smallholders

The FSA section was collaborating with both maize breeding
and agricultural economics in including local maize in fertilizer
trials. However, the identification of local maize as the pre-
dominant variety in the majority of farm cropping systems and the
key treatment in on-farm trials had to "a basic misunderstanding
about the role of FSR in Malawi agricultural research" (Baker, et
al., 1983:39). This point is considered below.

7. Establish a Research/Extension Liaison System

The FSA section worked both in the field and at the policy
level to build a strong link between research and extension. ADD
staff participated in design and execution of diagnostic surveys
and design, monitoring, and analysis of on-farm trials. The
section also met with ADD and National Rural Development
Programme (NRDP) staff to explain adaptive research and FSR.

8. Survey Farmer Adoption of DAR Recommendations

The foundation for achieving this objective had been laid
with a pilot survey conducted during August and September, 1982,
among smallholders in Lilongwe ADD. The data from this survey
were being analyzed at the time of the evaluation.

9. Have Publications Developed by TA Team and Malawi Staff

The FSA section had produced a number of publications on
FSR-related issues in Malawi. However, because only one of these
publications had been authored by a Malawian, the evaluation team
felt that greater effort needed to be placed on developing and
producing co-authored or section publications, and on selecting
Malawians for the FSA section (or in-service training for the
existing staff) who can contribute to the publication of FSA
reports. Otherwise, the evaluation cautioned, MARP could be
criticized that the FSA section is producing publications which
are "the sole output (and, implicitly, contain only the ideas) of
the FSA section head" (Baker, et al., 1982:40).

Despite the FSA section's progress in achieving these
objectives, the evaluation team found that:

(1) In comparison to "standard" FSR methodology and proce-
dures, the...FSR approach, as exhibited by the FSA section,
was extremely well carried out up to the time of harvest of
the 1981-82 cropping season; while, on the other hand,

(2) Some mainstream DAR researchers, some of the UF tech-
nical assistance, and some upper-level MOA officials are

(a) not in sympathy with the FSA approach to FSR, or
(b) of the opinion that the FSA approach to FSR is not
research (Baker, et al., 1983:40).

Also, the evaluation concluded that the FSA section was "rela-
tively isolated both from other UF technical assistance and from
the mainstream DAR researchers and higher-level MOA officials"
(Baker, et al., 1983:40).

While the diagnostic phase of the FSA section's approach to
FSR had been helpful in assessing farmers' needs,

problems arose when the FSA section headed the subsequent
trial design phase. Some DAR officials and...UF research
staff believe the trial design plase should have been a
joint exercise, where agronomy takes the lead with [the] FSA
section assisting. ...the original job description for the
Farming System Analyst position states that one of the...4
duties of the FSA is...: "Assist the Research Coordinator
and research officers in...selection and evaluation of

smallholder research projects to ensure incorporationn of
local smallholder farming systems data into research plan-
ning." (PP, ANNEX A, p. 11) (Baker, et al., 1983:40).

Apparently, as the evaluati. team found, the "agricultural
scientists...did not like the idea of a social scientist design-
ing, implementing, harvesting and analyzing agronomic on-farm
trials" (Baker, et al., 1983:41). Also, the evaluation team
concluded that there had been

very little UF team interaction between the diagnostic
survey stage and the farm trial design phase. .
Instead of assisting the rest of the team in design of
trials, the FSA section head had employed a more direct
approach.... There were few alternatives, however, as
the...DAR staff and UF technical assistance had no formal
mandate to work in an inter-disciplinary mode; thus the FSA
section was forced to rely on recruiting voluntary
assistance. The FSA section head was forced into a
choice between proceeding using whatever manpower and
agronomic advice was available and willing to participate in
1981-82, or waiting another season to initiate on-farm
trials. As the FSA technical assistance was only funded for
the first two years of the five year project, delaying the
trials would have meant th-t the objectives of the FSA
workplan would have fallen short of achievement (Baker, et
al., 1983:41).

