• TABLE OF CONTENTS
HIDE
 Front Cover
 Title Page
 Table of Contents
 Integrating gender issues into...
 The effect of storage loss rates...
 Gender issues in farming, part...
 Informal agricultural communication...
 Farming systems research and the...
 Farm-level performance of improved...
 The use of computers in field research,...
 Agricultural development in the...
 Trial designs and logistics for...
 Rural women in irrigated and rain-fed...
 Instructions to authors






Group Title: Journal for farming systems research-extension.
Title: Journal of farming systems research-extension
ALL VOLUMES CITATION PAGE IMAGE ZOOMABLE PAGE TEXT
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00071921/00007
 Material Information
Title: Journal of farming systems research-extension
Alternate Title: Journal for farming systems research-extension
Abbreviated Title: J. farming syst. res.-ext.
Physical Description: v. : ill. ; 23 cm.
Language: English
Creator: Association of Farming Systems Research-Extension
Publisher: Association of Farming Systems Research-Extension
Place of Publication: Tucson Ariz. USA
Publication Date: 1990-
 Subjects
Subject: Agricultural systems -- Periodicals -- Developing countries   ( lcsh )
Agricultural extension work -- Research -- Periodicals   ( lcsh )
Sustainable agriculture -- Periodicals -- Developing countries   ( lcsh )
Genre: periodical   ( marcgt )
 Notes
Dates or Sequential Designation: Vol. 1, no. 1-
General Note: Title varies slightly.
General Note: Title from cover.
General Note: Latest issue consulted: Vol. 1, no. 2, published in 1990.
Funding: Electronic resources created as part of a prototype UF Institutional Repository and Faculty Papers project by the University of Florida.
 Record Information
Bibliographic ID: UF00071921
Volume ID: VID00007
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 22044949
lccn - sn 90001812
issn - 1051-6786

Table of Contents
    Front Cover
        Front Cover
    Title Page
        Page i
        Page ii
    Table of Contents
        Page iii
    Integrating gender issues into farmer-participatory research: The case of vegetable IPM technology generation in Calamba, Laguna, Philippines, by M. M. Hoque and C. B. Adalla
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
    The effect of storage loss rates on the valuation of maize stored traditionally by farmers and removed periodically for food, feed, or sale in Cameroon, by Dermot McHugh
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
    Gender issues in farming, part II: Selected case studies in development and extension of farm and other equipment to women in northern Nigeria, by R. N. Kaul
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
    Informal agricultural communication patterns in a remote area of Bangladesh, S. M. A. Hossain, B. R. Crouch, and Shankariah Chamala
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
    Farming systems research and the rural poor: A political economy approach, by S. D. Biggs and John Farrington
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
    Farm-level performance of improved cassava varieties in the humid forest zone of Nigeria, by F. I. Nweke, H. C. Ezumah, and D. S. C. Spencer
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
        Page 95
        Page 96
    The use of computers in field research, by P. A. Tatian
        Page 97
        Page 98
        Page 99
        Page 100
        Page 101
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
        Page 109
        Page 110
        Page 111
        Page 112
        Page 113
        Page 114
        Page 115
        Page 116
        Page 117
        Page 118
    Agricultural development in the Nazareth Region, Israel: Twenty-five years of extension, by Hisham Yunis
        Page 119
        Page 120
        Page 121
        Page 122
        Page 123
        Page 124
        Page 125
        Page 126
        Page 127
        Page 128
        Page 129
        Page 130
    Trial designs and logistics for farmer-implemented technology assessments with large numbers of farmers: Some approaches used in Botswana, by G. M. Heinrich and S. Masikara
        Page 131
        Page 132
        Page 133
        Page 134
        Page 135
        Page 136
        Page 137
        Page 138
        Page 139
        Page 140
        Page 141
        Page 142
        Page 143
        Page 144
        Page 145
        Page 146
    Rural women in irrigated and rain-fed rice farming in the Philippines: Decision-making involvement and access to productive resources, by D. Timsina, A. L. Ferrer, T. Paris, and B. Duff
        Page 147
        Page 148
        Page 149
        Page 150
        Page 151
        Page 152
        Page 153
        Page 154
        Page 155
        Page 156
        Page 157
        Page 158
        Page 159
        Page 160
        Page 161
    Instructions to authors
        Page 162
Full Text
Volume 3, Number 2
1993


journal
for Farming Systems
Research- Extension









59 FRa d th ualPo







Journal
for Farming Systems
Research-Extension


Volume 3, Number 2, 1993


Published by
the Association for Farming Systems Research-Extension








Journal for Farming Systems Research-Extension


Editor
Timothy R. Frankenberger
Office of Arid Lands Studies
The University of Arizona, Tucson

Associate Editors
Daniel M. Goldstein and Philip E. Coyle
Office of Arid Lands Studies
The University of Arizona, Tucson

Production and Layout
Diedre Muns, Daniel M. Goldstein, and Philip E. Coyle
Arid Lands Design, Office of Arid Lands Studies
The University of Arizona, Tucson








The Journalfor Farming Systems Research-Extension is published by the Association for
Farming Systems Research-Extension (AFSRE), an international society organized to
promote the development and dissemination of methods and results of participatory on-
farm systems research and extension. The objectives of such research are the development
and adoption through participation by farm household members of improved and
appropriate technologies and management strategies to meet the socioeconomic and
nutritional needs of farm families; to foster the efficient and sustainable use of natural
resources; and to contribute toward meeting global requirements for food, feed, and
fiber.
The purpose of the Journal is to present multidisciplinary reports ofon-farm research-
extension work completed in the field, and discussions on methodology and other issues
of interest to farming systems practitioners, administrators, and trainers. The Journal
serves as a proceedings for the annual international Farming Systems Symposium from
which selected and refereed papers are included. It also welcomes contributed articles
from members of the AFSRE who were unable to attend the symposium. Contributed
articles will be judged by the same review process as invited articles.


ISSN: 1051-6786









Journal for Farming Systems Research-Extension
Volume 3, Number 2, 1993


CONTENTS


1 Integrating Gender Issues Into Farmer-Participatory Research: The Case of
Vegetable IPM Technology Generation in Calamba, Laguna, Philippines
M.M. Hoque and C.B. Adalla

13 The Effect of Storage Loss Rates on the Valuation of Maize Stored
Traditionally by Farmers and Removed Periodically for Food, Feed, or Sale
in Cameroon
Dermot McHugh

25 Gender Issues in Farming, Part II: Selected Case Studies in Development
and Extension of Farm and Other Equipment to Women in Northern
Nigeria
R.N. Kaul

39 Informal Agricultural Communication Patterns in a Remote Area of
Bangladesh
S.M.A. Hossain, B.R. Crouch, and Shankariah Chamala

59 Farming Systems Research and the Rural Poor: A Political Economy
Approach
S.D. Biggs and John Farrington

83 Farm-Level Performance of Improved Cassava Varieties in the Humid
Forest Zone of Nigeria
RII. Nweke, H.C. Ezumah, and D.S.C. Spencer

97 The Use of Computers in Field Research
P.A. Tatian

119 Agricultural Development in the Nazareth Region, Israel: Twenty-Five Years
of Extension
Hisham Yunis

131 Trial Designs and Logistics for Farmer-Implemented Technology Assessments
With Large Numbers of Farmers: Some Approaches Used in Botswana
G.M. Heinrich and S. Masikara

147 Rural Women in Irrigated and Rain-Fed Rice Farming in the Philippines:
Decision-Making Involvement and Access to Productive Resources
D. Timsina, A.L. Ferrer, T. Pars, and B. Duff








Integrating Gender Issues Into
Farmer-Participatory Research:
The Case ofVegetable IPM Technology
Generation in Calamba, Laguna, Philippines1

M.M. Hoque and C.B. Adalla 2



ABSTRACT
This paper highlights the authors' experiences integrating gender issues
into the generation and verification ofcrop-protection technology. The
project described was an off-shoot of a pilot project on the participatory
verification of rice IPM technology, in which farmers had expressed the
desire to investigate their vegetable pest problems. Initially, the project
focused on technology generation with male farmers. After identifying
the roles of men, women, and children in the farming households,
appropriate training was given to enhance the productivity of women.
The involvement of male and female farmers in the two-and-a-half-year
study made them aware of an approach to reduce losses in vegetables
through application of insecticides based on economic threshold levels
rather than on the sight of pests in their fields. Researcher/farmer
interaction raised the farmers' consciousness ofthe health implications of
pesticide over-use. Where they harvest for home consumption, women
now set aside rows of their vegetables to remain unsprayed.


INTRODUCTION
A project on the development of insect integrated pest management (IPM) on
vegetable crops was initiated in the latter part of 1987 in Looc, Calamba,
Laguna, Philippines (Figure 1). This was an effort to address one of the
expressed needs of the cooperators in an earlier rice IPM technology verifica-
tion project that was introduced in the village through a Food and Agriculture
Organization grant in 1986. After several meetings and casual interviews with
the rice farmer cooperators, the researchers decided to set up a research

1 Paper presented at the 11 th Annual Farming Systems Research-Extension Symposium, Michigan
State University, East Lansing, October 5-10, 1991.
2 Associate Professors, National Crop Protection Center and Department of Entomology, respec-
tively, College of Agriculture, University of the Philippines at Los Bafios, College, Laguna, the
Philippines.






HOQUE AND ADALLA


Figure 1. Map of Looc, Calamba, Laguna, Philippines.
demonstration with vegetables following the approach used in the rice
project.
IPM basically integrates two or more strategies (e.g., cultural and biolog-
ical, or the use of resistant varieties and chemical pesticides) when pest
population levels become economically damaging. IPM technology on
vegetables in the Philippines is still in its component technology generation
stage, although the generation of IPM components for rice has been going on
for years. Public concern over the potential health hazards ofeating foods that
are heavily laden with pesticides has been on the rise. Certainly the use of IPM
technology for vegetables would greatly minimize pesticide problems. Hence,
the studies conducted under this project theoretically should have developed
baseline information towards the development of an effective, economical,
and acceptable IPM approach for vegetables. As the project progressed,
however, the researchers found it necessary to study the participation of the
family members in some vegetable-farming households. Thus, this report
represents a more qualitative account of their observations.
Integrating gender issues into the generation and verification of crop
protection technology (IPM) particularly in vegetables requires an under-
standing of the roles, resources, and interests of women as well as men in
agricultural production. This paper recounts the processes that the research-
ers went through to enable them to learn about and enhance women's roles


Journal for Farming Systems Research-Extension






GENDER ISSUES AND IPM TECHNOLOGY


in vegetable production and pest management. The inclusion of women in the
project resulted in greater farmer interest and better information for the
researchers. This paper highlights the process used in identifying female
farmers' roles in crop production, as well as the activities undertaken to
improve their skills and knowledge of the use of the technology both to
complement their husbands' work and as users of the technology themselves.

Description of Cropping Practices and Study Site
The study site is located approximately 3 km from the town ofCalamba and
can be reached by motor vehicle (Figure 1). The majority of the vegetable
fields are situated within a 2-km radius bordering Laguna de Bay. A large
proportion of the fields can be planted only during the first half of the year
before the arrival of monsoon rains. With typhoons and heavy rains arriving
as early as May, fields bordering the bay may be submerged most of the year.
The majority of the farmer cooperators have parcels of land in this condition
but some also have fields that can be cultivated throughout the year. Farmers
in the study site planted a variety of vegetable crops and also practiced relay
and intercropping of such crops as okra in between string beans, okra after
beans, eggplant after tomatoes, eggplant after beans, white gourd after beans,
and other combinations.
Twenty-six farming families joined the vegetable project in 1988, 18 in
1989, and 12 in 1990. The decrease in the number of cooperators between
1988 and 1990 was due to one or more of the following reasons: (1) farmer's
failure to comply with what was agreed upon at the start of the project, (2)
changes in the crop grown by the cooperator to a crop that may not be among
the researchers' priority crops, and (3) failure to attend the monthly meetings.


METHODOLOGY
We used a participatory methodology wherein farmers played important roles
in almost all of the activities. In the highly technical component generation
studies, the farmers were less directly involved; nevertheless, they were kept
abreast of what was going on in their fields. Focus was given to four of the
vegetables identified as the most widely grown in the area based on a survey
in 1987. Research data were gathered daily by the research assistant, who was
hired particularly for the project and resided in the village.
The project leader (Hoque) visited each cooperator at least once every two
weeks. The frequent visits of the project staff with the farmers established a


Vol. 3, No. 2, 1993






HOQUE AND ADALLA


congenial working relationship between them. During these farm visits,
project staff engaged farmers in discussions of farm activities, production
problems (including crop protection), and domestic life. The project leader
documented these observations by tape recorder or in a record book and,
whenever possible, recorded them in a computerized format.


DISCUSSION AND RESULTS

Research and Technology Generation
An example of the insect control strategy generated with the farmers was
a need-based application of control based on the economic threshold level
(ETL) of the pests. The ETL-based insecticide application strategy is based
on the pest-population level obtained through monitoring. Insecticide
spraying is done when the designated larval population reaches a level of 2.7
larvae/20 plants. This figure was arrived at by examining 80 random hills at
any time when the insect population became evident. A previous experiment
was performed in the same locale in 1987 (Hoque and Saavedra, 1988) on the
major bean pests, namely bean leaf folder (Lamprosema indicate), cutworm
(Spodoptera litura), and pod borer (Maruca testulalis). On the other hand,
the farmer's (FP) use of pesticide is based on their perception of the pest level
and degree of damage, without quantifying the pest population.
An economic analysis of the cost and return using partial budgeting showed
the differences between the two pest management practices. For string beans,
the need-based (ETL) plot and the farmers' practice plots (FP) produced
equal quantities of harvested pods and returns per unit area for three seasons
(Table 1). The difference, however, lies in the cost of insect control in FP,
which was almost 50 percent higher than the ETL plot for 1988 and 1989.
For the first two years, the frequency of pesticide application during
priming was consistently and significantly higher in the FP (means of 6.0 and
5.1 in 1988 and 1989, respectively) than in the ETL plot (means of 1.7 and
2.9). In the 1990 trial, the difference between cost of control and frequency
of spraying was greatly reduced in FP compared with that of the previous two
years. A correlation analysis was run on the data to determine the relationship
between the cost of insect control versus volume of undamaged pods. The
results showed that as spraying frequency increases, the volume of undamaged
pods decreases (r = -0.416, P<0.05). However, the returns above pest control
cost did not vary significantly between the treatment plots, a finding which


Journal for Farming Systems Research-Extension






GENDER ISSUES AND IPM TECHNOLOGY


Table 1. Comparison of Farmers' Insect Control Practices Versus Need-based
Chemical Protection in String Beans (Looc, Calamba, Laguna, Philippines).'

1988 (n=18)b 1989 (n=21) 1990 (n=12)
Factors ETLc FPc ETL FP ETL FP
Total harvest (kg, 1st to 8th harvest)
first class 11.44 10.54 9.08 8.34 7.44 7.17
second class 2.39 2.35 2.18 2.26 1.58 1.32
total 13.38 12.89 11.26 10.60 9.02 8.49
Cost of control (Peso) 6.49 11.41 5.88 11.81 6.02 8.99
Returns (Peso)
gross 46.49 43.11 64.49 62.13 61.11 57.61
above pest control 40.00 31.70 60.61 50.32 55.09 48.62
Frequency of spraying
before priming 4.5 5.8 4.6 7.1 5.5 5.8
during priming 1.7 6.0d 2.9 5.1 2.4 3.2
Mean number of larvae/
20 plants 1.98 2.06 1.98 1.9 2.41 2.36
a Per 20 m2.
b N = number of paired fields for the season.
c ETL = Economic Threshold Level and FP = Farmer Practice Plot.
d Differences are significant at P < 0.05.


strongly supports the researchers' recommendation not to spray during
harvesting.
The strategy developed with the farmers for the management of bean pests
has shown that it yields higher profits with minimal use of insecticides during
priming. In spite of what was shown to the farmers, they were still reluctant
to follow what was practiced in the ETL plots. The participatory approach
used in this experiment may have paved the way for the adoption by some
farmer cooperators. However, this still remains to be seen because the
pressure to produce vegetables free from insect damage makes it difficult to
convince farmers to change their current pesticide-based control practice. On
the other hand, the researchers are capitalizing on the increasing costs of
pesticides and the growing concern of women for pesticide-free crops and a
cleaner environment to eventually persuade farmers to shift to the ETL
strategy or to IPM when the technology is packaged.

The Strategy for Integrating Gender Issues into the Project
Though the biological scientists were all women, the initial study on
vegetables in 1987 was not planned with attention to gender issues because
women were not perceived as different from men as users of crop-protection


Vol. 3, No. 2, 1993





HOQUE AND ADALLA


technology. Contrary to the researchers' expectations, however, women play
a significant role, especially in vegetable production. Hoque and Saavedra
(1988) observed that farmers' wives participated in about 15 percent of
production activities. The women control the purchase of pesticides and
fertilizers and harvesting and marketing of produce. Table 2 shows the total
involvement of women in production. Overall, women's role in decision
making in vegetable production (45.4 percent) was greater than that of men
(27.3 percent), particularly in hiring labor for weeding (50 percent), harvest-
ing (100 percent), and watering and irrigating the fields (87.5 percent). This
is understandable because Filipino women are traditionally the keepers of the
family purse, besides being guardians of family health and education.
The observations of the women's involvement in vegetable production
shifted the researchers' attention in 1988 to integrating women into the
vegetable project. Two major criteria for selecting vegetable cooperators in
the 1988 International Development Research Centre (IDRC)-funded project
were (1) willingness to participate in the research activities to be conducted
on their farms and (2) involvement of women in production of both rice and
vegetables. The focus on women's participation was made because of the
authors' belief that addressing women as well as men might pave the way for
a faster adoption of the technology developed. However, some difficulties
were encountered in selecting cooperators who met both of these major
criteria. Because the primary objective of the vegetable project was the
generation of IPM technology with women's concerns being secondary, the
researchers decided to settle for whatever number of vegetable-farming
households met the first criterion. The vegetable cooperators were recruited
in their fields as they tended their farms, instead of by using a benchmark
survey, as had been used in the rice project as a basis for selection. This was
because project funds were released at the peak of the planting season. To have
waited for the results of the survey would have meant a big delay in starting
the study. In the initial 26 cooperating households, 10 women were active in
vegetable production. These households remained with the project all three
years of its operation.
During the recruitment of cooperators, it was apparent that the importance
of women's involvement was not yet well recognized by either the women
themselves or by their husbands. The women perceived that their involve-
ment was but a natural duty to their families. They themselves took these
duties for granted. It was only after several regular farm visits, casual


Journal for Farming Systems Research-Extension






GENDER ISSUES AND IPM TECHNOLOGY


Table 2. Degree of Participation by Men, Women, and Children in Vegetable
Production in Looc, Calamba, Laguna, Philippines.'

Activities Mb W M&W W,&C All
(%) (%) (%) (%) (%)


PRODUCTION
Land preparation
Seedbed preparation
and care
Seed selection
Planting/transplanting
Cultivation
Weeding
Fertilizer application
Watering/irrigation
Purchase of pesticides
and fertilizers
Spraying
Percent contribution to
production activity
POSTPRODUCTION
Harvesting
Sorting and packing
Hauling
Marketing
Percent contribution to
postproduction
DECISION MAKING
Buying of seeds
Buying of fertilizers
and pesticides
Hiring labor for
spraying
weeding
harvesting
irrigation
Vegetable selection
Percent contribution to
decision making


100
75

0
12.5
100
50
80
50
0

100

56.8

12.5
0
100
0


0 0 0
0 25 0


12.5
12.5
0
0
20
0
100

0


87.5
37.5
0
25
0
25
0


0 0


0
0

0
37.5
0
12.5
0
12.5
0

0


14.5 20 2.5 6.2


62.5
37.5
0
100


29 51.6


12.5
25
0
0


9.7 6.5 3.2


12.5 37.5
25 0


37.5
50
100
87.5
0


12.5
37.5
0
0
100


45.4 27.3


a There were eight respondents, with an equal number of men and women; seven were current
cooperators who planted tomatoes and beans. "Women" refers to wives of cooperators and/or
adult women hired in the postproduction activities.
b M = Men, W = Women, and C = Children.
c Percent contribution is the average of all the activities.

interviews, and conversations with the researchers that the farming house-
holds recognized the important roles women played in vegetable production.
Seven of the ten wives who were actively involved in vegetable production
were seen frequently in the fields. When they were asked what motivated their
involvement in vegetable production activities (supposedly a man's activity),


Vol. 3, No. 2, 1993





HOQUE AND ADALLA


almost all answered "for economic reasons," i.e., to augment family income.
Most of them apparently came from farm families and as such already had
experience with vegetable farming. Most of the women were literate and had
experience with keeping records on their farms.
The researchers' regular farm visits (a must in the generation and verifica-
tion of IPM technology) added a special dimension to the process of getting
the women interested in participating in the project. The researchers'
unhurried visits with the farmers, for example, helped the researchers' gain the
confidence of both men and women cooperators. This procedure was rather
expensive and time consuming on the part of the researchers, but was very
important for the future adoption of the technology being generated and
verified. For example, during farm visits when harvesting was going on, the
researchers sat down with the cooperators, hired vegetable harvesters, and
middlemen and helped sort the produce. It was on these occasions that the
researchers engaged the group in important discussions about such matters as
the social implications of the IPM technology in general. Most of those
involved in these discussions were women, because the majority of the hired
laborers were women.
Often the researchers were offered vegetables to take home with them. On
many occasions, the researchers graciously turned down the offers explaining
that the pesticide residues in vegetables are health hazards. This opened up
wider discussion of the health problems caused by residues in vegetables
consumed by the public. The women reported cases of lingering illnesses in
their village and wondered whether these were due to pesticides.
A special seminar was held each time the results of the IPM experiments
were analyzed. These were held periodically, corresponding with the different
growing periods of different vegetables. Each farm family was provided with
a copy of the economic analyses of their own farm. Those who were not able
to attend this special seminar were personally visited by the project staff, who
explained the results to them, emphasizing the technology based on the
minimum use of pesticides.
The majority of the farmers who attended the meetings were the husbands.
Occasionally, their wives also attended. The wives who attended the meetings
were those frequently seen in the fields attending to their vegetables. During
those sessions, the farmers were surprised to learn that the profits derived from
their plots (FP) were lower than the researcher-managed (ETL) plots. In
terms of pest-control inputs, the researcher-managed plots had fewer inputs
compared with those of the farmers' plots.


Journalfor Farming Systems Research-Extension






GENDER ISSUES AND IPM TECHNOLOGY


By the second year of the project, the technology had been verified for two
years in the area. The women had become more enthusiastic, and the
researchers suggested that they meet at least once a week for casual conversa-
tion in their fields while they were shown how to monitor the pests. The
women decided on the day and time for these meetings. About six to seven
women attended the meetings, with a woman rice farmer as one of the
participants. This woman served as the catalyst and inspiration for the others
to regularly attend the meetings.
At the start, the researchers showed the results of the pesticide residue
analyses of vegetable samples. The data were explained in terms ofthe possible
health consequences of consuming vegetables sprayed with insecticides before
the recommended postharvest interval. The group's fear was aroused when
they saw the results of the pesticide residue analysis, but it was a fear that
spurred their interest in the technology being verified in their fields.
Has a focus on women in the project made any difference? The answer is
yes. Indeed, it made a significant impact on both male and female farmers as
well as on the female researchers. This can be gleaned from the following
accounts of the women and the observations of the researchers.
Although there was no formal assessment of the knowledge and skills
gained in this process, the authors observed a significant change in the
women's attitudes towards acquiring skills that enable them to make decisions
on when to apply the control. The researchers also observed that towards the
end of the project the women had become more assertive insofar as buying the
"right" pesticide was concerned. Whereas it used to be the husband who
prescribed what the wife would buy, she could now discuss intelligently what
pesticide was needed to control a particular pest. This stemmed from the fact
that both husbands and wives are acquiring similar skills from the IPM
researchers. Another remarkable change was in the women's confidence and
self-esteem. This was clearly shown when two of the female cooperators
accepted invitations to attend a three-day workshop and conference on IPM
at the University of the Philippines at Los Bafios. Both acted as discussants
in two of the sessions.
Aling Mereng is a rice farmer who often attends the meeting of the women
vegetable farmers. Her enthusiasm rubbed off on the other women and she
proudly claimed:
Actually to practice IPM is not difficult. What is difficult is getting started.
Personally I thought it would interfere with my household responsibilities


Vol. 3, No. 2, 1993






HOQUE AND ADALLA


because we were requested to attend meetings and training sessions particularly
on weekends, which of course is the schedule for pressing my children's school
uniforms. Then I thought of the professional staff who were likewise mothers
and yet spending their Sunday afternoons with us, so I gave it a try and from
then on I got adjusted to the situation.
Another woman, Aling Priscing (a vegetable farmer) said:
I used to have our crops sprayed at the sight of insects damaging our vegetables
while my husband often contradicted me on this practice. He even chided me
for making the chemical dealer all the more richer by my practice. But now I no
longer spray without consulting the IPM project technician who resides in the
village.
Based on experiences such as these, we are convinced that integrating
women in the project did make a significant difference. The researchers'
interaction with them has raised the women's awareness of the health
implications of pesticide overuse, and they now set aside rows of their
vegetables to remain unsprayed for home consumption.
The project staff, who are female biological scientists, were sensitized to the
importance of integrating gender issues into their research on technology
development. In this project, where a participatory methodology was applied,
we observed firsthand how an appropriate pest control technology was
developed based on the perceived needs of farmers and the identification of
the clientele or technology users. It was evident that for a better adoption of
a technology, the roles of women and/or other members in the production
line must be identified. In so doing, the use of developed technology may be
ensured.

Some Lessons Learned and Recommendations
The two-and-a-half-year period with the farmers was not long enough to
have created a significant impact insofar as technology adoption was con-
cerned. This points to the need for a longer duration of projects of this nature
for the farmers to fully appreciate and adopt the technology generated with
them.
One important consideration in the introduction of an intervention
(technology or training) to enhance women's productivity, as in the case of
vegetable farming, is to observe closely their involvement. Because they
themselves consider their farm activities as a natural consequence of being
married to farmers, upon initial inquiry they may not be explicit about their
roles. Having identified their roles in vegetable production (e.g., pest


Journal for Farming Systems Research-Extension






GENDER ISSUES AND IPM TECHNOLOGY


control), the women could be taught monitoring skills that would enable
them to make correct decisions on when to apply appropriate control
measures.
Once research had begun, learning about and addressing women's roles
contributed strongly to meeting the primary objective of verifying insect IPM
technology. Many women actively engaged in vegetable production as
farmers and hired laborers. All female farmers were solely responsible for the
purchase of pesticides. Being included in training made them better collab-
orators with their husbands in testing and applying IPM. Women are also the
guardians of family health and were therefore especially important to reach
with information on the dangers of pesticide residues in food.


ACKNOWLEDGMENTS

The authors gratefully acknowledge the International Development Research
Centre (IDRC) in Ottawa, Canada for the funds provided to the University
of the Philippines at Los Bafios, which enabled them to conduct the project
on IPM Extension and Women. Special thanks to Dr. Kenneth T. Mackay for
his interest in promoting the project and to Dr. Hilary Feldstein for editing
the manuscript.


REFERENCE
Hoque, M.M., and N.L. Saavedra. 1988. The role of women in the optimization of
inputs for vegetable production in a rice-based farming community: A case study. In
Filipino women in rice farming systems. University of the Philippines at Los Bafios-
PIDS-International Rice Research Institute.


Vol. 3, No. 2, 1993











The Effect of Storage Loss Rates on the
Valuation of Maize Stored Traditionally by
Farmers and Removed Periodically for Food,
Feed, or Sale in Cameroon'

Dermot McHugh2



ABSTRACT
Farmer surveys consistently have shown that maize storage losses are of
primary concern to farmers in Cameroon and that storability ranks high
among the characteristics used by farmers in their assessment of new
varieties. Although new, high-yielding maize varieties have been well
received for their taste and cooking characteristics, farmers complain that
the large, denty (soft) grains do not store aswell under traditional storage
conditions as the small, flinty (hard) grained local varieties.
This qualitative deficiency is not captured in partial budget analyses, in
which economists value net harvest maize yield at a single price. Even the
reduction ofnet yield by an estimate ofstorage loss before valuation does
not adequately model the true relative benefits to Cameroonian farmers,
who store their maize from harvest to harvest, removing it piecemeal for
food, feed, or sale throughout the year. This paper proposes a budgeting
procedure for examining the effects ofquantitative and qualitative storage
losses on the valuation of maize. Hypothetical data is used to show how
the new higher-yielding but less storageworthy varieties might be better
evaluated.


INTRODUCTION
Maize (Zea mays) and sorghum (Sorghum vulgare) are the principal cereal
crops in Cameroon, with annual production figures of 600,000 and 500,000
metric tonnes respectively (National Directorate of the Agricultural Census,
1987). Whereas sorghum production is limited to the northern dry savanna
zone (2 of 10 provinces), maize is grown throughout the country, with the
single exception of the Extreme North Province. The western highland zone
(above 1,000 m in altitude), situated in the North West and West Provinces

1 Paper presented at the Tenth Annual Association for Farming Systems Research-Extension
Symposium, Michigan State University, East Lansing, October 14-17, 1990.
2 Institute de la Recherche Agronomique/National Cereals Research and Extension Project/
Testing and Liaison Unit, Bambui, Bamenda, Cameroon.
13






MCHUGH


(Figure 1), makes up only 10 percent of the land area and 20 percent of the
population of Cameroon and yet produces over 60 percent of the national
maize crop.



AFRICA


Figure 1. North West Province of the Republic of Cameroon.


Journal for Farming Systems Research-Extension





STORAGE LOSS RATES AND MAIZE VALUE IN CAMEROON 15

In this zone, maize is the daily food for most of the population (more than
2 million). It is usually eaten as fufu-corn, a stiff porridge served with a leafy
vegetable (jamajama) or stew. Other forms in which maize is consumed
include cornchaff (fried with beans), corn-kokey (a steamed pudding, often
prepared with dried fish and hot pepper), pap (a thin porridge eaten in the
morning), roasted green ears, and cornbeer.
Maize almost always is grown in association with other crops, including
beans (Phaseolusvulgaris), groundnuts (Arachishypogaea), irish potato (Sola-
num tuberosum), cocoyam (Colocasia esculenta and Xanthosoma sagittifoli-
um), cassava (Manihot esculenta), sweet potato (Ipomoea batatas), yam
(Dioscorea spp.), and coffee (Coffee spp.). The crops in the association are
determined by altitude, soil type, economic and cultural factors, and farmers'
preferences. Crop associations typically vary from farm to farm, either in crop
combination or planting pattern. Mean maize grain yields in the zone range
from 1 to 3 tons per hectare (McHugh, 1988).


MAIZE STORAGE

In the mid- to high-altitude zone of the North West Province, maize is
harvested from late July until early November, depending upon planting date,
variety, and altitude, which ranges from 1,000 to 2,400 m. Maize is either (1)
dried in a loft above the kitchen fire (banda) and stored there in the husk or
(2) completely or partially dehusked and air dried on raised platforms
(wunchum and tapkwa) or hung from the eaves of the house and then stored
in a rectangular bamboo silo (nchang) by the ear. The first drying and storage
method is practiced at the lower elevations where harvest occurs at the height
of the single rainy season (March-October), and the second at higher
altitudes, where it is harvested shortly before the onset of the dry season
(November-February).
A minority of farmers use a chemical insecticide (actellic 2 percent powder)
to control storage insects (principally the maize weevil Sitophilus zeamais),
which are the main sources of storage losses. Others use local aromatic plant
materials; the majority use no control methods whatsoever.
The maize is stored from one harvest to the next, barring early depletion
of the stock as a result of insufficient production, unexpectedly large demands
on the supply, and/or excessive storage losses. The maize is removed
piecemeal over the storage period at weekly to fortnightly intervals for
consumption, sale, or animal feed. During removal from the store, the maize


Vol. 3, No. 2, 1993






MCHUGH


ears are dehusked and graded into two categories: "good maize" (for
consumption or sale) and "bad maize" (for animal feed or sale as feed, or
rarely, outright disposal).
The maize is usually sorted while still on the cob, so that an ear with mostly
damaged (weeviled or mouldy) grains is set aside for animal feed and other ears
with only a few bad grains are kept for family consumption. As a result, some
undamaged grains go into animal feed and some damaged grains are left in the
maize destined for human consumption or sale as such.


VALUATION OF MAIZE

In comparing a new high-yielding maize variety with the farmer's variety,
farming systems economists typically use net yields from on-farm trials,
evaluate them at a single price (e.g., price at harvest or mean annual price) and
deduct variable costs to obtain the net benefits. The marginal net benefit
(MNB) and marginal variable cost (MVC) are then used to calculate the
marginal rate of return (MRR) for farmers adopting the new variety:
(1) MRR = MNB/MVC x 100.
The partial budget approach described above provides a valid comparison
of maize varieties assuming, among other things, that (1) in nonyield
characteristics the new variety is at least equal to the old and (2) valuing the
total quantity of maize at a single price accurately reflects the economic value
of the maize to the farmer.
Examples of nonyield characteristics are taste, cooking quality, and stor-
ability. New varieties released by the Institut de la Recherche Agronomique
(IRA), the national research institute, to date have taste and cooking charac-
teristics that are more than acceptable to farmers. However, farmers complain,
often bitterly, of their short storage life. The large, soft, denty grains of the
"improved" maize varieties are apparently more susceptible to attack by the
maize weevil than the small, hard, flinty grains of the typical "local" variety.
As a result, the yield advantage of the new varieties is compromised by large
storage losses, both quantitative and qualitative, when stored for extended
periods.
It is not valid to value maize at a single price; because they consume maize
throughout the year, farmers are more likely to value each quantity of maize
removed from the store at current prices, rather than at a fixed price. For
example, farmers removing a bucket of maize for consumption in April when


Journal for Farming Systems Research-Extension






STORAGE LOSS RATES AND MAIZE VALUE IN CAMEROON 17

it can be sold at'1,000 francs CFA3 will certainly not value it at the 400 francs
CFA they would have received in October, immediately after harvest.
Poor storage quality has led to limited adoption of the improved varieties
by farmers. This qualitative deficiency is not, however, captured in conven-
tional partial budget analyses, which paint an overly optimistic picture for the
new varieties. To obtain a proper value assessment consumption patterns,
storage loss rates (quantitative and qualitative), and changing prices and
quality standards over the year have to be factored into the analysis.
In the following section, a model (budget procedure) is proposed that
values maize at the time of removal from the store, while adjusting for grain
quality and weight losses in storage and changing quality standards. An
illustrative example is presented, showing how the model might be used to
assess yield gains for an improved maize variety, when it is a poor storer relative
to the local variety.


THE MODEL

The underlying assumptions of the "stored-maize valuation model" are
1. Maize is removed piecemeal from the store throughout the storage period
(<12 months).
2. Maize is removed to meet target food or sale quantities (good maize). Feed
(bad) maize is a by-product of screening (sorting) the food/sale maize.
3. Quality standards for maize consumed by the household or sold in the
market tend to decline as one moves away from the immediate postharvest
period and toward the next harvest, as stores are drawn down and prices rise.
[Quality is measured by the percent damaged (weeviled or mouldy) grains in
the maize.] The quality of feed maize will also decline.
4. Farmers value their maize at the time of end use, or (for purposes of the
model) at the time of removal from the store.
Assumptions 1, 2, and 3 have been confirmed by observation and in
conversations with farmers in the North West Province. Assumption 4 is taken
as rational and justifies the model.
As the maize is removed from the store it is sorted to satisfy the quality
requirements defined by the farm family or market, with the rejected maize
being fed to the farm animals (chickens, pigs, and goats). Over the storage
season, the quality of the maize in the store declines (i.e., percent damaged
grains increases). Therefore, more and more maize must be removed to
3 US$1 = 280 francs CFA.


Vol. 3, No. 2, 1993





MCHUGH


provide the same quantity of food-quality maize for consumption or sale and
more and more will be rejected (i.e., fed to the animals).
However, it has been observed that quality standards change as the quantity
of maize in the store is reduced. The family is willing (or obliged) to accept
lower quality maize for consumption as supplies diminish. Maize fed to
animals also suffers a decline in quality. These facts partially offset the effect
of the decline in the quality of the stored maize on the amount of maize that
must be removed to satisfy the food-maize requirement. All of these factors
have been built into the "stored-maize valuation model," along with the
effects of seasonal price variations.
By adjusting for storage loss rates (both quantitative and qualitative),
changing quality standards imposed by the farmer and the market, quantities
of food/sale maize removed in each period, and market price trends, the
model should provide a better assessment of the value of maize produced and
stored by the farm family than conventional budgetary approaches that use net
harvest yields and a single price.
The principal variables in the model include the total quantity of maize in
the store (Q), the quantity of food/sale maize removed from the store (C),
the quantity of feed maize removed from the store (Fl), the total quantity of
maize removed from the store (R), the marginal storage weight loss (Mi), the
proportion of damaged grains in the stored maize (Di), the proportion of
damaged grains in food/sale maize (At), the proportion of damaged grains in
feed maize (0), the price of maize (P.), and the total value of the stored maize
(V). The subscripts (i) identify the storage period. All the variables are defined
and the relevant formulas presented in Appendix A.
Two types of storage losses are represented in the model: (1) quantitative
loss (proportion by weight of original quantity) and (2) qualitative loss
(proportion of damaged grains). It is assumed that both types of losses are a
function of the weevil population, and that the weevil population grows
exponentially. Both types of losses are, therefore, represented by an exponen-
tial function in the model:
(2) L = a (e 1 )
where Li = the cumulative potential storage loss in the it period; 0 < Li. 1;
i = the storage period (month); i = 1...12; a and b are estimated.
The following assumptions are made in the examples presented in the next
two sections.
1. Quantities of maize are weighed at constant moisture content.


Journal for Farming Systems Research-Extension






STORAGE LOSS RATES AND MAIZE VALUE IN CAMEROON 19

2. Sixty kg of food quality maize is removed from the store at the beginning
of each month.4
3. The tolerance for damaged grains in food maize increases linearly from 0
to 5 percent over the year.
4. The percent damaged grain in feed maize also increases in a linear fashion
from 10 to 90 percent.5
5. Each quantity of food maize, whether for consumption or sale, is valued
at the time of removal at prevailing market prices.6
6. Each quantity of feed maize is valued at 40 percent of the food maize price.7
7. Any remaining maize in the store at the end of 12 months is removed and
valued at prevailing prices.


APPLICATION OF THE MODEL

Several storage-loss level scenarios were evaluated using a computer spread-
sheet programmed for the model. Results are presented in Table 1, where the
initial quantity stored was set at 1,000 kg.8
The 1st column figures represent the weight loss that would occur if all the
maize was left in the store for the full 12 months. The 2nd column indicates
the quality of the maize after 12 months in storage. The 3rd column shows
the actual total weight loss, as a percentage of the initial stored weight (i.e.,
1,000 kg). Because maize is being removed throughout the storage period,
the actual weight loss (column 3) is less than the potential loss (column 1).
The quantities of food-quality maize remaining in the store at the end of the
12 months are shown in column 4. The total value of the maize under each
combination of quantitative and qualitative loss assumptions (columns 1 and
2) is shown in column 5.
If there were no storage weight losses and quality remained constant, the
value of the maize to the farm family would be 55,400 francs CFA. If, on the
other hand, the potential weight loss was 60 percent and the proportion of
damaged grains in the stored maize at the end of the year was 70 percent, the
value would be 43,020 francs CFA (representing a 22 percent loss in value).

4 This is indicative of removal rates reported in an ongoing TLU farmers' maize storage survey in
collaboration with the FAO Food Loss Reduction Program, 1990.
5 The figures for food and feed maize quality trends, at this point, represent educated guesses, while
awaiting results of the TLU/FAO-FLRP storage survey.
6 Using a smoothed-trend line for prices observed at the selected markets in the North West
Province.
7 Ratio reported by farmers who sell feed maize.
8 Dry shelled maize equivalent.


Vol. 3, No. 2, 1993






MCIIUGII


Table 1. Effect of Various Quantitative and Qualitative Loss Levels on the Value of
Stored Maize to the Farm Family (Stored Maize Valuation Model Estimates).
Potential Damaged grains Actual total Remaining Total value
cumulative at the end of weight loss quantity of food of the
weight loss the year maize in the store maizea
(%) (%) (%) (kg) (CFA)
0 0 0.0 340 55,400
10 20 4.1 226 51,370
40 3.9 134 49,040
20 40 7.0 116 47,880
60 6.6 51 46,440
30 40 9.6 101 46,890
60 9.0 42 45,700
40 60 11.2 35 45,020
80 10.0 3 44,940
50 60 13.1 28 44,440
80 11.5 2 44,580
60 70 17.4 0 43,020
a Conventional partial budget valuation (net yield x price at harvest) = 35,000 CFA.


IMPROVED VERSUS LOCAL MAIZE VARIETY

For illustrative purposes, let us assume that the local (hard-grained) maize
variety suffers a potential 10 percent weight loss with 20 percent damaged
grains at the end of 12 months and that the improved (soft-grained) variety
suffers quadruple these losses (i.e., 40 percent and 80 percent). From Table
1, we can see that the total value figures for the same initial stored quantities
of local and improved maize arc 51,370 and 44,940 francs CFA, respectively.
Starting with the same quantity in store, the farmer who has switched to the
improved maize experiences a 6,430 francs CFA or 12.5 percent loss in value.
However, farmer-managed trials in 1989 demonstrated that improved
maize with a moderate application of fertilizer yielded 84 percent higher than
the local maize at the farmer's fertilizer rate (NCRE, 1989). Using these trial
results, a partial budget is developed in Table 2 comparing the improved and
local varieties, using, first, the conventional valuation method (net yield times
a single farmgate price), and then values generated by the simulation model.
The assumptions with respect to relative storage losses mentioned in the
previous paragraph are retained, as are the seven assumptions discussed above.
The conventional valuation yields a very high marginal rate of return (314
percent). Using the model-generated values reduces the MRR by more than
nine-tenths to 26 percent. In this instance, it would mean that a recommen-
dation to adopt the new variety would not be forthcoming.


Journalfor Farming Systems Research-Extension






STORAGE LOSS RATES AND MAIZE VALUE IN CAMEROON 21

Table 2. Partial Budget Comparing a Local Maize Variety With an Improved Variety
(0.5 ha).'
Conventional Model
valuation valuation
Item Local Improved Local Improved

Net yield (kg) 928 1704 928 1704
Total benefit (CFA) 32,480 59,640 48,411 56,680
Total variable coasts
(CFA) 6,062 12,625 6,062 12,625
Net benefit (CFA) 26,418 47,015 42,349 44,055
MRR 314% 26%
a Based on results of farmer-managed trials (NCRE, 1989).
b Seed, fertilizer, and labor.


Of course, there are a number of other factors that could affect the
analysis-a different pattern of maize removal from the store, use of storage
chemicals, price changes, etc. Also, some factors have been left out, including
the increased labor required to sort the improved maize and the cost of
increased storage space. If they had been factored into the analysis, a switch
to the improved variety would probably be shown to be even less advisable.
It is presumed that farmers do take all of the above into consideration (in
their own fashion), which explains, at least in part, why many refuse the new
denty varieties despite their large yield advantage, beautiful car aspect, and
admirable cooking quality. On advice of the TLU, the NCRE highland maize
breeder has taken steps to develop harder-grained varieties for the western
highlands of Cameroon.


CONCLUSIONS

In surveys and farmer assessments associated with on-farm variety trials,
farmers in the western highlands of Cameroon have frequently criticized the
new higher yielding maize varieties for their short storage life. This charac-
teristic stems in part from the denty grain type that is positively correlated to
yield. The local varieties usually have small flinty grains and are more
storageworthy.
The same problem exists in east and southern Africa, notably in Zambia
(Adams and Harman, 1977), where farmers typically grow SR52 hybrid maize
on part of the farm for sale shortly after harvest, while growing the local variety
on the rest of the farm for storage and family consumption over the year.


Vol. 3, No. 2, 1993






MCHUGH


Cameroonian farmers lack markets for large quantities of maize. They are
therefore obliged to store most ofit, removing small quantities periodically for
food, sale, and feed.
A model (budgeting procedure) was developed that values each quantity
of maize removed from the store at the prevailing farmgate price. The model
accounts for storage losses (quantitative and qualitative), rate of maize
removal from the store and changing qualities of food and feed maize over the
storage season.
The model was used to estimate the effect of various loss levels on the value
of stored maize to the farm family. Finally, an illustrative example was
presented showing how the model could provide a better assessment of yield
gains from a higher-yielding but poorer-storing improved maize variety.