The evaluation team noted that identifying the importance of
local maize was a scientifically important (and acceptable)
outcome of the diagnostic phase of the FSR methodology being
implemented by the FSA section. However, as the evaluation team
also noted, basing the first round of on-farm trials on local
maize varieties seemed to go counter to the GOM policy of quickly
increasing per hectare yields in smallholder fields. Further,
from an agronomic point of view,

it was assumed by the...DAR researchers, based on years of
experience that the improved varieties...are genetically
superior to the local varieties in their ability to yield
well under high doses of nitrogen fertilizer and good
management. What the...[FSA section's on-farm trials]
measured, however, during only one growing season, was the
response of an improved versus a local variety using DAR-
recommended levels of fertilizer in the farmer's cropping
system (which, in the case of the...[Phalombe area],
included both sunflower and cowpea) under his (or her) own
management. Thus, the improved variety was subjected to two
conditions for which it was not specifically bred (Baker, et
al., 1983:41).

Analyzing this situation, the evaluation team noted that
very few agronomists/breeders would place as much emphasis on one

year's data as did the FSA section. However, as the evaluation
also noted, the MOA/DAR and the TA team misinterpreted the
implications of the on-farm trials (OFTs).

What the results indicate is not that there are no dif-
ferences between varieties, but that in the particular ADD
farmer system and under the unique farmer management during
the 1981-82 season, there were no statistically significant
differences between varieties. Further, the importance of
considering alternative sets of recommendations for differ-
ent levels of farmer resources was pointed out. The...MOA/
DAR-UF research team should have used this information as...
positive feedback from the farm level to refine on-station
research priorities to address the' issues raised by the
OFTs. They should not have reacted negatively to the
results of the OFTs. ...the way in which the FSA section
reported...results should have been positive "we believe
more on-station work could be done on improved varieties
grown...with other crops, and perhaps more thought should be
given to higher fertilizer recommendations for improved
varieties than for local ones", rather than "there are no
differences between local maize and the improved variety."

Once the...actors began to go separate ways, subsequent
contacts became less frequent and opinions about..."others"
solidified and became self-reinforcing. The FSA section
viewed the OFTs as ultra-high priority and dedicated much
time to them; other UF scientists had their own programs and
priorities, and little inclination to visit trials into
which they had little or no input; some MOA officials con-
tinued to lament...that the FSA section was taking the lead
in agrcnomic farm trials (Baker, et al., 1983:41-42).

Between the 1981-82 and 1982-83 seasons, the Chief Agricul-
ture Research Officer (CARO) and the UF Chief of Party decided to
put a temporary stop on the FSA section's OFTs until such time as
agronomy could be "officially" involved in the effort. Thus, at
the time the evaluation was conducted, there was:

no second round cf OFTs during the 1982-83 cropping season.
The FSR methodology has come to a halt between the OFT "ob-
servation trial" (season one) phase and the OFT "verifica-
tion trial" (season two, etc.) phase. ...these issues
should have been resolved so there could have been follow-up
trials this year with the farmers in...ADDs where trials
were in place last year. However, if such a pause in the
farm trial phase is both temporary and positive, in that it
leads to a reinstitution of communication between the FSA
section and the...UF technical advisors, and between the UF
team and the acting CARO vis-a-vis FSR methodology and
implemertaticn plans, then...there is...reason to be
optimistic that the farm trial phase will resume in the next
croppng eason. All sides need to stop accusing the

FSA section of being too autocratic, and the FSA section
needs to see itself as a participant in the adaptive
research trials...(Baker, et al., 1983:42-43).

One "lesson" that may be "learned" from this experience is
that while MARP included a TA team position for a farming system
analyst, the Malawians and the TA team did not share a common
vision as to the objectives and methodology of an FSR/E approach.
Indeed, the evaluation team concluded that "one of the great
obstacles" in the way of realizing the potential of introducing a
FSR/E approach at the ADD level, was "the divergent views held as
to how Malawi should approach on-farm research" (Baker, et al.,