REFERENCES

Adams, J.M., and G.W. Harman. 1977. The evaluation of losses in maize stored in a
selection of small farms in Zambia with particular reference to the development of
methodology. Report G109, Products Institute (Tropical Development and Research
Institute). Ministry of Overseas Development, London, 149 pp.
McHugh, D. 1988. Maize-based cropping systems in the Ndop Plain of the North West
Province, Cameroon-A monitored farm survey, with labor utilization data. National
Cereals Research and Extension Project, USAID/IITA, Bambui, 90 pp.
National Cereals Research and Extension (NCRE) Project. 1982-89. NCRE annual
reports, 1982-89. USAID/IITA, USAID Mission, Yaound6, Cameroon.
National Directorate of the Agricultural Census. 1987. 1984 Agricultural Census-
Traditional sector. Ministry of Agriculture, Yaound6, Cameroon.


Journalfor Farming Systems Research-Extension





STORAGE LOSS RATES AND MAIZE VALUE IN CAMEROON 23

APPENDIX A: STORED MAIZE VALUATION MODEL,
VARIABLES, AND FORMULAS

Li The potential cumulative weight loss for stored maize (due to insects,
mold, and/or rats) at the end of the i" period (proportion of original
quantity).9 0 < Li < 1
M The marginal weight loss for stored maize during the i" period, expressed
as a proportion of the quantity of maize remaining in the store at the end
of the previous period (Q,).
M = ( L-L)/(1-L) 0 < M< 1
Di The proportion of damaged grains in the stored maize at the beginning
of the i" period. 0 < D. < 1
Ai Observed proportion of damaged grains in food and/or market quality
maize at the beginning of the i* period. (This represents the acceptable
levels of damaged grains in maize destined for human consumption in the
i* period. It is assumed that quality standards will decrease over the course
of the storage season, as supplies are drawn down and maize prices go up.
Therefore, Ai should increase as one moves away from the harvest.)
0 < A 1
Ci The quantity (weight'0) of food or market quality maize removed from
the store at the beginning of the i" period. [kg]
W Excess quantity of damaged maize removed from the store in the process
of removing Ci food-quality maize at the beginning of the i" period. [kg]

Wi=C,(1-A)/(1 -D)-
= C, (Di- A ) / ( 1- Di)
O Observed proportion of damaged grains in feed-quality maize at the
beginning of the i' period. 00. O 1
Xi Additional quantity of maize removed from the store to build up the
proportion of undamaged grains in feed-quality maize to observed levels
(O0). [kg]
( D, X + W1) / (W + X, )or
X= (OW -W D ) /(D, O )
Fi The quantity of feed quality maize removed from the store at the
beginning of the i* period. [kg]
Fi = W + Xi = C (A- D, )/(i D- O )


9 All measures (variables) are by weight.
10 Weights based on shelled maize at standard moisture content.


Vol. 3, No. 2, 1993





MCHUGH


R The total quantity of maize removed from the store at the beginning of
the i* period. [kg]
R= C + F
= C(Ai- Oi/(D 0,)
Q0 The (effective) quantity of maize remaining in the store at the end of the
i" storage period. [kg]
Q,( P fod-q y m e -te p M)
Pc Price of food-quality maize in the i* period. [CFA/kg]
P. Price of feed-quality maize in the ih* period. [CFA/kg]
V Total value of maize in farmer's store, valued at time of end use, while
accounting for storage losses and changing quality standards. [CFA]
V = ( P C + P F ) + ( PCC2 + P, F, ) +..+ ( P Cn + P F )


Journal for Farming Systems Research-Extension








Gender Issues in Farming, Part II:
Selected Case Studies in Development and
Extension of Farm and Other Equipment to
Women in Northern Nigeria'

R.N. Kaul 2



ABSTRACT
It is well established that women the world over perform a wide variety
of tasks. Such tasks as cooking (partly for daily sale),water fetching, milk
processing, and crop threshing were chosen for closer study, and this
paper reports the findings as case studies done in two states in northern
Nigeria. Sample surveys, including Rapid Rural Appraisals (RRA), were
done to assess the scope of the technological constraints and other
problems faced by women. Improved equipment development was
targeted to ease drudgery and to allow more output and release time for
other activities. Maximum emphasis was given to modifying existing
technologies using available infrastructural facilities. Women in groups
and as individuals were involved during the phase of the project devoted
to development/adoption of tools.
This paper discusses these approaches, illustrating successes and failures
with women foodsellers, milk processors, introduction ofcropthreshers/
sellers, water-carrier improvement, and formation of women's cooper-
atives. Additional tasks for income-generation (such as raw sugar
preparation, use of solar driers, etc.) are being introduced to generate
extra income in the time saved by use of better equipment. Care was
exercised to ensure that technologies being developed enable women to
retain their traditional tasks.


INTRODUCTION

Perhaps one of the most studied topics in recent years has been the role of
women in different sectors of the economy and their overall influence on

1Part I of this paper appeared as Kaul and Ali (1992) in the Journalfor Farming Systems Re-
search-Extension 3(1):35-46.
2Professor and Program Leader, Agricultural Mechanization Research Program, Institute for
Agricultural Research, Samaru, Ahmadu Bello University, Zaria, Nigeria.






KAUL


national development. Recent reviews by Carloni (1987), Blumberg (1989),
Saito (1990), and Stamp (1990) give excellent overviews of the global work
done on issues concerning women's roles, particularly in Africa. This interest
in women's issues is also reflected in a review of nearly a decade's worth of
Women in Development reports (World Bank, 1988). Many aspects of this
problem are now glaringly clear and well documented, although in the so-
called "men's world" most planners appear to be following the "ostrich"
approach of hiding from these realities, in that there are few serious efforts
aimed at integrating women into development projects.

Women's Roles: Established Facts
Assortment of tasks. Women perform a wide assortment of tasks, including
farming, water collection, firewood collection, cooking, etc. Many factors
necessitate this heavy load, but the most important are sociological (cultural
organization of a society) and biological (motherhood and childrearing).
Both of these are difficult or impossible to change, although there have been
"interventions" in education, birth control, and some forced changes through
cultural transplanting. Some of these interventions have induced changes, but
most were reversed once the intervention force was eliminated. The plain fact
remains that women work longer hours than men and their tasks cover a wider
spectrum. This has created impedcments to productivity related to time
constraints and inadequate productive inputs .
Timc-budget studies have indicated that fetching water for household and
animals and arranging and transporting fuel for cooking are two of the most
time-consuming jobs. Other transportation tasks, such as carrying crops/
seeds/fertilizer to and from fields and processing tasks like hulling, winnow-
ing, threshing, and pounding are also very time consuming. These studies
generally have confirmed that women who find ways to ease time constraints
use the time gained for income generation or for rest and care of family.
African households. Because husband's and wife's incomes are not integrat-
ed, African households arc not gender-neutral for any development effort. A
wife's contribution to labor is affected by the income it is to accrue. She will
tend to devote more labor to activities that can give her income, so that
developmental incentives must be geared to her separately.
Women, especially among the poor, play larger roles as family provisioncrs
and contribute a higher proportion to family subsistence than do men.


Journal for Farming Systems Research-Extension






GENDER ISSUES IN FARMING, PART II


Working with women. Studies have confirmed that:
Working with women in groups gives the advantages of lower costs of
extension and greater cultural acceptability, particularly if the extension agent
is a male.
African market women are known for their organizational skills and
strategies and play a major role in internal distribution of agricultural
commodities. Women tend to market surplus from their own account
production, although there may be regional variations.
Income-generation projects often are directed toward nonfarm craft
operations that deny women their rights as principal food producers; the crafts
have barely adequate markets.
New agricultural technologies often intensify the labor of women, as
these are not introduced into tasks done primarily by women.
Pastoral women, although of key significance in Africa in milk production
and processing, have been relegated to a comparatively low position in any
developmental effort.
Attention to African women farmers' needs, constraints, and incentives
may be the single most cost-effective approach to alleviating African food
crises.

The Issue
Accepting that women's roles are crucial in agricultural production alters
our perspective. What ties women to various unavoidable tasks are the long
hours they have to put in to cover such tasks. If something is to be done to
reduce these long hours of labor, women need to be given adequate farm
tools. This does not undermine the need for other inputs, like credit, fertilizer,
etc., but the most important inputs for reducing women's drudgery are tools.
Even the credit issue is not unconnected to procurement of farm tools,
equipment, and other inputs.

Focus of the Study
This study reports some attempts to develop and introduce appropriate
technologies for women under a financial grant from the Ford Foundation.
The work is being carried out at the Institute for Agricultural Research (IAR),
Samaru, Nigeria, which is one of the National Research Institutions of Nigeria
and is attached to Ahmadu Bello University.
IAR has several interdisciplinary research programs, including the Agricul-
tural Mechanization Program that is focused on the evaluation, development,


Vol. 3, No. 2, 1993






KAUL


and management aspects of agricultural equipment, especially for the small-
scale sector. Several items of equipment already have been developed using
local material and technical know-how (Kaul, 1987). The Institute currently
has mandate for five northern states of Nigeria: Kaduna, Kano, Katsina,
Bauchi, and Sokoto.
The work reported here covers the following aspects of our efforts to reach
women farmers:
Understanding the nature of women's groups; in this connection, a
survey (Atala and Tarfa, 1991) was conducted in Kaduna to find the pattern
of such groupings.
Selecting tasks done exclusively by women in order to understand the
nature of constraints on those tasks and to undertake development work for
easing problems. The following tasks were chosen:
cereal threshing, particularly maize, sorghum, and millet;
food vending (Atala et al., 1991);
milk processing (Ali et al., 1991) and related activities done by Fulani
women, who are the only local source of milk and water transportation
(Adeoti and Kaul, 1991).
Understanding the problems of pepper-mills (Asota and Kaul, 1991),
which are used for grinding pepper, vegetables, and a few cereals, and which
are operated exclusively by women.
Both structured and RRA procedures were used. Equipment was selected
from among the prototypes already developed at IAR(Kaul, 1987) and some
from new or adopted designs. Selected women's groups tried the equipment
prototypes. Based on feedback, the equipment was modified if possible, and
then retried with the same groups. This process was continued until the
groups were satisfied with performance. Frequent follow-up visits continue
to be made to ensure no operational problems.


FINDINGS


Interaction with Women's Groups
In recent years Nigeria has aroused considerable awareness of women
through its internationally recognized Better Life Program (BLP), which is
personally directed by the First Lady. This BLP, together with other women's
organizations like the National Council of Women Societies (NCWS), have
had a positive effect.


Journal for Farming Systems Research-Extension






GENDER ISSUES IN FARMING, PART II


In a survey involving all Local GovernmentAreas (LGAs) of the state, it was
revealed that women claim to have organized themselves into 7300 groups.
Of these, only 300 (4 percent) were found to have been registered as
development groups or associations with either NCWS or BLP. These groups
are for specific tasks or a combination of tasks (Multipurpose Cooperative
Groups). Practically all activities were covered by one group or another,
involving major crop and livestock farming and other conventional activities,
such as trading in clothing, food, and fuel. It was obvious that the BLP and
NCWS have, in fact, made an impact and that women are receptive to new and
improved technologies. Their most frequently reported problems were lack
of modern tools and equipment (e.g., threshers, dehuskers, weeders, and so
on), lack ofaccess to modern inputs (e.g., fertilizer and seed), and poor market
outlets.
This study concentrated in two states: Kaduna and Kano. Rapport was
established with the directorates of the BLP and NCWS, and their chief
executives were brought in for discussions and demonstrations of some of the
available prototype equipment designed and constructed locally under the
institute's agricultural mechanization program. This was followed by inviting
women leaders of relatively dynamic groups from various Local Government
Areas of the two states. More than 100 such leaders came for discussions and
demonstrations. They also made suggestions on likely problems with some
prototypes and identified need for some new ones. This process set the farmer
collaborative work in motion. The LGAs are the official grassroots-level
organs of federal and state administration, and all have officers elected by a
democratic poll. Each LGA has arrangements with Directorates ofWomen for
liaison on women's affairs.
In each state two or three LGAs were chosen for initial attention, such as
Soba, Igabi (Rigachukun) in Kaduna, and Danbatta and Tudunwada in Kano.
A complete baseline survey was done at each of these LGAs to establish the
existing state of the art; this is documented in the Pre-Technology Introduc-
tion Baseline Survey ofWomen's Technology in Kaduna and Kano States, an
internal report.

Introduction of Threshers
A multicrop thresher, developed earlier under the agricultural mechaniza-
tion research program, was chosen for adoption by women's groups, based on
the favorable response received from visiting women farmers and discussions


Vol. 3, No. 2, 1993





KAUL


with them. The machine employs a peg-type cylinder and concave arrange-
ment for threshing, whereas sieving is done by a combination of sieves and
blower. The machine is designed to thresh sorghum, wheat, and millet and
to give output in the range of 60-80 kg/hour, depending on crop and variety.
The first machine was installed and introduced at Danbatta. This place has
a history of existing women's groups who do manual threshing of millet and
sorghum in an organized manner. Eight to ten women constitute a subgroup,
which takes the unthreshed crop from customers against a flat rate of N=2.00
or N=3.00 per bag (100kg grain).3 The subgroup further divides the crop
between two or three women, who pound it against a fixed stone. The
threshed grain mixture is winnowed by traditional dependence on wind. The
unthreshed heads repeatedly are pounded and winnowed, until most of the
grains are out.
The thresher operation was explained and three or four subgroups volun-
teered to try it.
Machine modifications based on user feedback. Once the machines were
operational, the women gave the following feedback, which was used to
improve the machines.
1. The petrol engine was difficult to start. The women had difficulty in
starting the petrol engine, owing to flooding and ignition problems and to
overheating caused by continuous use.
Although the engine was a new one and our technicians explained its care,
after a few days of operation the women gave up and concentrated on another
machine that was run by a diesel engine; this machine's working hours
increased to almost 12 hours daily. In order not to dampen the women's
interest, the second machine was installed with financial support from the
LGA.
2. Grain crackage. The millet available in the area apparently is of many
varieties, and customers even bring in for threshing two-year-old stock, which
often is damaged by insects. It became a problem to convince women that part
of the crackage is due to insect infestation. Despite machine adjustments,
grain crackage persisted. Ultimately, we tried putting hard rubber tubes on
beater tips, a successful strategy. The crackage was negligible. The rubber tips
had to be tightened on the beater bar after about 100 hours of operation; to
improve access to the thresher cylinder, the entire cylinder cover was put on


3N= Naira, approximately US$0.10 at the December, 1991 rate of exchange.


Journal for Farming Systems Research-Extension






GENDER ISSUES IN FARMING, PART II


hinges on one side (instead of bolts on both sides), which made it possible to
check the rubber tubes' condition easily.
It was a great problem to strike a balance between acceptable grain crackage
and unthreshed heads.
3. Low output. The women also complained that the output from the
machine was relatively low for the volume of their crop. Studies were done
to determine the amount of time spent by women using traditional methods
and using the machine (Table 1). Output using the machine was five to seven
times greater than with traditional methods. We disconnected the sieving


Table 1. Comparative Threshing Figures on Millet and Sorghum at Dambatta, Kano.

Method Average Time Time
No. wt. of for Time for Total
in crop in thresh- for winnow- Total grain
group operation ing resting ing time collected Output
(kg) (mts) (mts) (mts) (mts) (kg) (kg/hr)


CROP: SORGHUM
Pounding
pestlec/
mortar) 2


Beating
against
stone


23 35 3 4 42 7 10


2 9.5 15 6 3.5 24.5 6.5 15.9


Mechanical
thresher 2


54 15


20 35 42 72


CROP: MILLET
Pounding
(pestle/
motar) 2
2nd Pounding
Beating
against
stone 2
2nd Pounding
Mechanical
thresher 2


43 97
7


39 66
9

33 9


25 15
7


148 25
4


7 17 102 22 15.9
S 3 5

7 16 18 67.5


Supplemental
(manual)
threshing 10 4 14 5 21.4

Subtotal 19 11 30 23 46

a Leads not threshed by machine were pounded manually.


Vol. 3, No. 2, 1993






KAUL


mechanism, which was the main factor for reduced output, and set the
machine only to thresh the crop. The heavy fly wheel also was removed, and
this brought down the power need to about 2kw capacity. For sieving, we
provided a manual sieve and it was possible to shake out the unthreshed heads
in a relatively short time (Table 1). The unthreshed heads then were manually
pounded; the time taken and output from the second threshing is shown
under "supplemental manual time" in the same table. Total grain output also
is recorded in Table 1 and reveals the potential of the machine in-relation to
earlier use. Apparently, the real demand is for easing the actual threshing
operation, while the manual winnowing is relatively easy.and fast.
The women's groups now are using the machines almost 12 to 14 hours
daily. Their feedback on machine convenience indicates that it is far more
comfortable and does not cause eye problems as frequently as with the
conventional method. Also, the final samples from the machine are stone- and
sand-free, which is not true of traditional pounding or beating. We are
attempting to develop a simple winnower/grain cleaner to supplement the
work of the thresher.
Women's participation gave us a far greater insight into machine develop-
ment problems and issues, and partly answered the question of why these
threshers were not getting the attention of users: we were targeting male
farmers.

Introduction of an Improved Tool for Butter Extraction from Milk
Most of the local production of milk in Nigeria is done by Fulanis. A survey
of milk processing and other activities among Fulanis revealed that activities
ranging from milk collection to its processing and marketing are done by
Fulani women, who also must do routine household tasks and the fetching of
water and firewood. Most spent about 17 hours a day on these tasks. It was
further observed that the most time-consuming activities were marketing,
including carrying of milk and milk products in weights ranging from 5 to 20
kg and covering distances up to 10 km, and water fetching. Churning of milk
(to get butter) was listed as most fatiguing, involving shaking of soured milk
in traditional gourds for 10 to 30 minutes, depending on the volume of milk
in the gourd. Thus butter extraction is the most energy-demanding and time-
consuming task in traditional systems. Under the existing method, women
shake small quantities of milk in a specially shaped gourd, in an up-down-side
swing motion, for up to 20 or 25 minutes. Consequently, someone having


Journal for Farming Systems Research-Extension






GENDER ISSUES IN FARMING, PART II


more than 2 liters of milk to churn will have to repeat the process several times,
each operation requiring effort and time.
A simple churn was developed using a clay pot and a wooden spindle, with
a rotor on one end and supported on a bearing, which is turned by twisting
a rope. Parameters for height, grip size, etc., were based on best available
anthropometric information on women. Since all Fulanis have a simple thatch
and mud house, it was difficult to find a strong existing pole in the house to
support the device. Consequently, a separate stand became necessary,
especially to ensure some portability (and especially during rains). Effort was
made to use available construction blocks. Adequate training was given to
users and the churn was installed at about ten sites. Simultaneously, in
collaboration with the National Livestock Project Department (NLPD) of the
federal Ministry of Agriculture, two churns were installed at each of three
locations chosen by NLPD. NLPD has a scheme to settle the Fulanis on large
hectarage of acquired grass reserves, with the ultimate aim of using them as a
system for supplying the milk and meat needs of the entire country.
Userfeedback. After the churn was introduced, we started making improve-
ments on it based on user feedback, including:
1. Use of claypot. It was only after setting up the units that we learned
that Fulanis have a superstition against using a clay pot in the house. This,
together with the discoloration it induced in milk, led to rejection of the
churn.
It was decided to revert to use of the traditional calabash and the churn was
fitted in it. We later introduced plastic buckets, both because the cost of
calabash is quite high and because it is difficult to find two calabashes of
identical size and shape.
2. Fixing device. The cement block used to hold the churn was observed
to cause problems and detaching the churn rotor from its stand was time-
consuming. These problems were corrected by having a wider base, which can
be pegged to any soft floor, and by having the rotor and shaft as easily
detachable units.
3. Size of churn. Users suggested having a choice of two models-one
for small quantities and another for larger quantities-depending on the
available milk ofa particular family. Two units, of approximately six- and ten-
liter pot capacity, are now in use.
Feedback from the larger groups at grass reserves has been encouraging,
especially as they appear to be able to save time for other activities, which are
now being monitored. A nucleus cooperative introduced by NLPD at one of


Vol. 3, No. 2, 1993





KAUL


Table 2. IAR-Improved Milk Churn Performance with Users.
Average liters Previous average
Length of milk per week liters milk per
Number of usage processed week processed
Location units (weeks) with chur with gourd
Kachia
(Kaduna) 2 38 120 Nominal
Sarbon gari
(Jos) 2 22 210 14-42a
a Depending on ownership of cows.

the reserves (Kachia in Kaduna) revealed that, with the ease of churning, the
women have pooled milk and up to 15 liters are churned at one time. The
butter and byproducts (like Nono4) also are marketed as an activity of the
cooperative, which was not previously done on this scale or as regularly. The
NLPD also has introduced the churn at similar cooperatives at Jos and Bobi
(Niger), which are now being monitored.
Table 2 gives the preliminary data obtained from Fulani groups at the two
cooperative sites.
Some of the constraints experienced in milk processing are related to lack
of organized marketing strategies and to storage of milk and milk products,
which is receiving attention. The only operational problems reported with the
churn pertain to wearing out of the rope and leakages around the pot cover.
This has since been rectified, using better materials and fittings.

Development of a Hand-Pushed Water Carrier
Head porterage is the main vehicle of water conveyance in the area of study,
which was located in Soba and Igabi LGA of Kaduna. It was postulated that
if more water could be transported with the same human effort, it would
greatly ease the task. Table 3 gives some parameters that were useful in
developing an alternative.
A conventional wheelbarrow principle was adapted to develop a pushcart.
The emphasis was on lightness to suit women, a capacity of up to 80 liters,
access to narrow tracks, and case of fabrication, as well as limited versatility to
transport other materials. Three such devices were given to farmer women to
try.

4 Nono is a Hausa word for sour milk, which also can be mixed with millet to yield Fura, another
popular dish.


Journal for Farming Systems Research-Extension






GENDER ISSUES IN FARMING, PART II


Table 3. Pattern of Supply and Usage of Water in Typical Households (Average Val-
ues for the Three Villages).'

Parameters Survey locations Overall

Rigachukun Soba Biye Average values
Household size
(No. of persons) 14 10 9 11
Daily water usages (liters)
i) for cooking
and drinking 58 43 41 47
ii) For washing
cooking utensils 40 31 28 33
iii) For washing clothes 322 89 77 163
iv) For bathing 199 107 88 131
v) Other uses
(animals, etc.) 31 19 10 20
Total per household (liters) 650 288 244 294
Estimated water usage per
person per day (liters) 45 28 26 36
Water source well & tap well well well
Distance of water source
from dwelling (km) 0.3 1.4 1.5 1.0
Size of container for
fetching water (liters) 19 24 25 23
Estimated required trip
frequencies (numbers) 34 12 10 17
Approximate trips needed
with water carrier 8 3 2 4
a Full details in Adeoti and Kaul (1991).

User feedback. The user reaction was quite positive and some changes were
possible based on their feedback, including:
When not transporting water, the cart could be adjusted to take other
materials, such as fertilizer.
Owing to vibration, the unit needed frequent tightening. Use of spring
washers, instead of flat ones, has improved performance.
The device is being accepted, as is evident from the requests pouring in
for more information on its pricing and availability.


Problems of Foodsellers
One of the visible activities of women in all towns and suburbs is
foodsclling, which includes some form of food processing and cooking.
Although usually unrecognized as an organized group, foodsellers serve a vital
need by providing ready-to-eat snacks and full meals. In a way, they act as a


Vol. 3, No. 2, 1993





KAUL


nerve center where trading goes on. The activity involves women and a sizable
number of children.
A structured questionnaire (Atala et al., 1991) directed to women food-
sellers in Zaria township ofKaduna state was intended to study the foodsellers
as a class and to study their operations with a view to understanding their
socioeconomic and technological constraints.
The survey revealed that their major problems included lack of capital, the
high cost of raw materials (food items) and fuel, and inefficient tools and
equipment. It was interesting to note that 44.5 percent of respondents rated
a simple tool like a knife to be problematic, owing to its frequent breakage,
bluntness, or even easily getting lost. The pestle, because of its heaviness and
the tiredness it causes, and cooking stoves, because of excessive fuel needs, also
were rated high problematic areas.
The nature of problems with tools is being studied in its totality to look for
alternatives. One ofthe main aspects being studied for improvement presently
is ways of reducing fuel consumption, which, apart from being expensive, is
related to degradation of the environment through felling of trees.
Two approaches have been taken. One is to adopt a stove that can replace,
or at least supplement, the "three-stone" method, and another is to introduce
devices that reduce cooking time and are strong. Again, by working closely
with actual sellers, an attempt is being made to take the variety of stoves
currently being used (other than the three-stone one) and to select the best
qualities of each and combine them in one stove. Simultaneously, devices like
the Moin-Moin steamer, which already is available in the market, is being
studied to replace the existing steaming process in the hope that it will induce
overall efficiency.

Other Prototypes in Development
Several other prototypes are in development and are intended to be tested
and modified in close collaboration with women's groups. Some of the
prototypes include: a maize dehusker-sheller, a solar drier for vegetables, an
improved pepper grinder mill, a groundnut oil extractor, and a sugarcane
crusher. These will be reported in a subsequent paper.


Journal for Farming Systems Research-Extension





GENDER ISSUES IN FARMING, PART II


TECHNOLOGY IN FARMING SYSTEMS RESEARCH-
EXTENSION

One of the focal points in farming systems research is the idea that farm
operations are to be taken as a system in which the contribution of each
component and its interdependence with other components is recognized.
The role of technologies is a common denominator in practically all operations
and can be a limitation in an otherwise well-conceived project to upgrade
existing practices. Increase in yields attributed to higher planting population,
better weeding, timely spraying, harvesting, and threshing, and transportation
are all indirectly dependent on appropriate mechanical and other technolo-
gies. This paper has highlighted the need for focusing on this aspect in a
farming systems research program and has further illustrated the manner of
evolving some of the equipment by closely working with actual users. In
particular, it has brought to light the need to develop equipment specific to
women, who contribute more than 60 percent of food production and nearly
100 percent of food processing, and who have a distinctive role in the food
production chain.

Working with Women
The author has been associated with equipment development and adoption
for more than three decades in Nigeria and elsewhere. One of the things he
has learned is that the gender issue is a very crucial aspect in any development
plans in Nigeria and in most countries in Africa. There has been a perhaps
unintentional undermining of the role of women in farming, primarily for lack
of knowledge. One could not, therefore, agree more with Stamp (1990) and
Blumberg (1989), who have brought to light the main limiting constraints in
African agriculture, especially technological limitations.
We are among the recent converts to this basic truth, as revealed through
our efforts in introducing equipment to female small-scale farmers. The big
question, however, is whether those in policy planning can be made to
appreciate the role of women.


Vol. 3, No. 2, 1993






KAUL


REFERENCES

Adeoti, J.S., and RN. Kaul. 1991. Development and testing of a hand pushed water
carrier. In Research Report on lAR-Ford Foundation Project on Technology for Women
(IAR Project G.II.8.16.3), July.
Ali, A.M., RN. Kaul, and T.K. Atala. 1991. Milk processing and other activities among
Fulani women: A survey of Zaria Area (Kaduna State). In Research Report on IAR-
Ford Foundation Project on Technology for Women (IAR Project G.II.8.16.3), June.
Asota, C.N., and RN. Kaul. 1991. Pepper-mills: A sample survey of their types and
operational problems. In Research Report on IAR-Ford Foundation Project on
Technologyfor Women (IAR Project G.II.8.16.3), July.
Atala, T.K., RN. Kaul, M.A. Ali, and S.B. Tarfa. A sample survey of women food sellers
in Zaria township, Kaduna State. In Research Report on IAR-Ford Foundation Project
on Technology for Women (IAR Project G.II.8.16.3), May.
Atala, T.K.,and S.B. Tarfa. 1991. A surveyofwomen groups in Kaduna state. In Research
Report on IAR-Ford Foundation Project on Technology for Women (IAR Project G.
11.8.16.3), May.
Blumberg, RL. 1989. Making the casefor thegender variable: Women and the wealth and
well-being of nations. Office of Women in Development, AID Report No. 1.
Washington, D.C.
Carloni, A.S. 1987. Women in development: AID's experience, 1973-1985. Evaluation
Report 18(1).
Kaul, R.N. 1987. Agricultural mechanization as a limiting physical technological
constraint in agricultural production in the semi-arid region of Nigeria. Proceedings,
Symposium on Food Production in the Semi-Arid Regions of Sub-Saharan Africa,
Nairobi, Kenya, May 19-23, 1986.
Kaul, R-N., and A. Ali. 1992. Gender issues in farming, part I: A case for developing
farm tools for women. Journal for Farming Systems Research-Extension 3(1):35-46.
Saito, K.A. 1990. Proceedings, Symposium on Household Food Security and the Role of
Women, Harare, Zimbabwe, January 21-24. Washington, D.C.: The World Bank.
Stamp, P. 1990. Technology, gender, and power in Africa. Technical Study 63e,
International Development Research Centre, Canada.
World Bank. 1988. Women in development: A review of selected economic and sector
reports, 1980-87. Washington, D.C.: The World Bank.


Journal for Farming Systems Research-Extension









Informal Agricultural Communication

Patterns in a Remote Area of Bangladesh'

S.M.A. Hossain, B.R. Crouch, and Shankariah Chamala2



ABSTRACT
This study investigates the communication channels through which
information on two agricultural practices is disseminated in a remote area
of Bangladesh. Two major farming practices, high-yielding varieties
(HYV) of rice cultivation and treatment of sick cattle, were examined.
The second research question investigates the nature (monomorphic vs.
polymorphic) of opinion leadership. Seventy-five farmers were inter-
viewed for this study in six neighboring remote villages of Bangladesh.
The results indicate the existence of two contrasting communication
patterns. The information-seeking pattern in the case of HYV cultivation
was characterized by a large number ofopinion leaders being available for
consultations with other knowledgeable farmers outside the sample
villages, and by a multistep interpersonal communication pattern. In the
case ofsick cattle treatment, only three local traditional veterinary doctors
were identified as key opinion leaders. Unlike HYV rice cultivation, the
communication process in the treatment of sick cattle was direct.
Opinion leaders consulted regarding HYV rice cultivation were different
from those consulted regarding sick cattle treatment. Finally, the holistic
nature of this sociometric study demonstrated the breadth and scope for
extension work. The potential usefulness ofindigenous-based, appropri-
ate, applicable, and acceptable technology is also outline.


INTRODUCTION

In agriculture, communication between farmers is a major process by which
information concerning new and improved farm practices is spread. It
produces greater effects than any other medium in terms of knowledge gains,
attitude formation and changes, and overt behavioral change. Such informal
interpersonal relations in a rural community act both as sources of social
1 A version of this paper was presented at the Australian Sociological Conference, The University
of Queensland, Brisbane, December 12-16, 1990.
2 Department of Agriculture, University of Queensland, Brisbane, Australia, currently on leave
from Bangladesh Agricultural University, Mymensingh; Rural Development Consultant, Bris-
bane, Australia; and Department of Agriculture, University of Queensland, Brisbane, Australia.





HOSSAIN ET AL.


pressure and of social support, and each relates interpersonal relations to
decision-making and sustainable agricultural development (Chamala, 1987).
As a result of the interaction of two or more persons and the needs of the
group, informal influentials emerge (Gibb, 1969). Studies of farmers' sources
of information have shown the importance of personal sources (knowledge-
able persons) as opinion leaders, a position that is earned and maintained by
an individual's ready social accessibility, adherence to appropriate social
norms, and a relatively high degree of technical competence (Castillo, 1979;
Adams, 1982).
Communication channels that operate informally among farmers should
not, however, be regarded as completely distinct and separate from the more
formal, institutionalized, and often impersonal channels that enter the farm-
ing system from outside. Rather, within this influence structure, information
is distilled and edited through the "gatekeeper" effect. That is, a multistep
pattern of interpersonal influence prevails (Rogers, 1983).
In less-developed countries, where little or no mass media are available,
such informal interpersonal channels of information remain the single most
reliable and important source of information (Lingamneni, 1981). Such
interpersonal relations in traditional societies are multidimensional rather
than unidimensional, the latter being more typical of developed societies
(King, 1985).
Specifically, communication of information on HYV rice cultivation and
sick cattle treatment is of crucial significance for the farmers of Bangladesh.
The importance of these two farm practices warrants some elaboration. Rice
production is central to the agricultural economy of Bangladesh. In recent
years, nearly 80 percent of the cultivated area has been planted with rice. Just
over a third of the total rice area is allocated to HYVs (BBS, 1990).
The importance of cattle in Bangladesh agriculture can hardly be overem-
phasized, as it is the primary source of draught power. According to the latest
agricultural census, about 90 percent of working cattle are used for cultiva-
tion. Cattle also are an important source of animal protein and dairy products
(BBS, 1990). Cattle also provide important supplementary income in many
households (Alauddin and Tisdell, 1991).
This paper analyzes the spread of information through informal interper-
sonal communications about these two main agricultural practices. Further,
it identifies the degree ofopinion leaders' influence in the farming community
and investigates the use pattern of radio and selected print media by farmers.


Journal for Farming Systems Research-Extension






AGRICULTURAL COMMUNICATION PATTERNS


STUDY LOCALE

The field investigation was carried out in the Atigram Union3 of Manikganj
district, Bangladesh (Map 1). The union under study was situated in a
relatively remote area 29 km from extension centers. Access to the study area
is difficult; the only road is narrow and interrupted by several canals, pools, and
a river. The area is low-lying and prone to flooding during the wet season,
during which transportation is readily disrupted. Otherwise, the study area
was characteristic of other regions of the country in terms of physical,
demographic, and socioeconomic conditions.


METHODOLOGY

Northway (1967) defined a sociometric test as "a means for determining the
degree to which individuals are accepted in a group, for discovering the
relationships which exist among these individuals, and for disclosing the
structure of the group itself." To accommodate the sociometric requirements
of this study, six neighboring villages were selected; their names are not given
in this paper.
Seventy-five farmers were selected by the sociometric technique of snow-
ball sampling (Rogers and Kincaid, 1981; Knoke and Kuklinski, 1982) to
identify agricultural opinion leaders. Two areas of farming enterprises, HYV
rice cultivation and cattle raising, were chosen as those on which farmers
would seek information. Although these two farming enterprises are different
in nature, they represent the predominant crop-livestock polyculture system
in Bangladeshi farming.
Each respondent was asked for the names of two farmers he considered
worth talking to about HYV rice cultivation and for the name of one
considered a preferential consultant for sick cattle treatment.4
Sociomatrices were prepared. At least three incoming choices were
considered to indicate an opinion leader. Northway's (1967) target socio-
gram was used as a basis for plotting the communication pattern, but some
modifications were made for reporting the present study. In making a target
sociogram, the original concept of drawing concentric circles was followed.

3 A union is a local government unit covering an approximate area of 20 mi2 and a population of
20,400 (BBS, 1989).
4 Pretesting disclosed that it was not feasible for farmers to provide two names in relation to sick
cattle treatment.


Vol. 3, No. 2, 1993





HOSSAIN ET AL.


15 75 Km.
0.~


BAY OF BENGAL


International boundary
District boundary BURMA
o Capital city
Study area


Map 1. Location of Field Investigation, Atigram Union of
Manikgani District, Bangladesh.


On the basis of the rank order of the respondents, which is based on the total
score of the respondents on an issue, concentric circles are drawn as numbers
of ranks plus one. For example, if there are six rank orders of the respondents,
then seven concentric circles are drawn. Except for the innermost circle, which
may be kept for any legend information, the respondents with relatively higher
sociometric scores falling under relatively higher order of ranks are found near


Journal for Farming Systems Research-Extension


Legends





AGRICULTURAL COMMUNICATION PATTERNS


the inner circles. Those who received no choices are found at the outermost
circumference of the circles.
Extent of exposure to mass media was measured by including one electron-
ic medium (radio) and six print media (daily newspaper, agricultural maga-
zines, circular letters, leaflets, bulletins, and newsletters). This exposure was
measured by the numbers of times a farmer listened to agricultural programs
on the radio and/or read selected printed materials. Farmers responded by
checking one of the following categories: "frequently," "occasionally,"
"rarely," and "never." The weighting of scores ranged from 4 for "frequent-
ly" to 1 for "never."


RESULTS


Sociometry of Rice Cultivation
Respondents chosen by other farmers for consultation on any aspect of
HYV rice cultivation were regarded as rice cultivation opinion leaders. Total
score for each nominated farmer was calculated on the basis of the number of
preferences he received. These are arranged in order of their score in Table 1.
Key communicators were considered to be those farmers with greater than
the arithmetic mean of the cumulative sociometric score, that is, 3.24. Those
with less than the mean score were called communicators. Those never
nominated (zero score) were named noncommunicators. These three cate-
gories were denoted by a circle, a hexagon, and a rectangle; numbers of
respondents falling into these three categories were 14, 23, and 38, respective-
ly.
First choice. Figure 1 shows the key communicators, communicators, and
noncommunicators in the respective villages and their positions.
It is evident from the sociogram (Figure 1) that a large number (ten) of
farmers in Village 5 (V5) perceived farmer No. 55 as the best farmer to consult.
Clearly, the consultation pattern (in all six villages surveyed) indicates that an
influential person in a small group of farmers is connected with another group.
In these small groups at least one influential member is connected with a
farmer outside the study area. This shows that opinion leadership choices cut
across village boundaries. Such external systems of contacts have generated
effective intersystem linkages for maximizing informational entropy for the
receiver village system. Havelock (1986) portrayed the internal dynamism of
such communication systems as one in which the end desired state is seen as


Vol. 3, No. 2, 1993






IIOSSAIN ET AL.


Table 1. Sociometric Scores of the Respondents for HYV Rice Cultivation.
Position Rank order Farmer no. Total score
Key communicators 1 55 13
2 1 6
3 13,29,34,42,66 5
4 14,18,24,35,48, 4
56,69
Communicators 5 2,3,4,5,25,30, 3
39,40,43,54,64,65
6 63 2
7 6,17,26,33,41, 1
50,58,62,73,75
Mean=3.24


the "needs" of knowledge of the user village (sampled) system while "solu-
tions" are provided by the resource village as part of the external system.
These intersystem connections between groups or village systems could be
called "weak ties" (Granovetter, 1973) by stressing their strength and
importance. Granovetter's term expresses the degree to which an interperson-
al communication network overlaps with low communication proximity.
Weak ties are considered the major basis of intergroup connectivity and
account for informational flow about activities outside any organized system
(Friedkin, 1982). It also is observable in Figure 1 that Farmers 1, 29, and 55,
being the most influential, are consulted by the other subgroup leaders. This
indicates the presence of "social distance" between them and the other
farmers. In each village there is an influential gatekeeper in each subgroup of
farmers. The multistep process of communication provides apriori reasoning
for delayed diffusion and adoption of farm practices. Havelock (1986)
demonstrated another reason: Within a system, procedural configuration
(i.e., the internal flow pattern that dictates how knowledge and other material
is transferred and transformed within the system) is kept in place by a
prevailing ideology, or a long-standing tradition, or both. However, mainte-
nance of important procedural configurations and, most particularly, the
introduction of new ones, also depends upon opinion leaders who, in some
sense, embody the idea of the system and personify its procedural configura-
tion. These leaders act as subjective system boundary membranes; that is,
through their selective perception they allow relevant and necessary things to
enter the system and provide barriers to the entry of others. This kind of
selective permeability preserves the integrity of systems, blocks invasion by


Journal for Farming Systems Research-Extension






AGRICULTURAL COMMUNICATION PATTERNS


V4


44 45 46 4







54 2 2 8 0
60 1 8






Legends51 63 64 Farmer absent/








Q Key communicator
~ Commun.catoo V3
t NonItcommud ntor 1









--- Village boundary
VI. V2, V3, V4, V5, and V6 are anonymous
names of the villages surveyed

Figure 1. Key Communicators, Communicators, and Noncommunicators
in otherSix Villages.













other systems and dangerous environmental elements, and ensures survival
through controlled inflow ofinformation. However, when the idea penetrates
into the group system, the group procedural configuration rapidly increases
the extent of diffusion and intensity of adoption. The presence of smaller
group leaders in each village also supports the theory that in a stratified
community each stratum has its own opinion leaders, rather than a few such
persons performing this role for the whole community (Rd1ing et al., 1981).
74 73 9 66 1 % 29 25 33 32

71 %
68 % 24 23




28

Legends 37 36
O Key communicator
OCmunicator V3
ElNon-communicator
-- Village boundary
VI, V2, V3, V4, V5, and V6 are anonymous
names of the villages surveyed

Figure 1. Key Commuficators, Communicators, and Noncommunicators
in Six Villages.

Other systems and dangerous environmental elements, and ensures survival
through controlled inflow ofinformation. However, when the idea penetrates
into the group system, the group procedural configuration rapidly increases
the extent of diffusion and intensity of adoption. The presence of smaller
group leaders in each village also supports the theory that in a stratified
community each stratum has its own opinion leaders, rather than a few such
persons performing this role for the whole community (Rbling et al., 1981).


Vol. 3, No. 2, 1993





HOSSAIN ET AL.


Second choice. The second choice is plotted and presented in Figure 2,
which shows that no one farmer in any of the villages studied is dominating
in the seeker-sought dyad.
It can be seen clearly in Figure 2 that, although the same farmers occupy
the key positions in the second choice, there are some changes in preferences.
Respondents 13, 25,42, 54, 56, and 64 received the most second choices. No
influential farmer indicated any preference for farmers in other villages. It also
is noticeable that in Villages 1,4, and 5 (V1, V4, and V5), mutual preferences
exist among key communicators, between key communicators and commu-


Legends

O Key communicator s
SCommunicator V3
D Non-communicator
--- Village boundary
V1, V2, V3, V4, VS, and V6 are anonymous
names of the villages surveyed

Figure 2. Key Communicators, Communicators, and Noncommunicators
in Six Villages


Journal for Farming Systems Research-Extension





AGRICULTURAL COMMUNICATION PATTERNS


nicators, and among communicators. This verifies close communication links
among subsystems of an internal system. This linkage might be owing to
common interests, such as acquiring more technical knowledge from other
leaders and retaining their positions in the village. Another possible reason for
this linkage of influential farmers could be that the selection of friends is based
on the similarity of their function as farmer opinion leaders with comparable
economic standing. It might also be possible that they cross-check the
credibility of information and share their personal experiences of the innova-
tion. This sort of functional homophily has been reported by Crouch (1970).
Still another reason, this one associated with kinship, cannot be overlooked;
that is, the social relationship of brothers, especially Farmers 1 and 18.
Villages 2 and 6 (V2 and V6) disclose highly interactive farmers' groups,
with each influential person in these individual villages connected with all
other influentials. These community members are enmeshed with tight-link
social relationships. In such a situation, diffusion and adoption of agricultural
practices needs social approval by these influential farmers (West, 1983).
It is evident from Figures 1 and 2 that the consultation pattern for the HYV
rice cultivation is widely diffused in all the six villages surveyed.

Stepped linkage
Keeping in view the first- and second-choice (Figures 1 and 2) networks for
HYV rice cultivation in these villages, one can easily identify the communica-
tion network steps consistent with the multistep flow of communication. The
majority of the influential farmers have links either within the endogenous
system of villages surveyed or with the exogenous system of unsurveyed
villages. This is more pronounced in first-choice than in second-choice
networks. In the case of first-choice, for example, the following network steps
can be identified. Only connections with more than two steps are illustrated.


Step 1 Step 2 Step 3
i) No.1-- > No.3--> No.5-- > outside sample village.

Step 1 Step 2 Step 3 Step 4
ii) No.17-- > No.13- > No.34- > No.39- > outside sample village.

Step 1 Step 2 Step 3 Step 4
iii) No.68- > No.69- > No.73- > No.65- >
Step 5 Step 6 Step 7
> No.64- > No.50--> No.55- >outside sample village.


Vol. 3, No. 2, 1993






HIOSSAIN ET AL.


In the case of second-choice:

Step 1 Step 2 Step 3
i) No.62- > No.64---> No.55 <- > No. 5.

Step 1 Step 2 Step 3 Step 4 Step 5 Step 6
ii) No.7----> No.66---- > No.75- > No.65- >No.69- > No. 66.

Step 1 Step 2 Step 3 Step 4
iii) No.37- > No.40-- > No.63- > No.34- >
Step 5 Step 6 Step 7
> No.13- > No.17- > No.14<-- >No.6.


Thus, the multistep flow of communication linking mass media eitherr
through listening to radio, having minor contact with print media, and/or
contacting other, more cosmopolitan, farmers in other villages) with networks
of interpersonal relations existed. Alternatively, channels like visiting outside
social systems and local interpersonal channels at the village level were
operating for HYV rice cultivation in the villages surveyed. This multistep flow
of information, with filters or gatekeepers at every step, increases the chance
for information to lose fidelity; that is, the larger the communication channels,
the more points or levels along the channel at which someone decides whether
or not the message shall continue to be transmitted in the same form without
further loss in quality and quantity (Sinha and Mehta, 1972). Noting another
reason, Van den Ban (1963) pointed out that the multistep process of
diffusion of innovations in a community occurs especially when farmers are
not actively seeking information on the innovation. Informal interpersonal
communication is their most important source of information.