Indicative of this problem was the project's difficulty in
defining the role of agricultural economics vis-a-vis the FSA
section. The evaluation noted that neither the Outputs section
nor the Logical Framework of the PP "list anything specific for
the economics section" (Baker, et al., 1983:30). Further, the
duties for the economist listed in the PP's Long Term Technical
Assistance -ob Descriptions differed from the Work Plan of the
economist provided by UF. While this work plan listed six
objectives, there was little or no emphasis on supporting the
development of the FSA section. Reviewing the program of the
agricultural economics section, the evaluation found that the
section's research interest concentrated more toward

addressing macroeconomic policy issues than...economic
constraints faced by Malawi smallholder[s]. There-
fore, the evaluation team recommends that the section spend
more time in (a) farm-level trial design and analysis of
trial results, (b) determining whether or not improved
treatments benefit the farmer more than they cost him, and
(c) collaborating with the adaptive research effort via the
FSA section. If this recommendation is followed, the...work
plan of the...agricultural economist will begin to look more
like the original job description for this position outlined
in the Project Paper (Baker, et al., 1983:34-35).

The evaluation team proposed a series of recommendations
aimed at getting project participants on the same wave length
vis-a-vis FSR/E. These recommendations included steps for
improving communication as well as for organizing and coordi-
nating the manpower, technical, and financial resources available
to the project. For example, the evaluation recommended that a
recently arrived UF agronomist be assigned, oriented, trained,
and counted on as the FSA section agronomic advisor to FSR trial
design, implemnetation, field observations, harvest, and
agronomic analysis.

Although noting problems in the project's implementation,
the evaluation team basically reaffirmed the validity of the FSR
model that the FSA section had been using, stressing that Malawi
be encouraged to use the following broad stages:

(a) diagnostics: already done extremely well;

(c) observation trials (On Farm Trials): again, already
done in some ADDs, but with too much emphasis placed on one
season's results;

(c) verification trials: a stage which may vary from one to
several years, depending on the amount of fine-tuning
involved in each set of trials per homogeneous ecological

(d) demonstration trials: these trials, 100% under the
supervision of extension...(Baker, et al., 1983:43).

One may question the evaluation team's definition of demon-
stration trials which the team further defined as follows: "so
named only when research is convinced that their recommended
technological innovations are superior to and less risky than the
current farm practice, as well as being non-disruptive of the
social setting of the farm family" (Baker, et al., 1983:43).
This definition is problematic because it assumes that the
researcher, rather than the farmer, is the final arbiter of
whether a given technology is sufficiently improved to merit
being recommended to other farmers.

Evaluation How was the project's performance measured or

Three "critical aspects" were identified as relevant to
assessing the extent to which the project was making progress
toward achievement of the project purpose. These aspects were:

1. Are the research programs being implemented tech-
nically sound, relevant to smallholders' needs and
conducted in a coordinated manner?

2. Is a research management system in place which
efficiently allocates financial and human resources in
accordance with project priorities?

3. Is there an adequate information dissemination system
which provides research results to the appropriate
clients of the research organization? (Baker, et al.,

1. Are the research programs being implemented technically
sound, relevant to smallholders' needs and conducted in a
coordinated manner?

The evaluation concluded that "some of the research being
conducted may not adequately take into account the actual condi-
tions and limitations faced by smallholders" (Baker, et al.,
1983:56). This, the evaluation team advised, could be corrected
by a closer coordination between the FSA section, the Agricul-
tural Economics section, and the commodity programs; and by
greater reliance on on-farm trials to increase contact between
researchers and farmers and between researchers and extension
staff. Further, the evaluation recommended that the FSA and
Agricultural Economics sections "be given a larger role...in
establishing commodity research priorities, designing research
activities and in conducting adaptive trials or tests" (Baker, et
al., 1983:57).

2. Is a research management system in place which efficiently
allocates financial and human resources in accordance with
project priorities?

The research resource allocation system in place at the time
the project was designed, and which was still in effect at the
time of the evaluation, allocated resources according to past
trends in expenditures (primarily to each research station and
not to each research program) and the availability of external
resources (primarily donor financing). While an improved system
of research management was not defined in the PP, the TA team was
assisting the DAR to plan for a management system that would
allocate funds and staff ba-ed on established research priorities
and actual progress toward research objectives.

The TA team members serving as acting heads of the various
research sections were tasked to develop research priorities for
their respective sections. While section-specific research
priorities had been established, the evaluation found

that such priorities have yet to reflect (a) the needs of
the smallholders and (b) national development priorities.
It was expected that multidisciplinary interaction through
the Farming Systems Analysis Section would determine the
research needs of smallholders, and that interaction and
coordination with the extension system would also identify
priorities...of small farmers and regional or area-specific
research needs (Baker, et al., 1983:59).