Sociometry of Sick Cattle Treatment
The ranking of respondents based on the total of their sociometric scores
obtained is shown in Table 2. Three farmers, numbers 6, 12, and 74, were
identified as key communicators, and eleven were identified as communica-
tors: ten with a minimum score of one and one (Farmer 1) with a score of two.
The informal consultation pattern for seeking information for the treat-
ment of sick cattle is depicted in Figure 3; seven respondents (numbers 31,
45,46,52, 53, 56, and 67) who own no cattle are not shown in the sociogram.
Unlike the sociograms on the HYV rice cultivation (Figures 1 and 2), the
sociometric choice pattern for sick cattle treatment seems to be polarized by
a few individuals occupying key communicator positions. In reality, these


Journal for Farming Systems Research-Extension






AGRICULTURAL COMMUNICATION PATTERNS


Table 2. Sociometric Scores of the Respondents for Cattle Husbandry.
Position Rank order Farmer no. Total score
Key communicators 1 6 19
2 74 12
3 12 11
Communicators 4 1 2
5 3,5,34,35,36,43, 1
54,55,58,62
Mean=3.85
three communicators (numbers 6, 12, and 74) are fakirs who treat cattle
diseases by traditional methods. Another 11 respondents received less-than-
average preferences.
In other words, all the sociometric choices were bestowed on only 14
respondents, with a majority of the preferences given to three fakirs. As
modern veterinary treatments are entirely lacking in the surveyed villages, the
traditional methods of curing cattle diseases were very prominent. The
increased pressure on land prevents farmers from considering growing
improved pasture for cattle. So, cattle husbandry in the study area is
concerned mainly with treating sick cattle. Better and scientific ways of animal
husbandry are not yet sought in the study area. Only 14 farmers were
consulted for information on sick cattle treatment and only three of them
occupied key positions, as indicated by hubs appearing as clear spoked-wheel
patterns of communication. Noticeably, those three traditional veterinarians
did not extend their choice links to others, but instead bridged the intervillage
subsystem of information links. Furthermore, of the 53 different links made
to connect three key influentials, the majority (60 percent) had a one-step
connection, 34 percent had two-step connections, and only 6 percent of links
formed three-step connections (Figure 3).

Web of Networks: Observed Patterns
There is a marked contrast in information-seeking patterns for growing of
HYV rice and cattle management. The communication network for informa-
tion on HYV rice production within the union is well developed: 37
respondents in the sample were consulted; 14 of them were key communica-
tors and the rest were communicators. This indicates that innovations in rice
production are widespread and known to more farmers than are treatments
for sick cattle. HYV rice cultivation is not technologically as complex as


Vol. 3, No. 2, 1993






HOSSAIN ET AL.


Vi

87
9
19
17
18


\5 54 29 V2


6 V6








Legends
Q Key communicator
51 23



869 35 3


V6 70 37
Legends V3

O Key communicator
Q Communicator
[ Non-communicator
.. Village boundary
V1, V2. V3, V4, V5, and V6 are anonymous
names of the villages surveyed


Figure 3. Key Communicators, Communicators, and Noncommunicators
in Six Villages.


treating sick cattle (Hossain et al., 1991). This, perhaps, limited the number
of local expert consultants on treating sick cattle to three. Crouch's (1970)
findings were similar; the information-seeking pattern for pasture was better
developed than were the patterns of information-seeking on sheep practices
in his study area in Australia.
This further reflects how farmers behave in seeking information in a
situation in which an adequate sociometric network has not yet developed
among farmers; the alternative is the formal agent. But the agent is not


Journal for Farming Systems Research-Extension






AGRICULTURAL COMMUNICATION PATTERNS


normally used even for subject matter for which the sociometric network is
well established (cf. Figures 1 and 2).

Exposure to Mass Media
Radio. Data presented in Table 3 show that more than one third (34.5
percent) of the opinion leaders listened to farm radio programs with various
degrees of frequency. By contrast, most followers did not use radio (95.7
percent) as a source of farm information.
Although opinion leaders used radio for farm information more than
followers did, the frequency of listening was not high. Little exposure to the
agricultural radio programs may be attributed to lack of ownership of a radio
set. If, however, a farmer has a radio, or access to it, low exposure to programs
could be owing to (a) preoccupation with other work, (b) unsuitability of the
program time, (c) irrelevance of the information in the programs, and thus (d)
lack of interest.
For the last three reasons, Lionberger (1986) blames professionals involved
in the information/knowledge dissemination process. He argues that these
professionals have only limited knowledge of the local situation and of the
informational needs of potential users. If the primary goal of an information
system-matching what is being transferred to what is needed-is not met, the
probability of its use decreases.
Print media. The print media were not used by the vast majority offarmers,
regardless of whether they were opinion leaders or followers (cf. Table 3). This
could be owing to respondents' very low literacy levels (Hossain, 1988).

Role Specificity of Opinion Leadership
In any community there are likely to be many networks based on many
social and functional relationships, because the size of each network is
relatively small. The polymorphism and monomorphism concepts based on
an individual's influence in a given community could be explained in terms of
social and functional networks. Thus, in any rural community no one specific
role dimension for opinion leadership needs to exist.
Table 4 reveals that of the 40 respondents identified as desirable sources of
information on two farming enterprises, the majority (72.5 percent) showed
specific (monomorphic) opinion leadership. This finding contradicts past
research results, which maintained that general (polymorphic) opinion lead-
ership is typical of farming communities in traditional societies or developing
countries (Rogers and Svenning, 1969; Agrawal and Pandcy, 1985). The


Vol. 3 No. 2, 1993






HOSSAIN ET AL.


Table 3. Nature and Extent of Exposure to Mass Media by Opinion Leaders
and Followers.


Medium

Radio


Extent of
exposure

Never
Rarely
Occasionally
Frequently


Daily Never
newspapers Rarely
Occasionally
Frequently

Agricultural Never
magazines Rarely
Occasionally
Frequently


Circular
letters



Leaflets


Opinion
leaders
No. %

19 (65.5)
3 (10.3)
4 (13.9)
3 (10.3)


(96
(3



(9(
(3



(9(




(9




(9
(6


Never
Rarely
Occasionally
Frequently

Never
Rarely
Occasionally
Frequently


Bulletins Never
Rarely
Occasionally
Frequently

Newsletters Never
Rarely
Occasionally
Frequently


(10(


Total 29 (10(


Followers
No. %

44 (95.7)

2 (4.3)


i.6) 43 (93.5)
3.4) -
- 2 (4.3)
- 1 (2.2)

.6) 45 (97.8)
3.4) -

1 (2.2)

.6) 45 (97.8)
3.4)

1 (2.2)

3.1) 44 (95.6)
6.9) 1 (2.2)

1 (2.2)

3.1) 45 (97.8)
S.9) -

- 1 (2.2)

).0) 45 (97.8)


1 (2.2)

).0) 46 (100.0)


present analysis, which is based on limited data, does not allow us to conclude
definitively that such specialized opinion leadership existed in the study area.
It also has been observed that this measure (specific vs. general opinion
leadership) is influenced by the diversity of topics on which opinion leadership
is measured (Rogers, 1983).


Journal for Farming Systems Research-Extension


1 (1.3)
75 (100.0)


--


Total
No. %

63 (84.0)
3 (4.0)
6 (8.0)
3 (4.0)

71 (94.7)
1 (1.3)
2 (2.7)
1 (1.3)

73 (97.4)
1 (1.3)

1 (1.3)

73 (97.4)
1 (1.3)

1 (1.3)

71 (94.7)
3 (4.0)

1 (1.3)

72 (96.0)
2 (2.7)

1 (1.3)

74 (98.7)






AGRICULTURAL COMMUNICATION PATTERNS


Table 4. Opinion Leaders' Role Dimensions Based on Two Farming Situations.
Leaders' role Opinion leaders
dimensions No. %
One 29 (72.5)
Two 11 (27.5)
Total 40 (100.0)

Moreover, both cattle husbandry and HYV rice cultivation involve some
degree of technical complexity. Crouch (1970), for example, equates
specialized opinion leadership with the level of knowledge and the level of
complexity of farm practices.
It is certain that as more innovations are adopted and the farm enterprise
becomes more complex (in either specialization or diversification), it will not
be possible for farmers to retain multiple roles in opinion leadership. It usually
takes 10 to 15 years for opinion leadership to emerge (Crouch, 1970). Thus,
it is neither practical nor possible for a farmer to develop competence in more
than two aspects of an enterprise in his effective working life. The latter point
is of considerable importance. It is logical (and, therefore, not surprising) that
influentials are much older than followers (Hossain, 1988). Time is the key
factor in acquiring knowledge and experience. This is why personal influence
is not confined to the more innovative adopter categories. The influential is
sought wherever he may be in terms of both social system and geography.
Because age is the key factor, farmers are likely to turn to more experienced
farmers, even in other villages.


DISCUSSION


Remote vs. Accessible Farming Systems
In this age of information there is a paradox: An information overload may
exist in one area while an information gap in another. Among others,
accessibility is a dominant reason. Since extension work with client farmers
involves various forms of individual and group contact, physical accessibility
plays a significant role.
Remoteness of the study area from the extension center, together with an
unreliable road transport system, resulted in insignificant contacts with the
official service system. There are many examples of this problem. Dey (1980)
observed a similar phenomenon in West Bengal, India. In another study,


Vol. 3, No. 2, 1993






HOSSAIN ET AL.


Shirazi (1983) found limited contact between Pakistani farmers and agricul-
tural extension officers in relatively remote villages, compared with villages
close to the agricultural extension center. Ogunsanya (1987) summed up the
situation by saying that poor rural road accessibility results in poor human-
resource mobilization for development.
There are differences in these two subsystems (remote vs. accessible) in
terms of reliance on and use patterns of available knowledge. Farmers in
remote locations who experience limited contact with government service
centers develop strong interpersonal linkage systems that enable them to rely
on the procedural configuration of their own internal problem-solving
subsystem. Also, probably owing to the perceived complexity of the technol-
ogy used, these internal knowledge utilization systems compel them to use the
existing design of indigenous wisdom. Furthermore, as this study shows, it
generates an area of specialization (monomorphism) in local opinion leader-
ship.

Improved vs. Indigenous Knowledge System
Ruling (1988: 33) defines an agricultural knowledge system as "a system
of beliefs, cognitions, models, theories, concepts and other products of mind
in which the (vicarious) experience of a person or group with respect to
agricultural production is accumulated." Significantly, different groups of
people have knowledge systems that either are improved (normally induced)
or indigenous.
Outlining the importance of indigenous knowledge systems, Ehrenfeld
(1987) remarked that agrarian subsystems that incorporate them can enhance
the repertoire of farm management for sustainable systems. Tisdell (1983)
showed that indigenous technology gives a lower expected return with a
greater degree of certainty; it also requires no specialized input. Improved
knowledge systems, on the other hand, are fraught with considerable uncer-
tainty determined by objective and subjective factors of ideal conditions that
must be met for them to operate. Stressing the importance of indigenous
knowledge systems for sustainable agricultural development, Beal (1986)
pragmatically argued that they deal with many important issues both social
(decentralization, local participation, peasant involvement, self-sufficiency,
self-determination) and technological (low-cost resources, appropriate tech-
nology developed out of experience and testing, technology representing a
strong potential for special targeting and utilization by broad populations of


Journal for Farming Systems Research-Extension






AGRICULTURAL COMMUNICATION PATTERNS


low-resource entrepreneurs, technology and practices that take into account
household and national goals, and technology and programs seen as interact-
ing well with the physical, biological, and socioeconomic environments).
In the study area, farmers used HYV rice (an improved practice) and
followed indigenous methods to cure cattle diseases, such as invoking
supernatural beings through incantations and using herbal medicines. This
seems an innovative blend of new practices and old ideas within a synergetic
function. However, based on the practical and social systems effects,
respondents found it easier to adopt traditional rather than modern veterinary
medicines, even though the former have little bearing on successful treatment
of cattle diseases. Farmers' acceptance of traditional methods could be owing
to the fact that it is very hard to find a modern veterinary practitioner and that
the cost of such treatment is exorbitantly high. Moreover, the veterinarians
with indigenous knowledge do not charge fees, so that their services are readily
available to farmers (Hossain, 1988).
In this practice there was very little penetration of improved knowledge
systems. For this reason indigenous knowledge systems emerged to meet the
needs of the local agriculturalists.


CONCLUSIONS
The consultation pattern for HYV rice cultivation was more diffuse than that
for sick cattle treatment, the latter being centered on three key communica-
tors. The villages were structured as small groups, some of which were
interlinked. The main characteristic of these groups was the existence of
opinion leadership in each group. In the case of HYV rice cultivation,
interpersonal relations were characterized by the multistep flow of communi-
cation. In the case of sick cattle treatment, however, contacts were made
directly with opinion leaders. On the basis of two different types of farming
enterprises, characterized by the complexity of HYV rice enterprise and the
apparent specialization of sick cattle treatment, it is not possible to decide
whether specialized or general opinion leadership prevailed. It is anticipated
that specialization will dissipate as more farmers become involved with
livestock and that some of these farmers will become opinion leaders in their
own right. This is particularly true as advice sought shifts from the traditional
question of animal disease to that of nutrition, breeding, and use of veterinary
medicine.


Vol. 3, No. 2, 1993






IIOSSAIN ET AL.


Extension workers should make systematic efforts to rediscover indigenous
farm wisdom, in other words, to develop and deliver locally validated
information and incorporate it in the new farming system. This would make
the technology more appropriate, applicable, and acceptable in terms of the
needs and nature of the local farming community.
This study also shows that sociometry is an extremely useful tool and that
it should be used by extension workers. The sociograms in this paper illustrate
their role in the clarification of social choices and relationships within external
and internal village subsystems. The patterns of communication and personal
relationships of the system are shown in a holistic way. The extension worker
should begin to operate indirectly within the total farming system.
Finally, the data in this study suggest that reaching remote client farmers
should be the highest priority of extension workers. In this context, the
evaluation unit of the extension system in Bangladesh should be strengthened
to enable its staff to evaluate the effectiveness of agricultural extension in the
country. Thisshould be done with the intention ofimproving the performance-
-as well as the image-of extension work and workers.


REFERENCES

Adams, M.E. 1982. Agricultural extension in developing countries. Longman.
Agrawal, G.K., and S.S. Pandey. 1985. Opinion leaders in rural communication: A study.
Kurukshetra 33:26-30.
Alauddin, M., and C.A. Tisdcll. 1991. The 'green revolution' and economic development:
The process and its impact in Bangladesh. London: Macmillan.
Bangladesh Bureau of Statistics (BBS). 1989. Statistical pocket hook of Bangladesh 1989.
Dhaka: Bangladesh Bureau of Statistics.
Bangladesh Bureau of Statistics (BBS). 1990. Statistical yearbook of Bangladesh 1990.
Dhaka: Bangladesh Bureau of Statistics.
Beal, G.M. 1986. A user-problem-necd-driven model: Farming systems research and
development. Pages 183-208 in G.M. Beal, W. Dissanayake, and S. Konoshima, eds.,
Knowledge generation, exchange and utilization. London: Westview Press.
Castillo, G.T. 1979. Beyond Manila: Philippine rural problems in perspective. Ottawa:
International Development Research Centre.
Chamala, S. 1987. Strategies to overcome communication deficiencies in achieving
sustainable farming systems. Journal of the Australian Institute of Agricultural
Science 53:164-9.
Crouch, B.R. 1970. Today, tomorrow, never: A sociological study of factors determining
the adoption ofagricultural innovations bywoolgrowers in the Yass River Valley, New
South Wales. Unpublished Ph.D. dissertation, Australian National University,
Canberra.


Journalfor Farming Systems Research-Extension






AGRICULTURAL COMMUNICATION PATTERNS


Dey, P.K. 1980. Differential extension contacts. IndianJournal of Extension Education
14:22-31.
Ehrenfeld, D. 1987. Sustainable agriculture and the challenge of place. Journal of
Alternative Agriculture 2:154-87.
Friedkin, N.E. 1982. Information flow through strong and weak ties in inter-organiza-
tional social networks. Social Networks 3:273-85.
Gibb, C.A. 1969. An interaction view of the emergence of leadership. Pages 214-22 in
C.A. Gibb, ed., Leadership: Selected readings. Penguin Books.
Granovetter, M.S. 1973. The strength of weak ties. American Journal of Sociology
78:1360-80.
Havelock, R.G. 1986. Modelling the knowledge systems. Pages 77-104 in G.M. Beal,
W. Dissanayake, and S. Konoshima, eds., Knowledge generation, exchange and
utilization. London: Westview Press.
Hossain, S.M.A. 1988. Opinion leadership in agriculture of Bangladesh. Unpublished
M.Agr.St. thesis, The University of Queensland, Australia.
Hossain, S.M.A., B.R. Crouch, and S. Chamala. 1991. Economic aspects ofinterpersonal
communication: A choice for extension work. Journal ofExtension Systems (in press).
King, P.G. 1985. Effectiveness of opinion leaders in the diffusion and adoption of
improved farming practices. South African Journal of Agricultural Extension 14:32-
7.
Knoke, D.,and J.K. Kuklinski. 1982. Network analysis. Beverly Hills: Sage Publications.
Lingamneni, J.R. 1981. Communication channels/sources in the innovation-decision
process: A comparative review of U.S. and Indian diffusion research. Pages 147-53
in B.Rl Crouch and S. Chamala, eds., Extension education and rural development,
Vol.1. Chichester: John Wiley and Sons.
Lionberger, H.F. 1986. Toward an idealized systems model for generating and utilizing
information in modernizing societies. Pages 105-34 in G.M. Beal, W. Dissanayake,
and S. Konoshima, eds., Knowledge generation, exchange and utilization. London:
Westview Press.
Northway, M.L. 1967. A primer of sociometry. Toronto: University of Toronto Press.
Ogunsanya, A.A. 1987. Rural accessibility problems and human resource development:
Case study from Nigeria. Journal of Rural Studies 3:31-42.
Rogers, E.M. 1983. Diffusion of innovations, 3rd cd. New York: Free Press.
Rogers, E.M., and D.L. Kincaid. 1981. Communication networks: Towards a new
paradigm for research. New York: Free Press.
Rogers, E.M., and L. Svenning. 1969. Modernization among peasants: The impact of
communication. New York: Holt, Rinehart and Winston.
Ruling, N.R. 1988. Extension science: Information systems in agricultural development.
Cambridge: Cambridge University Press.
Ruling, N.G., J. Ascroft, and F.W. Chege. 1981. The diffusion of innovations and the
issue of equity in rural development. Pages 225-36 in B.R. Crouch and S. Chamala,
eds., Extension education and rural development, Vol.1. Chichester: John Wiley and
Sons.
Shirazi, S.A. 1983. Communication and adoption of innovations in a remote and an
accessible village in Pakistan: A comparative study. Ph.D. dissertation, University of
Queensland, Australia.
Sinha, B.P., and P. Mchta. 1972. Farmers' need for achievement and change-proneness
in acquisition of information from a farm-telecast. Rural Sociology 37:417-27.


Vol. 3, No. 2, 1993






IIOSSAIN ET AL.


Tisdell, C. 1983. Conserving living resources in the Third World Countries: Economic
and social issues. International Journal of Environmental Studies 22:11-24.
Van den Ban, A.W. 1963. The communication of new farm practices in the Netherlands.
Van Gorocum & Co.
West, P.C. 1983. Collective adoption of natural resource practices in developing nations.
Rural Sociology 48:44-9.


Journal for Farming Systems Research-Extension








Farming Systems Research and the
Rural Poor:

A Political Economy Approach

S.D. Biggs and John Farrington '



ABSTRACT
This paper argues that the main constraint to sustained implementation
of FSR has not been a shortage of resources but the prevalence
among many FSR practitioners and planners of "central source"
perspectives, which cast technology development in a hierar-
chical mold and view science as politically and distributionally
neutral. By contrast, a long (but neglected) political economy
tradition stresses that innovations arise from multiple sources
and are conditioned by their historical, institutional, economic,
and. political contexts. Evidence in support of this tradition is
drawn from six case studies: the spread of Mexican wheats, learning by
centers from farmers, integrated pest management, wheeled toolcarriers,
and common properties. Wider adoption of political economy perspec-
tives broadens the basis ofresearch policy, of the philosophy and practice
ofscience, and of policy analysis. It requires cross-learning among social
and natural scientists within and beyond agriculture, and demands of
them a keen sense of personal responsibility.


SCIENTIFIC PHILOSOPHIES: "CENTRAL SOURCE" AND
POLITICAL ECONOMY PARADIGMS

This paper reviews two alternative paradigms of agricultural research and
technology dissemination, drawing out the implications of each for the policy
and practice of research. Its focus is on difficult farming areas, characterized
by combinations of low and erratic rainfall, infertile soils, hilly topography,
and weak infrastructure. The large numbers of rural poor-estimated at one
billion, or approaching 50 percent of the rural populations of developing

''he authors are associated with, respectively, the School of I)evelopment Studies,
University of East Anglia, Norwich NR4 7TJ, UK and the Overseas Ievelopment
Institute, Regents College, Inner Circle, Regents Park, London, NW1 4NS, UK.





BIGGS AND FARRINGTON


countries-and the diversity and complexity of prevailing agroecological and
socioeconomic conditions make these areas prime candidates for systems
approaches to technology improvement. Yet FSR has achieved only limited
success in these areas. Our aim is to review the characteristics of the prevalent
paradigm of research, indicating how it has influenced the implementation of
FSR and suggesting how these areas might be better served by an alternative
approach.
Many projects and programs concerned with FSR in difficult areas have
been influenced by donors' agendas or by farming systems methods developed
at international research centers. The philosophy and practice of these often
have been set in a "central source" (Biggs, 1990) mode of thinking, in which
technological and institutional innovations are created or generated at "cen-
ters" and then transferred to less developed "peripheries." Widespread
interpretation of "successes" or "failures" in terms of this central-source
model has helped to give rise to a set of myths, biases, and misperceptions that
act to reinforce the model.
By contrast, a long political economy tradition sets science and technology
in an historical, political, agroecological, economic, and institutional context.
Fundamental to this tradition is the view that technological change does not
necessarily follow hierarchical patterns but arises from multiple sources. A
further fundamental feature is its explicit recognition of the diversity of
agroecological and socioeconomic conditions in which the poor seek liveli-
hoods. By contrast, the central-source model's tendency to emphasize
generalities or widely adaptable technologies and institutional forms distracts
researchers' attention from the implications of diversity for future research
policy and practice.
The main elements of central- and multiple-source approaches
are summarized in Table 1. Following a review ofthe features and implications
of each approach, we argue from a number of case studies of FSR that the
political economy tradition provides a sounder basis than does "central
source" thinking for understanding the past behavior of research and exten-
sion systems and for directing their future.


Journal for Farming Systems Research-Extension






FSR AND THE RURAL POOR


Table 1. Characteristics of Alternative Approaches to Agricultural Research and
Technology Promotion.

Central-source- Multiple-source-
of-innovation of-innovation
Characteristics approach approach


1. Definition and role
of institutions






2. Stages in technology
research, exten-
sion, and adoption








3. Stages in research
capability develop-
ment


4. Structure of research
and extension system




5. Networks for materials
and information flows


Institutions clearly de-
fined by (1) international,
national, and regional sta-
tus, (2) research or exten-
sion role, and (3)
sometimes a crop or tech-
nology mandate

Emphasis on a linear se-
quential path of stages in
technology generation,
adaptation, and dem-
onstration (a problem-
solving approach)





Emphasis on the
transfer of research
capability from out-
side


Hierarchical and central-
ized. A few standardized
research and extension in-
stitutions


Coordination of informa-
tion and technology in
networks is systematically
performed according to
unambiguous criteria


Ambiguous in regard
to playing an interna-
tional or national role
and performing re-
search or extension
functions


Emphasis on continu-
ous interaction be-
tween researchers,
users, extension
agencies, and continu-
al assessment of the
political, economic,
technological, and in-
stitutional environ-
ments

Recognition of the pri-
or existence of local
R&D and diffusion ca-
pabilities and of how
outside interaction se-
lectively strengthens,
weakens, and im-
proves parts of these

Uniform/lateral/hori-
zontal/decentralized.
A multitude of diverse
research and promo-
tion institutions

Access to and control
over information and
materials in networks
is recognized as se-
lective and determined
by the interests of
different scientific,
political, economic,
and administrative in-
terest groups


Vol. 3, No. 2, 1993






BIGGS AND FARRINGTON


Table 1. Characteristics of Alternative Approaches to Agricultural Research and
Technology Promotion, Continued.


Characteristics


6. Use of data and
knowledge over
time


7. Treatment of time


8. Definition of tech-
nology








9. Sources of innova-
tion


Central-source-
of-innovation
approach

Assumes the system-
atic accumulation,
storage, and use of
knowledge

Minor consideration:
a "timeless" model in
which knowledge is sys-
tematically accumulated
and always used for in-
formed research policy de-
cisions



Frequently in broad,
unambiguous terms
such as packages of
technology or "how-
to" manuals. Defini-
tions and special- izations
of science dominate other
definitions and categoriza-
tions of users

(a) Central sources: Major
emphasis given to innova-
tions coming from formal
designated research insti-
tutions which are trans-
ferred to promotion and
institutional users

(b) Little emphasis on the
unpredictable nature of
innovations


Multiple-source-
of-innovation
approach

Assumes selective ac-
cumulation and selec-
tive use of knowledge
over time.

Critical feature:
whose knowledge
counts and who con-
trols research funds
determines the direc-
tion and content of re-
search, and who gets
the benefits at any
point in time

Emphasis on components
(whether materials, meth-
ods, or institutions),
which are combined local-
ly. Definitions and catego-
rization of science and
users given equal impor-
tance


(a) Multiple sources:
equal emphasis is given to
informal research by farm-
ers and other users, inno-
vations by local research
practitioners, re-
search-minded extension
agencies, administrators,
and others in real-world
situations as to innova-
tions from designated
central research institu-
tions

(b) Major emphasis on
unpredictable sources of
innovation


Journal for Farming Systems Research-Extension






FSR AND TIE RURAL POOR


Table 1. Characteristics of Alternative Approaches to Agricultural Research and
Technology Promotion, Continued.

Central-source- Multiple-source-
of-innovation of-innovation
Characteristics approach approach


9. Sources of innova-
tion (continued)






10. Political funding and
institutional context





11. Central/regional
allocation of research
resources














12. Orientation


(c) Institution innovations
can be created and devel-
oped and transferred from
centers to lower levels in
the system



No or little emphasis in
the model. The direction
of science is seen as (or
should be) independent of
political and other contex-
tual issues

Debate is primarily based
on agroclimatic and tech-
nical economics of scale in
research arguments (e.g.,
by geographically defined
regions, zones, and farms
systems). Pattern of re-
sources between central
and regional station. Pri-
marily based on the need
for regional stations in
different agroclimatic re-
gions to service central
stations or centrally con-
trolled research policy

"Supply push" with cen-
ters generating and pro-
moting good technology


(c) Institutional innova-
tions are, in a sense, al-
ways new, and their use
depends upon the specific
political, economic, tech-
nological, and historical
context of research

Central to the model.
The level and source
of funding determines
the direction and con-
text of research


Debate primarily based on
political, economic, and
institutional issues as well
as agroclimatic issues and
economics of scale in re-
search arguments. Pattern
of regional resource allo-
cation based primarily on
the need for strong re-
gional stations with power
to influence the research
policy of central stations,
and ability to effectively
use other research institu-
tions

Political economy factors
determine the availability
of research funds for dif-
ferent purposes. Propo-
nents of poverty reduction
R&D place emphasis on
giving resource-poor cli-
ents the ability to effec-
tively demand R&D on
their problems


Vol. 3, No. 2, 1993






BIGGS AND FARRINGTON


Table 1. Characteristics of Alternative Approaches to Agricultural Research and
Technology Promotion, Continued.

Central-source- Multiple-source-
of-innovation of-innovation
Characteristics approach approach


13. Dynamics of the
institutions in the
R&D systems


14. Rewards to research-
ers and to research
and extension institu-
tions





15. Determinants of the
behavior of research
and extension institu-
tions


There is a "natural pro-
gression" in the develop-
ment of research and
extension institutions

Returns to research ability
and contributions to
"good" science and exten-
sion work





Mainly directed by the
logic of stages in a scien-
tific problem-solving cy-
cle, e.g.: (1) stages in a
problem-solving process;
(2) the role an institution
is designated to play in
stages of technology gen-
eration and diffusion; and
(3) the stage of a coun-
try's research and exten-
sion capability
development


A system of institutions
that are always in disequi-
librium and undergoing
major structural change

Multiple criteria, e.g., (1)
returns to rent-seeking ac-
tivities; (2) returns to
profit motives; (3) returns
to good science; (4) re-
turns to administrative ac-
tivities; and (5) returns to
serendipity

Mainly directed by specif-
ic political, economic, and
institutional forces, e.g.,
(1) access to different
types of research and ex-
tension resources; (2) po-
litical power of
governments to pass and
implement legislation re-
lating to the flow and use
of technology; and (3)
strength of different so-
cioeconomic and other in-
terest groups to make
demands on research and
extension systems


The central-source approach is characterized by notions that science is
"value free," that different types of scientific institutions have distinct roles
and relate to each other in a hierarchical mode, that research should follow a
sequence of stages, and that the functions and institutional forms of research
and extension are necessarily distinct. Central-source thinking has exerted
several influences on the theory and practice of FSR, which are described
below.


Journal for Farming Systems Research-Extension






FSR AND THE RURAL POOR


Excessive Reliance on Quantification of "Rates of Return" to Research
Research managers, often under pressure from donors, frequently are
concerned with allocating scarce research resources so that they get "the best
rate of return on research investment." This has stimulated a growing
econometric literature on rates of return in leading international journals.
Farming systems research (FSR) is not immune to this type of analysis
(Martinez and Sain, 1983). These efforts face severe problems in quantifying
"inputs" and "outputs" and in determining timescale and causality (see
Hallam, 1990), but also give rise to two serious misperceptions.
First, they encourage the view that there is something called an investment
decision that can be isolated in a meaningful way from the political, economic,
and institutional context. In fact, however, a positive "rate of return" to, say,
wheat research would only be meaningful in the local context of, for example,
whether (and how far) prices had been maintained by government interven-
tion or whether fertilizers were accessible and subsidized. All that a high rate
of return means is that under specific contextual decisions a great deal of useful
R & D was done, which complemented the "costed" inputs of formal science.
The term conveys little about the quality of science or of research resource
allocation decisions in any general sense.2
Second, the apparent elegance of theoretical and econometric analysis
distracts attention from the more interesting issues research managers need to
address. For instance, if in the past some research managers did make "good"
decisions, got priorities "right," and managed the research program cost-
effectively, then how did they do it? How, for instance, did they second-guess
government intentions or forecast relative prices? How did managers forecast
that local and international political pressures would be such as to result in
"their" technologies being in demand? The critical issues for understanding
past research managers' decision-making rarely are discussed by economists
engaged in quantitative rate-of-return research. Their work, especially since
it generally shows high rates of return, has been used as a bargaining counter
to obtain more resources for research, but it offers little guidance on how
research decisions can be taken to meet future policy objectives, which is a
major concern of economists in the political economy tradition.



2 For a discussion and critique of "rates of return" research see Anderson, 1985; Bell, 1986; Wise,
1986; Biggs, 1989; and Hallam, 1990.


91. 3, No. 2, 1993






BIGGS AND FARRINGTON


Leaders, Followers, and Second-Generation Problems
A particular misperception associated with the central-source model is
conveyed in the term "second-generation problems." The common usage of
this term conveys that "we in centers (who had power to make sure that our
priorities were implemented first time around) could not have foreseen certain
outcomes that have now manifested themselves as second-generation prob-
lems." The implication is that although the center acted in the best way
possible in the past, some unforeseeable second-generation problems have
arisen and now need to be addressed. This may be so in certain cases.
However, this centralist language helps to obscure the fact that in many
circumstances there often were scientists and other interest groups who had
foreseen the second-generation problems, and still lost out in the political
processes of research resource allocation (Rudra, 1987).

Research Capability, Empowerment, and Leadership
A further myth arising from the central-source model is the political
neutrality of such terms as "leadership" and "strong research
capabilities." The strengthening of research capabilities means
little unless the full nature of that capability is described. Some FSR
programs are, in fact, concerned with what we regard as ultimately
desirable; that is, with efforts to strengthen local research capabil-
ities and to make them sustainable over the long term. However,
centralists often have sought to strengthen national agricultural
research capabilities so that they can better receive and adapt technology,
methods, and institutional forms transferred from other centers.
Support for FSR has been seen as a way of promoting this process. Rarely
do we see terms used in the mainstream literature denoting that FSR is
primarily concerned with increasing the control by poorer groups and their
protagonists over technology development, either through a long term
sustainable development of their own research capacity or through influence
on the research agendas of national and international centers.
These examples of ways in which the central-source model permeates the
theory, practice, and documentation of FSR highlight its tendency to accom-
modate, colonize, and absorb features of alternative research practice without
challenge. We venture to suggest that many FSR projects, which are seen by
their funders, supporters, and practitioners as "radical" new ways forward, are,
in fact, "special projects" within the centralist tradition. In practice, they may
serve as a way of defusing potential threats to the central-source model,


Journal for Farming Systems Research-Extension






FSR AND THE RURAL POOR


allowing it to be left in place and in control by diverting attention from
possible challenges to it. The empirical evidence on which such challenges are
based is growing rapidly. For example:
1. Numerous historical studies of science demonstrate that even basic
science rarely is value-free. For instance, it is clear that in the future the
directions of basic and strategic research in biotechnology will have major
implications for the welfare of poor people in developing countries. For
example, much work on molecular biology in animals is premised on the
objective of increasing the animals' response to the favorable feed and habitats
found in developed countries, whereas the ability to enhance feed supplies is
a major constraint in poor countries.
2. There is much empirical evidence to demonstrate that important
insights for new technology and for future research directions come from
those conducting applied research or working with users, that is, from
innovators in adaptive and applied parts of the formal system, from extension
personnel, and from informal R & D on the part of technology users.
3. There is evidence that many innovations and new perceptions of
problems did not, in fact, come from a research process that followed
problem-solving "stages" (Beveridge, 1957; Latour, 1987). Indeed, scarce
funds often have been wasted when scientists have, over many years, kept to
original blueprints or "stages," using them as a defense against challenges by
outsiders (including intended beneficiary groups) to the relevance of their
research.
Political economy approaches to FSR postulate that technology has mul-
tiple sources that render such approaches better able to respond to these
challenges. These political economy approaches are an expression of a long
tradition of alternative science in which researchers do not abstract research
practice from its historical, political, economic, and institutional context.3
Political economy approaches focus on sources, processes, and outcomes of
research as society passes from one state of disequilibrium to another. To a
large degree, research takes place in a worldwide marketplace where there is
always disequilibrium and potential conflict among interest groups of scien-
tists, entrepreneurs, and politicians. Public, private, and nongovernmental
organizations (NGOs) attempt to influence the direction of science and
technology promotion in such a way that surpluses, profits, and prestige
3 For recent literature on alternative models of science practice see Fay, 1975; Anderson, ct al,
1982; Lincoln and Guba, 1985; Clark and Clay, 1986; Clark, 1987; Aronowitz, 1988; Biggs and
Clay, 1988; Feyerabend, 1988; Latour, 1988; and Biggs and Farrington, 1991.


Vol. 3, No. 2, 1993





BIGGS AND FARRINGTON


Figure 1. The Political Economy Approach in a Dynamic Framework.

benefit one group of people rather than another. International centers
sometimes lead the way, and sometimes they follow the innovations of others.
Sometimes large-scale private companies gain the most from public-sector
research, and sometimes poor farmers gain. Sometimes networks and
information linkages help to strengthen and empower research capabilities
among resource-poor people, and at other times they cause ideas to be
appropriated by more powerful institutions in the system.
From a multiple-source perspective, a gap in science or a second-generation
problem may result as much from the uneven distribution of enquiry handled
by a dominant paradigm as from any notion of missing factual components in
our knowledge. The model recognizes the ambiguous nature of such terms
as "international research centers" or "national agricultural research institu-
tions." These terms often reflect institutional or legal categories rather than
inherent research capability. Thus the enhanced salaries and status of
researchers joining an international institute do not necessarily lead to
commensurate increases in research productivity vis-a-vis former colleagues
left behind to address what often are more complex scientific problems under
far more difficult conditions. Figure 1 illustrates the range of contexts that
have to be considered in multiple-source thinking, placing them in a dynamic


Journal for Farming Systems Research-Extension


Political Economic
context context




Historical context Farming systems Future context
( research policy
programs and projects



Time: Start Present Future End


Institutional Agro-economical
context context





FSR AND THE RURAL POOR


perspective. The work of an FSR group would be seen in a continuum of time
from left to right. The start ofa newproject or "new" methods and approaches
are always an ambiguous point in time, especially because what is new may
reflect merely an accounting decision to allocate funds to a specific group of
people or the ability of a speaker, academician, or researcher to claim novelty.
Whether past research in this area, even in the same institution, has been
recognized is determined by prevailing political, economic, and institutional
factors.
Similar factors are key determinants of how an FSR project is managed, as
well as of apparent successes and failures. In the context of planning future
policy and programs for FSR, it is clearly evident, if this framework is used, that
projections concerning future political, economic, and institutional reality are
as important for FSR practitioners as is consideration of internal organization,
methods, and techniques.


CASE STUDIES

The following case studies illustrate how a multiple-source paradigm in the
political economy tradition would view science in a different way than does the
central-source approach.

The Spread of Mexican Wheats in Asia
Wheat varieties were developed through on-station and on-farm research
in northwest Mexico in the 1940s and 1950s in response to clear national and
international commitment, both political and financial, to develop wheat
technology for farmers with irrigation. The work was highly farmer-oriented,
with a very high level of farmer-researcher interaction.
An innovative research method-the introduction ofoff-season growing of
wheat in a difficult environment-not only cut the time of variety develop-
ment in half (two crops instead of one crop per year), but it also subjected the
material to additional selection stress. It was serendipitous that, when
introduced into Asia, this material was superior to local varieties and to
improved materials from national research systems. The decision to import
large quantities of wheat seed from Mexico to India in the mid-1960s, with
a minimum of on-station and on-farm testing, was a political decision
(Hopper, 1978). Some scientists maintained that the risk of genetic vulner-
ability was worth taking, while others disagreed. The wheats were promoted
through large-scale public-sector intervention, comprising extension pro-


Vol. 3, No. 2, 1993






BIGGS AND FARRINGTON


grams, input subsidies, and price supports. The long-term sustainability of
these wheats was ensured by the priority funding for wheat research capabil-
ities in India, both before the introduction of Mexican varieties and subse-
quently. In the more difficult conditions of eastern India, many farmers
started growing Mexican wheat under rainfed conditions before formal
research had begun in those areas. They also developed ways of selecting and
storing seeds effectively despite high humidity in the off-season. Farmers and
local scientists in eastern India were helped by the fact that, when the seeds
were distributed, food prices were relatively high and inputs often subsidized.
In Bangladesh, the formal wheat research programs repeatedly responded to
changing contextual considerations. Senior wheat scientists were involved in
numerous, rapid, special-purpose surveys and were personally involved in
developing seeds for different parts of the country, learning from farmers and
extension staff, and developing research programs in a responsive and reflec-
tive fashion.
Thus, much of the success of wheat science in South Asia arose
from collaboration between two national research systems, Mexico
and India. CIMMYT, the International Maize and Wheat Improvement
Center, grew out ofpart of the Mexican wheat program and had major support
from the many national and international wheat researchers who worked in
India in the 1960s. CIMMYT has had a continuing involvement in the wheat
programs of India and Bangladesh, but transfers of knowledge and experience
have taken place in both directions and are neither as clear-cut nor as
unidirectional as adherents to centralist interpretations maintain.

International Centers Learn from Farmers and National Research and
Extension Systems in Animal Husbandry
Many of the advocates of FSR, particularly in international
centers, claim that learning from farmers and extensionists is a new approach.4
They would have been less inclined to label it as novel if stronger multiple-
source perspectives had been adopted. Learning from the tradition of altrna-
tive science has been with us for many years. As early as 1959 an Agricultural
Production Team in India (Verma and Singh, 1969) observed:
There is an urgent need for Animal Husbandry workers to observe and study the
practices and procedures used by villagers in the management of farm animals.
4 A major exception is the work of the International Potato Center in Peru (CIP, 1981; and Horton
and Prain, 1989), which is characterized by reflection on its processes of learning, including
learning from its own mistakes.


Journal for Farming Systems Research-Extension






FSR AND THE RURAL POOR


This should be done with a view to determining whether there are sound reasons
for what is being done or whether traditional and ill-afforded beliefs are reducing
efficiency and blocking progress. In the opinion of this team, such studies would
reveal many opportunities for worthwhile research on simple but important
problems.
In a discussion of the practical worth of such studies, Verma and Singh
bemoan the absence ofinterdisciplinary methods "to distill the problems from
field-soliciting research." For instance, they suggest that "the
behaviour of the owners in relation to their animals" is a fruitful
area for research and should be taken up by extension scientists.
Published articles in India in the 1960s clearly were drawing
attention to issues that now are finally being pursued by interna-
tional centers. The likelihood is that much further innovative work
in India and elsewhere remained unpublished. Clearly, many cur-
rent ideas are not new and the more humble of the international
fraternity realize that they are joining a tradition of alternative
science rather than being at the forefront of new development and
perceptions.

Vested Interests Against Integrated Pest Management (IPM)
Kenmore (1987) analyzes factors that gave rise to excessive routine
applications of pesticides on rice in many Asian countries and the changes
introduced by "alternative scientists." These factors included: administra-
tors' promotion of a subsidized "insurance" application of pesticide, the
influence on government of the pesticide industry, disregard by scientists of
pest resistance to pesticides and the diversity ofinteractions between pests and
other technical components at low infestation levels, and scientific and
extension promotion attitudes that asked "What chemicals do I need to save
my crop" rather than "How much will I get in return for spending on
chemicals" (Kenmore, 1987: 227). Through diverse types of on-farm
experimentation, surveys and meetings, and training programs, the IPM
program challenged both initial pesticide recommendations and agricultural
policy favoring pesticides in parts of India, Malaysia, the Philip-
pines, Sri Lanka, and Indonesia. The ability of the program to take
scientific advisers and agricultural policymakers to see on-farm
research in progress and to talk to farmers were major features in counteracting
the influence of groups that had been promoting the heavy application of


Vol. 3, No. 2, 1993





BIGGS AND FARRINGTON


pesticides. This work clearly is rooted in the conviction that FSRdoes not take
place in an economic, institutional, and political vacuum. The scientists
involved recognized that the neutrality of science generally is overstated, and
mobilized resources to create information and to gain influence so that science
was used to direct benefits to one set of clients and associated interest groups
and away from others.

FSR and Agricultural Engineering: Toolcarriers "Perfected yet Rejected"
Many of the efforts to develop animal-drawn wheeled toolcarriers for small
farmers have been set in a farming systems research approach. Reviewing a
wide range of these, Starkey (1988: 9-10) concluded:
To date about 10,000 wheeled toolcarriers of over 45 different designs have been
made. Of these the number ever used by farmers as multi-purpose implements
for several years is negligible. Farmer rejection had been apparent since the early
1960s yet as recently as 1986 the majority of researchers, agriculturalists,
planners and decision makers in national programmes and agencies and interna-
tional centers were under the impression that wheeled toolcarriers were a highly
successful technology. These impressions derive from encouraging and highly
optimistic reports.
Research on wheeled toolcarriers had involved an estimated US$15 million
of direct research costs. The "top-down" approach of engineers and the failure
of economists associated with these programs to introduce sufficiently pene-
trating analysis were among the main reasons for this waste of resources. Their
behavior fits squarely into a central-source paradigm.
Specific shortcomings ofthis research included: 1) the lack of realism of on-
station research and on-farm research; 2) excessive optimism in the more
accessible literature and little effort to "understand the actual field experiences
of earlier initiatives" (Starkey, 1988: 132); 3) the influence of vested interests
on reporting and publicity (Starkey succinctly summarizes the vested interests
of funding agencies, scientists, and their respective constituencies that define
the reality of the institutional context; none was prepared to face the threat to
funding or prestige that admissions of failure might have brought.); and 4)
lack of continuous dialogue with and respect for farmers.
Starkey's conclusions, although drawn largely from project experience,
suggest in the policy framework that FSR has to be seen in a political economic
context in which interest groups in different institutions compete for research
resources, prestige, and promotions.