However, the evaluation was encouraged by a proposal to create
adaptive research teams at the ADD level, and felt that this
could provide "a suitable mechanism for factoring the needs of
small farmers into...research priorities" as well as strengthen
linkages between research and extension (Baker, et al., 1983:59).

Thus, while the research management system in place at the
time of the evaluation was "only a ninor improvement over what
was in effect at the time the project was designed," the evalua-
tion concluded that there was "a much more generalized awareness"
of that system's problems (Baker, et al., 1983:59). However, the
evaluation was quick to point out a major negative factor,
namely, "that the duration of the present project is too short to
ensuree the proposed changes in the research management system
will be implemented" (Baker, et al., 1983:59).

3. Is there an adequate information dissemination system which
provides research results to the appropriate clients of the
research organization?

The evaluation concluded that one of the project's major
output targets was not'being met, namely, that the results of the
research being conducted by the project were not getting to the
extension service and in turn to the smallholder. Further, the
evaluation concluded that a major effort needed to be made to
develop links between research and extension.

Further, the evaluation noted significant disparities of
expectations among USAID/Malawi, the GOM and the TA team.

Compounding the problem of uncertainty of specific responsi-
bilities, there appears to be some disagreement between
USAID, the DAR (GOM) and UF about whether the primary
objective of the project is to (a) provide training,
equipment and physical facilities, (b) build a research
institution or (c) produce research results. The lack of
agreement concerning the fundamental objective and the lack
of a clear assignment of responsibilities have give rise to
occasional misunderstandings which have hindered the
implementation of the project. More importantly, the two
factors have left the project without effective leadership
and a clear sense of direction (Baker, et al., 1983:60).

In the PES, USAID/Malawi recommended that benchmarks need to be
more closely and carefully emphasized during project design, and
that the follow-on project to MARP "should not be of such an
ambitious nature" (PES, p. 10).

Institutionalization How did the project provide for the
implementing agency to develop a sustainable capability to
continue to perform the types of activities supported by the

The PP for MARP provided for a heavy input of training: 33
participants to receive long-term training to the M.S. or Ph.D,
level in the U.S.; in-country in-service training for all DAR
professional staff; and participation by selected DAR staff in
professional meetings, conferences, seminars, and workshops both
within and outside of Malawi. At the time of the second
evaluation, the number of Ph.D. candidates had been increased
from eight in the PP to an actual total of 12 trainees. As the
evaluation noted, the greater length of a doctoral program and
the increase in the number of'Ph.D. candidates would make it

more difficult for returning participants to benefit from
the technical assistance provided under the project. Most
trainees will return after the PACD. ...each trainee
will require more time to complete his or her program and
will be less likely to overlap with the UF technical
assistance team, thus jeopardizing the output of an estab-
lished and sustained program of research relevant to the
smallholder.... It will particularly affect the
ability of the newly-returned researchers to benefit from
the guidance of the technical assistance team and the
continuity the letter have provided while participants were
absent (Baker, et al., 1983:12, 15, 17).

According to the PP, "all of the training decisions were
based on the specific needs of research.... The training program
represents the summation of specific project needs for better
trained professional researchers." However, the evaluation found
that many of the disciplines originally identified in the PP were
changed during implementation of the participant training pro-
gram. For example, the PP provided for five M.S. candidates,
three in agricultural economics and two in applied anthropology
for a total of 96 person months, whereas the actual training
program was providing only three M.S. candidates, two in agricul-
tural economics and one in applied anthropology, for a total of
72 person months. Given the changes in the disciplines in which
the participants were being trained, as well as the problems
encountered in implementing the project's FSA section, it is not
clear that returning participants would necessarily be assigned
to work on the continued development of FSR/E. In this regard,
the Regional Inspector General's audit of the project in November
1982 found that the TA component "needed to be synchronized with
the long-term training component to insure a reasonable overlap
between returning trainees and AID-funded expatriate

On the other hand, the project had offered one or two in-
service training courses on FSR and the design and analysis of
on-farm trials. These courses were generally offered to 25-40
participants, which represents one fourth to one third of the DAR
professional officers. Further, project participants in the
agricultural economics and FSA sections received on-the-job
training in the use of microcomputers for data storage and
analysis and word processing.

Another dimension of institutionalizing FSR is the govern-
ment's ability to cover associated recurrent costs. While the
evaluation concluded that the GOM appeared to be meeting its
commitment for covering recurrent costs of the project, "growth
and expansion of research activities commenced by the project"
might be slow, "especially for those activities which entail
considerable transportation costs (i.e. farm surveys and on-farm
trials)" (Baker, et al., 1983:25).

Perhaps the greatest hope for eventual institutionalization
of FSR/E in Malawi was the project's apparent impact on increas-
ing "the sensitivity of research managers to the need and
capabilities of smallholders" (Baker, et al., 1983:63). Further,
with the establishment of the FSA and Agricultural Economics
sections, the evaluation concluded "that the basic mechanisms
exist to provide internal interaction and coordination between
research sections" (Baker, et al., 1983:66). However, the
evaluation recommended

that the UF team meet and come to an agreement on their
general FSR approach. ...that the UF team members, in
consultation with their principal Malawian research staff,
each decide how their own individual research programs can
contribute towards adaptive on-farm research for the
duration of this project. ...that...the UF team meet with
the CA.O and CAO to reach agreement on the role of, and
direction of, adaptive on-farm research conducted at the ADD
l-vel for at least the duration of the project and hopefully
beyond (Baker, et al., 1983:66-67).

In the last analysis, the evaluation recommended that
USAID/Malawi and the GOM consider a follow-on to MARP, with the
follow-on focusing on "developing the adaptive research units and
extension units that will form critical components of a system
for the dissemination of information and services to smallholders
as well as continuing to strengthen appropriate component
research" (Baker, et al., 1983:68). Further, the evaluation
recommended that the research activities initiated by the AID-
supported Women in Agricultural Development Project be integrated
into and supported by MARP and its follow-on project.

The need for long-term support for the development of
agricultural research and extension was also voiced by the first

The review team recognizes the need for the technical
assistance project to continue beyond the present contract
period. The research objectives cannot be fully met in
1984. The NRDP goal is expected to be realized in 15 to 20
years. It is important that technical assistance be
provided on a continuing basis well beyond 1984 if the
objectives of this project are to be realized fully (Thorne,


1983 Project Evaluation Summary of Evaluation Report for
Malawi Agricultural Research Project (612-0202).

Baker, Murl, Joan Atherton, Daniel Galt, Curtis Nissly, Frank
Mwambaghi, and Henry Mwandemere
1983 Evaluation Report for Malawi Agricultural Research
Project (612-0202). (XD-AAM-810-A)

Thorne, Marlowe D.
1981 Report of First Internal Evaluation of Malawi
Agricultural Research Project (612-0202), November-
December 1981. (PD-AAN-267)

Annex A. Project Description Sheet.

This Project Description Sheet lists the core, operational,
and generic constraints identified in this project, per the
following codes: core (C), operational (0), and generic (G). A
positive (+) sign after a constraint indicates that the project
was effectively coping with the identified constraint.3

Core Constraints (C)

C.I Farmer Orientation
C.2 Farmer Participation
C.3 Locational Specificity of Technical and Human Factors
C.4 Problem-Solving Approach
C.5 Systems Orientation
C.6 Interdisciplinary Approach
C.7 Complementarity with Commodity and Discipline Research
C.8 Technology Testing in On-Farm Trials
C.9 Feedback to Shape:
a. Agricultural Research Priorities
b. Agricultural Policies

Operational Constraints (0)

0.1 Stakeholder Understanding of FSR/E
0.2 Agricultural Research Policy/Strategy Defining Role of FSR/E
0.3 Long-Term Commitment of Resources
0.4 Existing Research Capability and Shelf Technology
0.5 Consensus on FSR/E Methodology
0.6 Capability to Process Farming Systems Data
0.7 Consensus on Criteria for Evaluating FSR/E
0.8 Links with Extension
0.9 Links with Agri-Support Services
0.10 Links with Farmer Organizations

Generic Constraints (G)

G.1 Project Management Structure
G.2 Government Funding to Meet Recurrent Costs
G.3 Staffing with Trained Manpower
G.4 Management of Training
G.5 Management of Technical Assistance
G.6 Factors Beyond a Project's Control

An analysis ; of these constraints in 12 FSR/E projects appears
in A Review of A.T.D. Experience with Farming Systems Research and
Extension Projects, A.I.D. Evaluation Special Study (forthcoming),
available from A.I.D.'s Document and Information Handling Facility
(pai instructions on last page of this report).