Journal for Farming Systems Research-Extension





FSR AND THE RURAL POOR


FSR and the Welfare of Agricultural Laborers: Threshers in Asia
In IRRI-sponsored R & D into axial-flow threshers in the Philippines,
scientists worked closely with small farmers and with local manufacturers,
resulting in widespread adoption of the technology. The main appeal of the
thresher lay in the reduced labor needed for harvesting, causing a transfer of
income from poorer to richer groups. The increase in employment in the rural
manufacturing sector did not compensate for the decrease in employment in
the agricultural sector (Duff, 1987).
When the thresher was transferred to Thailand, its impact was very
different, owing to different political, economic, and institutional contexts. In
this case, there was plenty of land available and the threshers spread in areas
where labor for harvesting was in short supply. Threshers broke this labor
bottleneck and enabled more land to be cultivated, with large net increases in
employment opportunities.
These examples demonstrate that the focus and impact of research on
mechanization have to be seen in their political, institutional, and economic
contexts. Numerous other studies, particularly on the impact of tractoriza-
tion, have demonstrated how technological change (and concomitant R & D)
have been driven by overt or implicit subsidies that work to the advantage of
some groups and to the disadvantage of others (Binswanger, 1978; McIner-
ney and Donaldson, 1975).
The welfare of laborers may also be impacted by the R & D process when
little consideration is given to standards of equipment engineering (Ashford
and Biggs, 1992). There have been considerable numbers of injuries in Asia
due to safety devices not being part of the equipment, or not being used with
wheat threshers (Clay, 1978).

FSR and the Management of Common Property Resources
If the management of community resources such as pastures, woodlands,
and watercourses-all of which generate both on- and off-farm
benefits-is to be enhanced in a sustainable fashion, social science
and political analysis will be required far beyond the realm of neoclassical
economics. Existing traditions of research for these areas have been largely
neglected. For instance, "action research" conducted in Asia for many years
(Castillo, 1972) has focused on change in village-level institutions, including
those responsible for the management of common property resources. By
contrast with FSR's preoccupation with classifying clients into homogenous


Vol. 3, No. 2, 1993





BIGGS AND FARRINGTON


groups and identifying technology for them, action research has been con-
cerned with the ways in which different socioeconomic groups relate to one
another and with who benefits and who loses in the short and long term.
Such research, however, has not always been free of central-source tenden-
cies. The early "Comilla model" for example, adhered too rigidly to imported
principles and scored poorly on efficiency, management, and equity criteria
(Khan, 1971). Other institutional developments in Bangladesh more closely
reflected the tradition of alternative science in the way local innovators-
villagers, academics, administrators-developed new institutions for group
membership and management of tubewells and low-lift pumps, as they
developed credit institutions for landless women and men.5 Researchers saw
no need for a "field laboratory" in the development of new institutions for the
poor. From the start, these alternative science researchers found room to
maneuver in the existing political and institutional context to create pro-poor
changes in technology and institutions.
In irrigation development, Martin and Yoder (1987) clearly describe how
viable common property institutions that benefit poorer groups can arise from
village-based innovations. Similarly, in soil erosion and common property
issues, the work ofJodha (1987) and others (Blaikie and Brookfield, 1987)
take a useful political economy approach to understanding the issues.


IMPLICATIONS FOR AGRICULTURAL SCIENCE AND
TECHNOLOGY PROMOTION


Philosophy and Practice of Science and Technology Diffusion
The case studies cited above draw attention to the relevance of historical,
political, economic, and agroclimatic contexts for the theory and practice of
FSR. In the past, much FSR literature has underestimated the importance of
these issues and overestimated that of new approaches, methods, and tech-
niques, frequently implying that the latter come from "centers of excellence."
In this sense, FSR has been driven by many of the attributes more widely
characteristic of central-source institutions, including expansionism, defense
of their own members' interests, and incapacity to place "basic" research or
"new" methods into wider contexts. In reality, innovations come from a range
5 The work of the Grameen Bank (Yunus, 1982), the Bangladesh Rural Advancement Committee
(Chowdhury, 1989), and Proshika (Wood and Palmer-Jones, 1990) are all examples of such
efforts.


Journal for Farming Systems Research-Extension





FSR AND THE RURAL POOR


of practitioners, including research-minded farmers, innovative extensionists,
and scientists in both field and laboratory, and in local, national, and
international contexts.
By contrast, scientists in the alternative tradition explicitly recognize that
science and technology diffusion processes are political, and that the volume
of resources allocated to one or the other depend on the character of
institutions that evolve and the ways in which they are managed. Nor
is basic research necessarily thought to be superior to applied or
adaptive research: These are value terms conditional upon the historical,
institutional, political, and economic contexts in which formal science is
practiced.
Alternative science finds less need to stratify its clients into homogeneous
groups that can be "targeted" by prescriptions, but more need to understand
how appropriate resources (including the human capacity to make personal,
moral, and value decisions) can be mustered to bring science and technology
from multiple sources to bear on poverty reduction in diverse and complex
contexts.
A further implication is that commonly encountered statements bemoan-
ing "lack of political will" to introduce one or another aspect of FSR are
excessively restrictive. Under a political economy approach, the political
context is no longer a "given" but a valid object of research by macroecono-
mists and political scientists in testing the strengths and weaknesses of a range
of institutional options.

Cross-Learning in the Social and Natural Sciences
It is rare to see references in FSR literature to research on technologies in
other sectors, such as health or rural industries, whether at household or
community levels. Yet political economy approaches arc common across all
of these areas (cf. Justice, 1986). One technique of potentially wide
application is the "user perspective" approach promoted by the Population
Council since the mid-1970s (Bruce, 1980; Zeidenstein, 1980). Nor, for
instance, do the ideas and practices of social marketing of different types of
contraceptives appear to have been on the reading lists of those in FSR
engaged in the linking of FSR to extension and development activities. In
agricultural engineering, perspectives from the appropriate technology move-
ment (Carr, 1985; Stewart, 1987) would inhibit the recurrence of the
excessively narrow approaches characterizing, for example, the introduction
of wheeled toolcarriers.


Vol. 3, No. 2, 1993





BIGGS AND FARRINGTON


Agricultural Research Policy
One of the more straightforward conclusions of a political economy
perspective is that on-farm and village-level research cannot be separated from
research at experiment stations. A recent review of "client-oriented" FSR
programs indicates the shortcomings, particularly in feedback from farmers,
that currently widen the gulf between them (Merrill-Sands et al., 1989).
These cannot adequately be addressed by the "special project" nature of much
FSR in which projects and programs are merely added on to existing
agricultural research services. Prevailing institutional and political impedi-
ments need to be addressed more directly if a range of perspectives on the
usefulness of existing technology and of future R & D priorities is to be
obtained in a sustained fashion.
Traditionally, research policy analysis undertaken by agricultural econo-
mists has narrowed itself to addressing such issues as the allocation of research
resources between major crops and technology types. It rarely has addressed
five other areas of research policy. These are:
1. Research institutional policy. At its simplest level there is a choice
between, at one extreme, a hierarchical system of technology generation and
transfer and, at the other extreme, a horizontal, federal system with the power
over any "center" being in the hands of a body controlled by groups in "the
periphery." The merits of each of these, and of a range of intermediate
possibilities, need to be considered in relation to specific contexts.
2. Research personnel policy. Adjustments to pay, prestige, and field
allowances will, in themselves, not guarantee the emergence of
wider perspectives, but they are an important precondition. Anoth-
er is a reduced overall emphasis on publications as a criterion for promotion-
scientists of practical inclination, particularly in the private commercial or
voluntary sectors, do not accord high priority to publication-and to a shift
in criteria for assessing the relative merits of different types ofpublication. For
instance, adherents to the central-source model favor publications in interna-
tional journals that meet orthodox editorial criteria and standards. However,
local publications may be more relevant to the resolution of local scientific and
policy issues. Reports of the field trips of Ladejinsky (1969) to the Punjab and
Bihar published in the Indian Economic and Political Weekly are good
examples of quick, focused surveys by an experienced observer that helped to
stimulate the current international literature on rapid rural appraisal methods.


Journal for Farming Systems Research-Extension






FSR AND TIE RURAL POOR


A political economy perspective would see those who wrote and reviewed
articles for international journals as self-selecting groups of actors. Citations
in many international journals give little evidence of a careful and systematic
review of historical or local sources of information. By drawing instead on the
dominant (and usually international) sources of literature, they often reflect
the propensity and availability of resources for publication rather than intrinsic
merit and relevance to the case in hand. If they were to adopt political science
perspectives, local research managers would rank papers and reports that
addressed local issues and were accessible to a wide diversity of scientific,
policy making, and academic audiences at least on a par with those published
in international academic journals.
A further dimension of personnel policy concerns the ethics of attribution.
Anthropologists have norms by which they quote sources of ideas obtained in
the course of their work, norms that economists do not appear to share unless
their ideas come from a notable member of their profession.
3. Research methods and techniques. The introduction of a political
economy perspective into FSR would require a range of methods and tools
broader than the conventional toolkit of surveys, trials, and household and
village meetings that are the stuff of conventional manuals. These new
methods and tools include: a check-list that prompts FSR researchers to
consider five points of the contextual analysis (the historical, political,
economic, institutional, and agroecological); pay-off matrices to help analyze
which interest groups gain or lose from alternative interventions (see Biggs,
1978; Stewart, 1987); information linkage matrices to analyze options for
changes in the controlling influence over information; institutional and
functional analysis (e.g.j., Merrill-Sands et al., 1986, 1989) to focus on the
behavior ofgovernment and donor institutions (Clay and Schaffcr, 1984); and
historical and reflective research to document the processes of past projects
and programs and to draw lessons on their effectiveness (see Biggs, 1983;
Collinson, 1988). Additionally, in contrast to much FSR that sets itself
objectives within parameters of conventional scientific practice and institu-
tions, political economy research methods would concentrate more on
understanding how local groups of poor people and of pro-poor researchers,
whether local or international, can be empowered6 to extend their own
research capabilities and to influence the science and technology agendas of
formal scientific institutions.
6. Many NGOs follow the empowerment principles ofFreire (1970), whose writings are rarely cited
in the FSR literature.


Vol. 3, No. 2, 1993





BIGGS AND FARRINGTON


4. The clients of research. Farming systems research conventionally has
concentrated on the crops being grown, yet many of the poorer rural
agricultural households have little or no land and much of their livelihood is
earned as laborers in agricultural or nonfarm activities. An enhanced focus on
rural poverty in FSR projects implies a change in type of client and in type of
technology.
5. Research on common property management. FSR has concentrated
almost exclusively on on-farm issues, yet a high proportion of the income of
poorer households is generated from the use of common property resources
(CPR). Management of CPR and conjunctive use of CPR and on-farm
resources long have been the focus of interdisciplinary action research projects
in Asia (Bottrall, 1981). There are ample opportunities for FSR to learn from
these and to complement them.

Personal Responsibilities
In asserting that science is not value-free, political economy approaches
imply that all science, whatever its source, carries with it social responsibility.
Accordingly, "escape hatches" to avoid responsibility for any poor perfor-
mance in FSR projects need to be much more closely scrutinized than
previously. These include lack of political support, lack of yield-increasing
inputs, lack of a supportive policy context, and lack of cooperative farmers.
Resort to these explanations during or after projects may imply that the
political, economic, and village-level contexts were inadequately investigated
at the beginning of the project.


CONCLUSIONS

This paper has argued that the difficulty in identifying new agriculture-related
technologies with and for the rural poor lies less in an overall shortage of
resources available for the communities in question and more in restrictive
views of how such resources should be used. Arguments in favor of political
economy perspectives are advanced on the grounds that processes of resource
allocation to one type of research over another and corresponding reward
systems are not neutral but are grounded in political, economic, historical, and
institutional contexts. Elements of such political economy approaches long
have existed in alternative traditions of science.
FSR practitioners in the international community have the room to
maneuver to stimulate wider adoption of political economy perspectives, but


Journal for Farming Systems Research-Extension





FSR AND THE RURAL POOR


they first must shift their own thinking and practice in this direction. Two
indicators are most useful in discovering whether FSRpractitioners of any type
or at any level in the international community are committed to alternative
political economy approaches. The first is whether they see the sustained
empowerment of those below them as their most important underlying
objective. For those in international agencies, this would mean empowering
those scientists in national research systems with a commitment to reducing
poverty to be able to control the policies ofthe international centers. For those
researchers who work with poor people at the household and community
level, it would mean empowering them to have control over national and
international R & D institutions. A second indicator is whether the writings
of those in centers adequately reflect that they are learning as much in some
aspects of science and technology as they are giving in others.
Many of the advocates of poverty-focused FSR now are stressing the
importance of learning from and working with poor people. Iron-
ically, by suggesting that what they are doing is new, some of these advocates
are, in fact, failing to recognize the intellectual and personal contributions of
many natural and social scientists who through their spoken and written work
have been following an "alternative" approach for many years. By doing so
against prevailing currents in funding and reward systems, these scientists
often have incurred high personal costs in their own careers. A final test of
whether current FSR practitioners have shifted ground will, therefore, be the
extent to which they begin to search out, cite, and use material from the
alternative science tradition they are now joining.


REFERENCES

Anderson, J.R. 1985. Criteria for assessing farming systems research. Agricultural
Administration 20:225-35.
Anderson, R.S., P.R. Brass, E. Levy, and B.M. Morrison, eds. 1982. Science, politics and
the agricultural revolution in Asia. Boulder, CO: Westview Press.
Aronowitz, S. 1988. Science as power: Discourse and ideology in modern society.
Minneapolis: University of Minnesota Press.
Ashford, T., and S.D. Biggs. 1992. Dynamics of rural and agricultural mechanisation:
The role of different actors in technical and institutional change. Journal of
International Development 4:349-74.
Bell, R.M. 1986. The economic gains from technical change in agriculture: Returns to
investment in what? (2nd Draft). Science Policy Research Unit, University of Sussex.
Beveridge, W.I. 1957. The art of scientific investigation. New York: Vintage Books.
Biggs, S.D. 1978. Planning rural technologies in the context of social structures and
reward systems. Journal of Agricultural Economics 29:257-74.


Vol. 3, No. 2, 1993






BIGGS AND FARRINGTON


Biggs, S.D. 1983. Monitoring and control in agricultural research systems: Maize in
northern India. Research Policy 12:37-59.
Biggs, S.D. 1990. A multiple source of innovation model of agricultural
research and technology promotion. World Development 18:1481-99.
Biggs, S.D., and E.J. Clay. 1988. Generation and diffusion of agricultural
technology: A review of theories and experiences. Pages 19-67 in I. Ahmed and V.W.
Ruttan, eds., Generation and diffusion ofagricultural innovations: Role ofinstitution-
alfactors. Aldershot: Gower.
Biggs, S.D., and J. Farrington. 1991. Agricultural research and the rural poor: A review
ofsocialscience analysis. School of Development Studies, University of East Anglia and
Overseas Development Institute, London. Ottawa: International Development
Research Centre.
Binswanger, H.P. 1978. The economics oftractors in South Asia. NewYork: Agricultural
Development Council.
Blaikie, P.M., and H. Brookfield, eds. 1987. Land degradation and society. Mcthuen.
Bottrall, A. 1981. Action research towards improved water distribution. Agricultural
administration irrigation management network paper 5c. London: ODI.
Bruce,J. 1980. Implementing the userperspective. Studiesinfamilyplanning 11(1):29-
34.
Carr, M. 1985. The AT reader: Theory and practice in appropriate
technology. London: Intermediate Technology Publications Ltd.
Castillo, G. 1972. Research and the action program, A/D/C. New York.
CIP. 1981. Farmers help scientists change research objectives. Circular 9(4), May.
Peru: Centro International de la Papa (CIP).
Chowdhury, A.M.R. 1989. Delivering research: The case of oral rehydration therapy and
tubewell irrigation in Bangladesh. Paper presented at a Workshop on the Future of
Alternative Sciences in Asia: Agricultural and Health Research and Delivery. Univer-
sity of British Columbia, Vancouver, October 27-28.
Clark, N. 1987. Similarities and differences between scientific and technological
paradigms. Futures February:26-42.
Clark, N., and E. Clay. 1986. The dryland research project at Indore 1974-80: An
institutional innovation in rural technology transfer. Journal of Rural Studies
3(2):157-173.
Clay, E. 1978. Wheat threshers. ADAB (Agricultural Development Agencies in
Bangladesh, Dhaka) News 7:29.
Clay, E.J., and B. Schaffer, eds. 1984. Room for manoeuvre: An exploration of public
policy in agricultural and rural development. Heinemann.
Collinson, M. 1988. The development of African farming systems: Some personal views.
Agricultural Administration and Extension 29(1):7-22.
Duff, B. 1987. Changes in small farm paddy threshing technology in Thailand and the
Philippines. Pages 95-139 in F. Stewart, ed., Macro policiesfor appropriate technology
in developing countries. Boulder, Colo.: Westview Press.
Fay, B. 1975. Social theory and political practice. London: George Allen and Unwin.
Feyerabend, P. 1988. Against method. Iondon: Verso.
Freire, P. 1970. Pedagogy of the oppressed. New York.
Hallam, D. 1990. Agricultural research expenditures and agricultural productivity
change. Journal of Agricultural Economics 41(3):434-9.


Journal for Farming Systems Research-Extension






FSR AND THE RURAL POOR


Hopper, D.W. 1978. Distortions of agricultural development resulting from govern-
ment prohibitions. In T.W. Schultz, ed., Distortions of agricultural incentives.
Bloomington, IN: Indiana University Press.
Horton, D., and G. Prain. 1989. Beyond FSR: New challenges for social scientists in
agricultural R& D. Quarterly ournalofInternationalAgriculture 28(3/4):301-14.
Jodha, N.S. 1987. A case study of the degradation of common property resources in
India. In P.M. Blaikie, and H. Brookfield, eds., Land degradation and society.
Methuen.
Justice,J. 1986. Policies, plans and people: Foreign aid and health development. Berkeley:
University of California Press.
Kenmore, P.E. 1987. Crop loss assessment in a practical integrated pest control program
for tropical Asian rice. Pages 225-41 in P.S. Teng, ed., Crop Loss Assessment and Pest
Management. St. Paul, Minnesota: American Phytopathological Society Press.
Khan, A.H. 1971. Tour ofTwenty Thanas: Impressions ofdrainage-roads, irrigation and
co-operative programmes. Kotbari, Comilla, Bangladesh: Bangladesh Academy for
Rural Development.
Ladejinsky, W. 1969. The green revolution in Bihar-The Kosi area: A field trip.
Economic and Political Weekly 6(36).
Latour, B. 1987. Science in action: How to follow scientists and engineers through society.
Milton Keynes: Open University Press.
Latour, B. 1988. The politics of explanation: An alternative. Pages 155-76 in S.
Woolgar, ed., Knowledge and reflexivity: New frontiers in the sociology of knowledge.
London: Sage Publications.
Lincoln, Y.S., and E.G. Guba. 1985. Naturalistic inquiry. Beverly Hills: Sage
Publications.
Martin, E.D., and R. Yoder. 1987. Institutionsfor irrigation management infarm-
managed systems: Examples from Nepal. IIMI Research Paper 5. Digana Village via
Kandy, Sri Lanka: International Irrigation Institute.
Martinez, J.C., and G. Sain. 1983. The economic returns to institutional innovations in
national agricultural research: On-farm research in IDIAP, Panama. Working Paper
4/83. CIMMYT Economics Program.
McInerney, J.P., and G.P. Donaldson. 1975. The consequences of farm tractors in
Pakistan. World Bank StaffWorking Paper No. 210. Washington, D.C.: World Bank.
Merrill-Sands, D., S. Biggs, S. Kean, S. Poats, J. McAllister, E. Moscardi, and S. Ruando.
1986. Methodology for the ISNAR study on the organisation and management of on-
farm research. The Hague: International Service for National Agricultural Research.
Merrill-Sands, D., P. Ewell, S. Biggs, and J. McAllister. 1989. Issues in institutionalizing
on-farm client-oriented research: A review of experiences from nine national agricul-
tural research systems. Quarterly Journal of International Agriculture 3(4).
Rudra, A. 1987. Technology choice in agriculture in India over the past three decades.
In F. Stewart, ed., Macro-policies for appropriate technology in developing countries.
Boulder, CO: Westview Press.
Singh, D.P. 1982. Management of an agricultural university. In R.S. Anderson, et al.,
eds., Science, politics and the agricultural revolution in Asia. Boulder, Colo.:
Westview Press.
Starkey, P. 1988. Animal-drawn wheeled tool carriers: Perfected yet rejected. Braunsch-
weig/Wiesbaden: Friedr, Vieweg and Sohn.


Vol. 3, No. 2, 1993






BIGGS AND FARRINGTON


Stewart, F., ed. 1987. Macro policiesfor appropriate technology in developing countries.
Boulder, Colo .: Westview Press.
Verma, M.R., and Y.P. Singh. 1969. A plea for studies in traditional animal husbandry.
The Allahabad Farmer 18(2):93-8.
Wise, W.S. 1986. The calculation of rates of return from agricultural research from
production functions. Journal of Agricultural Economics 37(2):151-61.
Wood, G., and R. Palmer-Jones. 1990. The watersellers: A collective venture by the rural
poor. West Hartford: Kumarian.
Yunus, M. 1982. Grameen bank project in Bangladesh-A poverty focused rural
development programme. Paper presented at Bangladesh Academy for Rural Devel-
opment, Comilla, Bangladesh, March 22-27.
Zeidenstein, G. 1980. The user perspective: An evolutionary step in contraceptive
service programs. Studies in Family Planning 11(1):24-9.


Journal for Farming Systems Research-Extension









Farm-Level Performance of Improved

Cassava Varieties in the Humid Forest Zone

of Nigeria



F.I. Nweke, H.C. Ezumah, and D.S.C. Spencer1



ABSTRACT


This paper presents the results of a field survey of cassava-producing
households in Ovia Local Government Area (LGA) of the then Bendel
state2, Nigeria. Cassava is the second most important staple food in sub-
Saharan Africa and improved varieties have been developed by research
institutions during the last 15 years. Farmers in the survey area have
widely adopted improved varieties, which they have fitted into their
production systems without changing any of their old practices. They
obtain 35- to 75-percent higher yields and incomes with the improved
varieties. Weed control and processing technologies that do not require
the use of expensive purchased inputs will raise the incomes of the
improved-cassava-variety producers. Labor is a major constraint to
cassava production.


INTRODUCTION

Cassava (Manihot esculenta Crantz), is the staple food of approximately 160
million people, or about 40 percent of the population of sub-Saharan Africa
(IITA, 1987). Only maize (Zea mays), which is consumed by about 200
million people, or 50 percent of the population, is a more important staple
food than cassava. West Africa produces about one-third of all the cassava in
sub-Saharan Africa. An additional one-third is produced in Central Africa,
while East and Southern Africa together account for the rest of the cassava
output. Zaire and Nigeria are the continent's leading producers of cassava,

1 F. I. Nweke and D.S.C. Spencer are economists; II.C. Ezumah is an agronomist; the authors
work at the International Institute of Tropical Agriculture, Ibadan, Nigeria.
In August 1991, Bendel state was divided into two new states, Edo and Delta. Ovia LGA is now
located in Edo state.





NWEKE


with Nigeria accounting for about 23 percent of the total output (Dorosh,
1989). About 75 percent of the cassava in West Africa is grown in the humid
forest and moist savanna zones. The objectives of this study are to assess the
performance of improved cassava varieties at farm level and to identify any
problems that may constrain the full realization of their potential.


STUDY METHODOLOGY AND THE STUDY AREA

This paper is based on a diagnostic survey conducted in 1987 in an area where
the International Institute of Tropical Agriculture (IITA)-derived cassava
varieties were known to have been widely adopted. The Rapid Rural Appraisal
(RRA) method was adopted, using data from farmer interviews at both the
village and individual level; data also were obtained from direct measurements
in the field, and from various secondary sources. The survey was conducted by
economists and agronomists.
Ovia Local Government Area of Bendel state, Nigeria (Figure 1) was
selected, using knowledgeable informants and secondary information. A large
number of the farmers in Ovia LGA are migrants who were displaced from
Isoko, Ughelli, and Ethiope LGAs of the same state as a consequence of
petroleum exploration.
Cassava is a staple food in the area and the agroecological conditions are
largely representative of the cassava-growing areas in the humid climate zone
ofWest and Central Africa. In addition, IITA-improved and some traditional
varieties of cassava are grown in the area by smallholder farmers. Three
villages, namely Ohosu, Igbogui, and Igwugun in Ovia LGA, were selected
randomly. A group interview was conducted with the farmers in each of the
villages, and 15 farmers from each village were then selected at random for
individual interviews. Yield and field area measurements were taken from the
cassava fields, some planted with improved and others with local varieties, of
the 45 representative farmers.
The field areas were determined by measurement with compass, tape, and
ranging poles. Yield estimates were made for fields that were at least 12 months
old. The estimation was based on a representative plot of 40 m2, except when
the field was small, in which case a 20 m2 plot size was used. There were one
or two plots per field depending on the size and heterogeneity of the field in
terms of soil and toposequence. Cassava stands within the sample plot were
first counted and then harvested. The roots and the tops were weighed
separately and the roots were counted.


Journal for Farming Systems Research-Extension






IMPROVED CASSAVA VARIETIES IN NIGERIA


10 5 0 0 20 30 40 50 60 Miles
10 0 20 40 60 80 100 Kilometres

Figure 1. Distribution of Improved Cassava Planting Materials,
Bendel State, Nigeria, 1987.


Vol. 3, No. 2, 1993





NWEKE


Separate questionnaires were used for group and for individual farmer
interviews. The group interview questionnaire covered information that was
not expected to vary from farmer to farmer; this included, for example,
information on cropping patterns, major constraints faced by cassava farmers,
and the availability of farmland. The individual farmer questionnaire covered
information on household size and composition, cassava varieties grown,
source of farm labor, farm expenditure, and disposal of farm products.
Secondary information on the distribution of planting material was collect-
ed from Shell Petroleum Company of Nigeria Limited (Shell-BP) in Warri; the
Bendel State Ministry of Agriculture, and the Bendel State Agricultural
Development Project, both in Benin City. Weather information was obtained
from the meteorology station of the Nigerian Institute for Oil Palm Research
(NIFOR) in Benin City. Price information was acquired from the Bendel State
Ministry of Finance and Economic Development, Benin City. Cassava pro-
duction labor data were culled from other surveys (Knipscheer, 1980).
Rainfall data compiled over 43 years in Benin city show a range of 1150 to
2200 mm/year, with a mean of 1914 mm/year. Rainfall distribution is
heaviest from March to October, and crop planting generally begins from the
second week of March but is less risky if started one month later. Major food
crops are cassava, maize, and plantains, although yam and egusi melon also are
grown. Tree crops are cocoa, rubber, and oil palm.


CROPPING PATTERN AND CALENDAR OF FARM
OPERATIONS

The farming system in Ovia LGA is characterized by intercropping, the major
intercrops being cassava/maize, cassava/plantains, and tree crops (oil palm or
cocoa)/cassava. The calendar of farm operations is described in Figure 2. Land
preparation, which is essentially slash-and-burn, is carried out between
December and March, although the peak is in February and March. Tillage is
minimal and is combined with planting. The farmers claim that higher cassava
root yield is possible in tilled fields and all indicated that they would till their
fields if they had a tractor or more labor. Cassava is planted from March to
September, but April and May constitute the peak period. Melon, yam, and
first-season maize are planted between March and May. March-planted maize
has a high risk of failure, arising from the possibility of inadequate rainfall (i.e.,
rains starting later than the second week in March). But the farmers usually
take this risk because of the opportunities of receiving a premium price for


Journal for Farming Systems Research-Extension






IMPROVED CASSAVA VARIETIES IN NIGERIA


Activity
1. Land clearing
2. Planting
cassava
maize (early)
maize (late)
melon
yam
3. Weeding
4. Harvesting
cassava
maize (early)
maize (late)
melon
yam
oil palm
5. Processing
cassava
melon
oil palm


Minor activity 1 Regular activity Peak activity

Figure 2. Calendar of Farm Operations Indicating Peak Periods in Ohosu Area,
Bendel State, Nigeria, 1987.

their early maize. Late maize is planted in August and September. Weeding is
done from April to October with hand hoes and cutlasses; herbicides are not
used. The weed problem is exacerbated by the practice of minimum tillage; 95
percent of the farmers consider weeding to be their most labor intensive farm
operation. Cassava harvesting is continuous throughout the year, although it
peaks between November and March.


Cassava Processing to Make Gari
Because cassava roots are not eaten fresh in the area, cassava production is
not complete until the roots are transformed into one of several products. Gari
and starch (for human consumption) are the major cassava products made in
the Ovia LGA. Cassava leaf is not eaten and cassava is not processed into
animal feed. Starch is extracted in the process ofgari making, but removal of
the starch reduces the gari quality. Because gari and not starch is the main
commercial product, onlyjust enough starch to meet household consumption
needs is produced by the farmers.
The peeling of harvested roots in preparation for processing into gari is
done manually with small cutlasses. Peeled roots are grated on a custom basis
using motorized mills. After fermentation and the pressing out of excess


Vol. 3, No. 2, 1993


O N D J F MiA M J J A S


......


It~f~lI~P/~T~T~


mali aumlll!a


j i j
i~T~I~TT5~3
E~,~s~T~s~~sss~
e~rTTT~T~TT~T~ I i i i i i i



j j i I _I I





NWEKE


water, the partially dehydrated mash is toasted over an open fire in a metal pan.
All processing operations are done almost exclusively by women and children.

Cassava Diseases and Pests
The major disease observed on the cassava fields in Ovia LGA was cassava
mosaic disease; yield loss from this disease can be high, ranging from 20 to 90
percent (Hahn et al., 1979). The most important pest observed, particularly
on local varieties, was the cassava mealybug. Cassava mealybug has been
reported to cause yield reductions ranging from 40 percent to 85 percent
(Nwanze et al., 1978). However, these losses have been reduced through the
intervention of biological control measures (Herren et al., 1986), as well as
through effective local enemies (Akinlosotu, 1983). Green spider mite also
was observed, but it was not as important a pest as the mealybug and appeared
to cause little damage in the area.

Labor Profile in the Ovia LGA
The busiest period for the Ovia LGA farmers is between March and May,
at which time they clear the land, plant cassava and its associated crops, weed
early-planted maize and cassava, and harvest and process not only cassava but
also oil-palm fruit. The slack period is between June and October when the
major operation is second weeding. Some harvesting and processing of
cassava, as well as some other operations, continue during this period, as minor
or regular activities (see Figure 2).
Farmers reported that they spent an average of 400 naira (equivalent to
about sixty US dollars) on hired labor for cassava production in 1986. Since
wage rates were about 6 naira/person-day, about 67 person-days of labor were
hired per farmer. Most of the hired labor was, however, employed by farmers
with larger-than-average farm size.
Family size in the area was large, with an average of 3.6 wives and 10
children. The farmer and his wives work full time on the farm. School-age
children help out when not in school. Markets are held once every five days;
the majority of the people attend. In addition, as Christians, they go to church
four days per month. Broadly speaking, therefore, they have about 20 days a
month, per adult, potentially available for farm work.
During the busy period (March to May), the average farmer who cultivates
0.82 ha (see below) of cassava needs 33 person-days for land preparation and
11 person-days for planting. The farmer also needs significant amounts of
labor for weeding, harvesting, and processing. In addition to time require-


Journal for Farming Systems Research-Extension






IMPROVED CASSAVA VARIETIES IN NIGERIA


ments for necessary social activities, farmers also engage in activities such as
oil-palm fruit harvesting and processing. They are, therefore, likely to face
labor bottlenecks during the busy season. The extent to which they can expand
cassava production, therefore, depends on their ability to hire labor.


DISTRIBUTION OF THE IMPROVED CASSAVA VARIETIES

Improved cassava varieties identified in the Ovia LGA are TMS 30572, TMS
30211, and TMS 30555, which originated from the IITA and are variously
called Ogoja, Agric, and Nitor by area farmers. The level of adoption, based
on the proportion of fields planted to improved varieties, is estimated at over
80 percent.
In the early 1970s, IITA, Shell-BP, the Federal Department ofAgricultural
Research, and Bendel State Ministry of Agriculture began cooperative efforts
at cassava selection and breeding for resistance against cassava bacterial blight
and for higher yield (IITA, 1976; Ohunyo, n.d.). In 1976, Shell-BP started
a cassava multiplication program at Agbarho in Ughelli LGA, using 66
varieties selected or bred earlier through their cooperative efforts. In 1977,
Shell-BP started distribution of planting materials directly to farmers.
The volume of planting materials distributed to farmers to date by Shell-
BP would be adequate to plant the entire farming area of Bendel state at least
twice over (Ohunyo, n.d.). However, the materials were not evenly distribut-
ed among all the Local Government Areas. Amounts supplied in Isoko,
Ughelli, Ethiope, and Okpe LGAs would be adequate to plant their entire
farming areas more than five times, but amounts supplied in Ovia, Orhrion-
mwon, Oredo, Warri, Burutu, Ndokwa, Oshimili, and Okpeho would not be
sufficient to plant the farming areas in these LGAs.
Cassava is not only an important staple food crop of farmers in the Ovia
LGA, its production and gari processing are the primary commercial activities
of the farmers. In addition, as migrants from Isoko, Ughelli, and Ethiope
LGAs, which had large supplies of the improved cassava planting materials in
the late 1970s, Ovia LGA farmers were well aware of the potential value of the
improved varieties. The unusually high adoption rate of over 80 percent
observed in the area can be explained by the fact that the improved varieties
were widely available to the farmers and that the farmers grow cassava for
commercial reasons. Other research findings indicate that smallholder farmers
are more likely to adopt new technologies for commercial production than for
subsistence production (Grilliches, 1957).


Vol. 3, No. 2, 1993






NWEKE


FRESH ROOT YIELD PERFORMANCE OF THE IMPROVED
CASSAVA VARIETIES

Various yield factors of both the improved and the local varieties at the farm
level are presented in Table 1 for 12- to 15-month-old cassava. The mean fresh
root yield of the improved varieties was 19.6 ton/ha, with a range of 11.5 to
26.3 tonne/ha.
This average yield compares favorably with that obtained by IITA in its
researcher-managed trials in three locations in other parts of the humid forest
zones of Nigeria between 1983 and 1985 (Table 2). This indicates that
farmers in the Ovia LGA are fully exploiting the yield potential of the
improved varieties even when no fertilizers are used.
The improved varieties produced yields 75 percent higher than local
varieties (Table 1). This difference is statistically significant at the one percent
level using the two tailed "t" test. The number of plants/ha is not significantly
different between the improved and the local varieties; hence, the difference
in root yield is attributed to the higher bulking capacity of the improved
varieties (Hahn et al., 1979). This is reflected in their average root weight
which is higher than the local varieties by nearly40 percent, and to the number
of roots per plant, which is higher than the local varieties by 45 percent.
The root yields ofthe improved and local varieties were regressed separate-
ly, with an average yield of the two (c) for each of the fields using the model
Table 1. Farm-Level Components for 12- to 15-Month-Old Improved and Local Va-
rieties of Cassava, Ovia LGA, Bcndel State, Nigeria, 1987.

Yield components Improved varieties Local varieties Standard error
Fresh root yield (tonne/ha) 19.6 11.2 3.59
Number of plants/ha ('000) 7.0 7.8 1.49
Number of roots/plant 3.71 2.56 1.67
Average root weight (kg) 0.77 0.56 0.18
Harvest index 0.49 0.38 0.06


Table 2. Results of IITA Trials Without Fertilizer, Using Cassava Variety TMS30572
at Three Locations in Nigeria, 1983 to 1985.
Fresh rubber yield at 12 months
ILocation 1983 (t/ha) 1984 (t/ha) 1985 (t/ha) Mean (t/ha)
IITA station (Ibadan) 22.4 29.9 19.1 23.8
Onne 23.4 24.5 15.5 21.0
Warri 17.3 16.1 13.5 15.7
Mean 21.0 23.5 16.0 20.2


Journal for Farming Systems Research-Extension






IMPROVED CASSAVA VARIETIES IN NIGERIA


Figure 3. Yields of 12-15 Month Improved and Local Cassava Varieties,
Ovia LGA, Bendel State, Nigeria, 1987.

of Eberhart and Russell (1966) and Hildebrand (1984). The environmental
index, e, represents the effects of farmer management skill, soil, and climatic
variables on performance of technology at the farm level. The result (Figure
3) shows that the improved varieties' function has both higher intercept and
slope than the local varieties' function. This indicates that in terms of fresh root
yield, the improved varieties harvested at 12 to 15 months have a higher
potential for fresh root yield and consistently outyield the local varieties, no
matter how favorable the environment or farmer management.
Since the Ovia LGA farmers also harvest some of their cassava at 16 tol8
months, yield samples also were taken from older cassava fields. The result
showed that the older the cassava, the less the difference in yield between
improved and local varieties. The percentage difference in yield declined from
75 percent at 12 to 15 months to 32 percent at 16 to 18 months. However,
sample size for the 16- tol8-month-old fields was too low because most fields
are harvested at 12 to 15 months by the farmers. Consequently, our conclu-
sion should be treated with caution.


Vol. 3, No. 2, 1993






NWEKE


COSTS AND RETURNS IN CASSAVA PRODUCTION

Farmland is relatively abundant in the area and is allocated free to farmers by
the Ovia local government. The allocation is on a long-term basis, as none of
the surveyed farmers had ever relinquished any land allocated to him.
Farmland, therefore, is assumed to have no opportunity cost outside cassava
production and is not costed in this analysis. There was an average of 0.82 ha
of cassava cultivated per farmer, with a range from 0.50 to 1.00 ha.
A major shortcoming of this study is that labor information was not
collected. In retrospect, we feel that it could have been collected by admin-
istering suitably framed questions on labor input on a field-by-field basis,
especially since the fields were measured using compass, tape, and ranging
poles. We have relied, therefore, on secondary data reported by Knipscheer
(1980). The accuracy of the following costs and returns analysis will depend
on the accuracy of the data, the extent to which the conditions under which
they were collected represent conditions in Ovia LGA, and the extent to which
our assumptions about the data are correct. Knipscheer reported a mean of
168 person-days/ha for production of cassava when no fertilizer was used.
Since it is not certain whether this was for production with improved or local
varieties, both the improved and local varieties are assumed to be produced
with the same amount of labor. This assumption may not be as subjective as
it sounds. Labor for land clearing, planting, and weeding is independent of
variety grown. Harvesting labor may, however, depend on yield, which varied
significantly between the improved and local varieties. While higher-yielding
varieties will require more labor to assemble the roots for transportation, they
may actually be easier to uproot because of their bulk. Labor employed in
processing activities, however, was ascertained from the farmers. They were
asked in group interviews to estimate the amount of labor required to process
400 kg of gari, or approximately a two-ton tractor-trailer load of fresh roots.
The transport cost used in this analysis is based on an average charge of 20
naira per 2-ton tractor-trailer load of fresh roots from the field to the gari mill;
in addition, there was a charge of 8 naira for transporting the resulting 400 kg
of gari from the mill to the market. The farmers reported an average
expenditure of 95 naira/year on purchase of farm tools, most of which are
replaced yearly.
The total cost for producing gari (Table 3) is about 25 percent higher per
hectare and 38 percent lower per ton for the production of improved varieties,
as compared with local varieties. The higher cost per hectare for improved


Journal for Farming Systems Research-Extension






IMPROVED CASSAVA VARIETIES IN NIGERIA


Table 3. Costs and Returns for Gari Production using Improved and Local Varieties
in Ovia LGA, 1986-87 Season (naira per hectare unless otherwise stated).
Item Improved varieties Local varieties
I. Harvest at 12 months,
sell at mean 1986 prices
Production labor 1,008 1,008
Processing 432 248
Farm tools 95 95
Transport to mill 195 112
Milling charge 195 112
Firewood 98 56
Sacks 23 13
Transport to market 78 45

Total cost 2,124 1,690
Total cost/tonne of gari 547 754

Gross revenue 2,744 1,568
Gross margin 1,646 700
Net revenue 618 -122
Net revenue/tonne of gari 152 -55

II. Harvest at 12 months,
sell at June 1986 price (net revenue/tonne) 313 106

III. Harvest at 16-18 months,
sell at mean 1986 price (net revenue/tonne) 181 26

IV. Harvest at 16-18 months,
sell at June 1986 price (net revenue/tonne) 341 185


varieties is a result of higher processing costs for the higher yield, while the
lower cost per ton is explained by lower fixed costs, such as the cost of tools
and labor (excluding processing labor), which are spread over a greater
output.
The price of gari at the time of this analysis (1986) was 700 naira/ton,
which was the mean price obtained by farmers using local varieties. The
average farmer who cultivated 0.82 ha would have earned a gross margin of
1350 naira if he produced using improved varieties, or 574 naira with local
varieties. Gross margin was estimated as gross revenue, less value of purchased
inputs.
The revenue estimates, however, depend on the reliability of labor data
since labor costs constitute about 70 percent of total costs. Furthermore,
revenue estimates are based on roots harvested at 12 to 15 months. Farmers


Vol. 3, No. 2, 1993






NWEKE


sometimes harvest older cassava. Using our survey data, yields of gari would
be 4.83 ton/ha for improved and 3.22 ton/ha for local varieties for 16- to 18-
month-old cassava. In addition, some of the cassava is harvested, processed,
and marketed at periods when prices are higher than the average used above.
As for most food products, the price of gari exhibits seasonal variability
reflecting variations in the availability of close substitutes and in the availability
of harvesting and processing labor. Gari prices usually reach their peak in June
during the period of scarcity of other starchy staples, such as yams and
plantains. At that time, women also are busy with the weeding of the fields.
In 1986, the price of gari in June was 860 naira per ton, compared with the
average for the year of 700 naira.
The sensitivity analysis reported in Table 3 shows that the net revenue
would more than double if gari were sold mainly in June and the roots were
harvested at 16 to 18 months. Delaying harvesting results not only in higher
net revenues, but also in narrowing of the net revenue gap between improved
and local varieties.


CONCLUSION

The improved cassava varieties appear to have fitted very easily into farmers'
existing cropping systems. The use of improved varieties has not induced any
change in the intercropping farming patterns in the Ovia LGA. Farmers were
able to interplant the improved varieties with maize, melon, and other crops
as they do with local varieties. The improved cassava varieties performed better
than the local varieties in terms of root yield, cost reduction, and revenue
generation. Because of these advantages, and because cassava production is
the primary commercial activity of the farmers in Ovia LGA, the improved
varieties are widely adopted in the area. The improved varieties appear to be
attaining their high yield potential in the farmers' fields in spite of weed
infestation and suboptimal plant density, which is the result of the farmers'
practice of intercropping.
Labor may be a major constraint to expansion in the production of cassava.
Labor is the major item of cost because all production and all processing
operations, except milling, are performed manually. Production costs may,
therefore, be reduced through the mechanization of more processing opera-
tions. Weed-control technology will reduce the labor input and likely will
increase returns from the planting of the improved cassava varieties.


Journalfor Farming Systems Research-Extension







IMPROVED CASSAVA VARIETIES IN NIGERIA


REFERENCES

Akinlosotu, T.A. 1983. Studies on the control of cassava mealybug (Phenacoccus
manihoti) and green spidermite (Mononychellus tanajoa) in southwestern Nigeria.
Journal of Root Crops 9(12):33-43.
Dorosh, P. 1989. The economics of root and tuber crops and plantains in Africa: An
overview. Ibadan, Nigeria: International Institute of Tropical Agriculture.
Eberhart, S.A., and W.A. Russell. 1966. Stability parameters for comparing varieties.
Crop Science 6:36-40.
Grilliches, Z. 1957. Hybrid corn: An exploration in the economics of technological
change. Econometrica 25(4).
Hahn, S.K., E.R. Terry, K. Leuschner, 1.0. Akabundu, C. Okali, and R. Lal. 1979.
Cassava improvement in Africa. Field Crops Research 2:193-226.
Herren, H.R., P. Neuenschwander, R.D. Hennessey, and W.N.O. Hammond. 1986.
Introduction and dispersal of Epidinocarsis lopezi (Hom, Encyrtidae), an exotic
parasitoid of cassava mealybug, Phenacoccus manihoti (Hom, Pseidocoaidae), in
Africa. Agricultural Ecosystems and Environments (in press).
Hildebrand, P.E. 1984. Modified stability analysis of farmer-managed on-farm trials.
Agronomy Journal 76:271-4.
International Institute of Tropical Agriculture (ITTA). 1976. Annual report. Ibadan,
Nigeria: International Institute of Tropical Agriculture.
International Institute of Tropical Agriculture (ITTA). 1986. IITA research: Priorities
and strategies, 1988-2000. Ibadan, Nigeria: International Institute of Tropical Agri-
culture.
International Institute of Tropical Agriculture. 1987. Root, tuber and plantain improve-
ment program. Annual report. Ibadan, Nigeria: International Institute of Tropical
Agriculture.
Knipscheer, H.C. 1980. Labor utilization data for selected tropical food crops (maize,
cassava, yam, and upland rice). Agricultural Information Bulletin 6. Ibadan, Nigeria:
International Institute of Tropical Agriculture.
Nwanze, K.F., K. Leuschner, and H.C. Ezumah. 1978. The cassava mealybug, Phenacoc-
cus sp, in the Republic of Zaire. PANS 25:125-30.
Ohunyo, P.U. n.d. Agbarho farm from 1974 to December 1981. Mimeograph.


Vol. 3, No. 2, 1993







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