Malawi/ARP Agricultural Research Project (612-0202)

Initial Authorization: 1979 (for 5 years)

Goal: "to increase agricultural production and real incomes of

Purpose: To strengthen the capability of the Department of
Agricultural Researcn (DAR) within the Ministry of Agriculture
"to provide socially acceptable and economically sound research
for smallholder needs in satisfactory quality and quantity and in
a form usable by the extension services."

Outputs: Not a FSR project per se but did provide support for
two new DAR sections: Farming Systems Analysis (FSA) and
Agricultural Economics. Outputs included strengthening of
quality and quantity of research programs in crop, livestock, and
technical areas relevant to smallholders; and field trials
completed by TA team and counterpart staff, and technology
packages developed.

Implementing Agency: Department of Agricultural Research (DAR),
Ministry of Agriculture.

TA Contractor: University of Florida.

Evaluations: Two -- a mid-term evaluation in 1981 (Thorne, 1981)
when most of the TA team members were arriving at post; and a
second evaluation in 1983 (Baker, et al., 1983).

Constraints: C.3, C.4, C.6, C.9, 0.1, 0.2, 0.3, 0.5, 0.7, 0.8,
G.1, G.2, G.3, G.4, G.5.


This CDIE Working Paper is a case study that was prepared for a
cross-cutting analysis of A.I.D. FSR/E projects, A Review of A.I.D.
Experience with Farming Systems Research and Extension Projects,
A.I.D. Evaluation Special Study (forthcoming). A total of 13 case
studies were prepared. These may be ordered from the A.I.D.
Document and Information Handling Facility, 7222 47th Street, Suite
100, Chevy Chase, MD 20815. Telephone: (301) 951-9647. Please
request CDIE Working Paper No. 112, followed by the required Case
Study No. and PN number.

Botswana Agricultural Technology Improvement Project (633-0221),
CDIE Working Paper No. 112--Case Study No. 1. (2W-ABC-073)

Gambia Mixed Farming and Resource Management Project (635-0203),
CDIE Working Paper No. 112--Case Study No. 2. (PN-ABC-074)

Lesotho Farming Systems Research Project (632-0065), CDIE Working
Paper No. 112--Case Study No. 3. (PN-ABC-075)

Malawi Agricultural Research Project (612-0202), CDIE Working Paper
No. 112--Case Study No. 4. (PN-ABC-076)

Senegal Agricultural Research and Planning Project (685-0223), CDIE
Working Paper No. 112--Case Study No. 5. (PN-ABC-077)

Tanzania Farming Systems Research Project (621-0156), CDIE Working
Paper No. 112--Case Study No. 6. (PN-ABC-078)

Zambia Agricultural Development Research & Extension Project (611-
0201), CDIE Working Paper No. 112--Case Study No. 7. (PN-ABC-079)

Nepal Agricultural Research and Production Project (367-0149), CDIE
Working Paper No. 112--Case Study No. 8. (PN-ABC-080).

Philippines Farming Systems Development Project-Eastern Visayas
(492-0356), CDIE Working Paper No. 112--Case Study No. 9. (PN-ABC-

Guatemala Food Productivity and Nutritional Improvement Project
(520-0232), CDIE Working Paper No. 112--Case Study No. 10. (PN-ABC-

Honduras Agricultural Research Project (522-0139), CDIE Working
Paper No. 112--Case Study No. 11. (PN-ABC-083)

ROCAP Small Farm Production Systems Project (596-0083), CDIE Working
Paper No. 112--Case Study No. 12. (PN-ABC-084)

Vignettes of Core, Operational, and Generic Constraints in 12
A.I.D.-Funded Farming Systems Research and Extension Projects, CDIE
Working Paper No. 112--Case Study No. 13. (PN-ABC-127)

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs