Front Cover
 Title Page
 Crops research in Ecuador
 Demonstrations of on-farm research...
 Institutionalization of on-farm...
 Concluding comments

Group Title: CIMMYT economics program working paper ; 01/83
Title: Creating an on-farm research program in Ecuador
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00080091/00001
 Material Information
Title: Creating an on-farm research program in Ecuador the case of INIAP's production research program
Series Title: CIMMYT economics program working paper
Physical Description: 28 p. : ill. ; 28 cm.
Language: English
Creator: Moscardi, Edgardo Ruben
Publisher: Centro Internacional de Mejoramiento de Maiz y Trigo
Place of Publication: Mexico D.F
Publication Date: 1983
Subject: Agriculture -- Research -- Ecuador   ( lcsh )
Genre: non-fiction   ( marcgt )
Spatial Coverage: Ecuador
Statement of Responsibility: Edgardo Moscardi ... et al.
 Record Information
Bibliographic ID: UF00080091
Volume ID: VID00001
Source Institution: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 38465952

Table of Contents
    Front Cover
        Front Cover
    Title Page
        Title Page
        Page 1
        Page 2
    Crops research in Ecuador
        Page 3
    Demonstrations of on-farm research procedures
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
    Institutionalization of on-farm research within INIAP
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
    Concluding comments
        Page 27
        Page 28
Full Text

The Case of INIAP's Production Research Program

Edgardo Moscardi, CIMMYT*
Victor Hugo Cardoso, INIAP
Patricio Espinosa, INIAP
R6mulo Soliz, INIAP
Ely Zambrano, INIAP

CIMMYT Economics Program Working Paper, 01/83

* Joined University of Florida, mid 1982.
The views expressed in this paper are the authors and do not necessarily
represent the official views of CIMMYT or INIAP.


The report which follows describes the evolution of a
new research activity within Ecuador's National Institute
for Agricultural Research (INIAP). The entity, the Produc-
tion Investigation Program (PIP), concentrates on on-farm
research. This report describes the research procedures
which orient a part of Ecuador's PIP research effort. The
research itself has certain distinguishing characteristics.
It involves collaboration between biological scientists and
social scientists (for the most part economists), it focuses
attention on the needs of representative farmers, and it is
area specific. The tone of the paper is one of advocacy as
CIMMYT believes that collaborative, on-farm, area-specific
research, focused on the needs of representative farmers, is
an essential step in the development of effective

Ecuador is one of the countries in which the CIMMYT
economics program staff cooperated closely with national
researchers in on-farm research activities. Initial contacts
were made in 1976 and, over the course of the next several
years, both INIAP and CIMMYT made substantial commitments to
the development of effective guidelines for carrying out
farm level-research. The experience in Ecuador was augmented
by that in other countries where CIMMYT staff and colleagues
from national programs were actively engaged in such

The essential elements of the process which emerged
are: (1) the identification of potential research areas in
terms of national priorities, (2) the delineation of
tentative recommendation domains, (3) the organization of
exploratory survey work, (4) the implementation of more
intensive surveys where needed, (5) the pre-screening of
information to identify leverage points for biological
research, (6) the initiation of on-farm experimentation
under the conditions of representative farmers and oriented
by the survey process, (7) the adjustment of subsequent
experimentation in terms of yearly results, and (8) the
orientation of relevant experiment station research in terms
of the findings from survey work and from on-farm

The following paper shows how the work in Ecuador
evolved, how it has contributed to the development of
improved technologies for the maize growers of two
recommendation domains, and how the process has now been
institutionalized by INIAP. We believe that the INIAP
experience offers solid evidence of the utility of on-farm
research. Beyond that, INIAP provides one model of how such
research can be organized and administered within a larger
research program.

The paper is based on materials provided by the
authors. These were edited by CIMMYT, with the edited copy
then reviewed and corrected by the authors to arrive at the
current version. Work on the paper started in early 1982. At
that time Moscardi was CIMMYT's regional economist. Cardosa
and Zambrano were agronomists working in the Imbabura
region. Soliz and Espinosa, both economists, were the first
and second directors of PIP.

Similar reports, based on the experience of other
countries, will follow in the near future. We hope that
these reports will encourage an ever wider application of
on-farm research as decision makers see the utility of the
process and the alternative forms for its implementation.

Donald Winkelmann
Director, Economics Program



Agriculture is the predominant economic activity of the
Ecuadorian people. Approximately 43 percent of the national
workforce is engaged in agriculture, which provides 21
percent of the gross domestic product and 40 percent of all
export earnings. Yields of most food crops are quite low and
showed little or no improvement during the 1970s. National
Agricultural development goals focus on raising basic food
production with particular concern for improving the welfare
of small farmers, who dominate the agricultural sector in
terms of numbers but not in terms of their contribution to
national production.

The National Institute for Agricultural Research
(INIAP), established in 1962, is charged with the
organization and execution of a national research system to
improve the productivity of Ecuadorian agriculture. INTAP
has seven experiment stations throughout the country: four
on the coast, two in the highlands, and one in the eastern
Amazonian lowlands. Since its creation, INIAP has employed a
typical research organization with various research programs
and departments organized basically along disciplinary
lines. Research programs (wheat, potatoes, maize, coffee,
beef cattle, etc.) focus on genetic improvement; research
departments (soils, entomology, pathology, agricultural
economics, communications) play a supporting role as the
different commodity-oriented programS develop improved
cultivars and races of livestock. The research programs,
supported by disciplinary research activities of the various
departments seek to develop improved "technological
packages" for Ecuador's major corps and livestock species.

Until 1976, INIAP scientists had been engaged in
research at two levels: experiment station research and
regional trials conducted in farmer's fields. Regional
trials sought to test technological components--varieties
and agronomic practices developed on experiment
stations--under varying soil and climatic conditions at the
farm level to determine yield potential and first
approximations of production recommendations. Normally,
these regional trials were placed on relatively large farms
to assure adequate management and to obtain datA with
reasonably high statistical confidence levels. Based on
these regional trials, packages of recommendations were
formulated and farmer field days were organized to extend
recommended technologies to farmers.

Beginning in 1976, INIAP added a third level of
research centered directly at the farm level. This research
featured farm-level efforts to determine the production

circumstances facing farmers in different production regions
and a series of on-farm experiments carried out on the
fields of "representative" farmers under their conditions.
This effort was to aim at developing and verifying
technologies appropriate to the needs of representative
farmers. A number of factors prompted INIAP to expand its
research system to include on-farm research. The over-riding
concern stemmed from the fact that little impact on
productivity was occurring from research on the basic food
crops, particularly within the dominant small-farm sector.
The absence of impact on basic food production led INIAP
research leaders and others to question whether appropriate
technology indeed existed for those basic food producers.
Beyond that, it was asked if the current research approach
could be expected to lead to effective technology for most
of Ecuador's farmers. These apprehensions were particularly
evident in highland maize and wheat production, where yields
on the total area planted to these crops were either static
or on the decline. Indeed, part of the initial interest of
INIAP's leaders in on-farm research was motivated by their
desire to ascertain the suitability of the Institute's
existing recommended technologies for cereal production,
particularly for small farmers.


After discussion within INIAP, several highland farming
areas were identified in which maize was an important crop
and where small farmers occupied most of the land. Moreover,
a range of biological and economic circumstances indicated
that different technologies were required to serve the needs
of different farmers in these regions. An area amounting to
50,000 ha in the provinces of Imbabura and Pichincha was
selected as the first site to carry out on-farm research
aimed at representative farmers. Although a wide range of
crops were grown in the area, maize was a dominant crop in
the farming system, covering 30 percent of the target
research area. The preferred maize in this region had a
large, soft, floury-type grain and was generally grown in
association with climbing beans. The region was within the
research responsibility of the Santa Catalina experiment
station, which provided vehicles, equipment, inputs, and
personnel services to the on-farm researchers. Funding was
largely supplied through a loan already made to INIAP by
the Inter-American Development Bank (IDB). This was later
supplemented by funds from the Swiss government.

By 1978, the importance of having the principal on-farm
research staff live within the research area was recognized.
Such a stationing at the local level represented a departure
from the existing modus operandi of INIAP in which virtually
all of the scientific staff were stationed at research
stations located near the larger cities and towns. The funds

made available by INIAP/IDB to carry out the project
permitted a special allowance to cover the costs of
relocating the project staff and their family to the study
area. The project staff also received additional
compensation for them to work away from an INIAP experiment

The Survey Sequence

Scientists from the maize improvement program based at
the Santa Catalina experiment station and from INIAP's
department of economics, along with CIMMYT economics staff,
began their research efforts in 1976 by travelling in the
area and talking with farmers and store keepers and others
about the various problems associated with maize production
and marketing within the region. New information generated
by this "exploratory survey" combined with the existing
knowledge of INIAP's maize scientists about production
problems in the target research area led to a preliminary
delineation of major aqroclimatic zones and farming systems.
This was a first step toward the formation of recommendation
domains, and the tentative identification of key questions
and hypotheses for a formal survey questionnaire to be
administered to a random sample of farmers in the area. Six
recommendation domains were tentatively identified in the
study area using criteria such as altitude, precipitation,
and soil type. Aerial maps from Ecuador's Military
Geographic Institute with a scale of 1:50:00 were utilized
to randomly select the land parcels and farmers for this
survey. A random sample of 230 farmers was identified and

The questionnaire focused on issues related to maize
technology along with immediately competitive and
complementary activities and sought to identify research
opportunities to increase the productivity of farmer
resources. Farmers were asked questions clustered around
eight groups of variables:

1. cropping patterns and practices;

2. storage practices;

3. consumption;

4. crop marketing and input acquisition;

5. agroclimatic characteristics of the

6. socioeconomic characteristics of farm

7. significance of institutional factors,
such as access to information and technical
assistance on production practices; and

8. an evaluation of farmer preferences for
earlier-maturing maize varieties.

Two teams of interviewers (two scientists per team)
were organized to undertake the formal survey. The
interviewers were first-degree agronomists (agricultural
technicians) who first received three weeks of training
before beginning their formal survey work at the Santa
Catalina station in survey procedures and in the production
and physiological aspects of the floury maize crop. In a
period of roughly two months, these technicians completed
their interviews with the farmers selected for the survey.

INIAP's department of agricultural economics staff did
the coding of the questionnaires and the data were then
analyzed at CIMMYT, Mexico using computer programs developed
to analyze such survey information. A report on the
conclusions of this survey activity, titled "Production
Practices in Ecuador," was prepared in 1977. This report
provided a statistical analysis of the groups of variables
included in the survey questionnaire.

A second survey was undertaken in 1979-80 to re-examine
a number of the factors originally considered in the 1976
survey. In particular, maize prices, marketing patterns,
input prices, and nutritional issues were re-examined.
Results of this survey reconfirmed the importance of maize
for home consumption, identified several quality issues
related to local preferences for white and yellow grain for
different food products, and reconfirmed that farmers were
highly interested in early-maturing maize and bean varieties
that would allow for an intensification in cropping

Delineation of Recommendatation Domains

Data from the informal and formal surveys confirmed
that the two provinces contained at least six different
recommendation domains. As each recommendation domain would
require on-farm experimentation and research resources were
scarce, it was decided to concentrate attention on the three
domains found in the Imbabura province. About half of the
respondents from the survey were located in these three
recommendation domains and maize was particularly dominant
in their farming systems.

Considerable agroclimatic variation existed among the
112 farm surveyed in Imbabura. While some differences in

economic characteristics were identified, biological factors
remained dominant in the delineation, and soil type were the
major criteria used in their formation.

Average Average General
Recommendation Domain Altitude Precipitation Soil
(meters) (mm) Type

Ibarra 2,300 650 Clay
Cotocachi 2,350 700 Sandy
Otavalo 2,500 855 Loam

Other factors also were considered in the delineation
of the three recommendation domains, e.g., the proportion of
level land, availability of supplemental irrigation,
incidence of pest problems, and whether maize was grown as a
monocrop or in association with climbing beans. Table 1
describes a number of the circumstances and practices of
farmers in the three recommendation domains.

Prescreening Components for On-farm Experimentation

Based on the anylisis of the survey information for
each recommendation domain, researchers identified key
factors as relatively more important for investigation in
the on-farm experimentation phase. The importance of variety
to the farmer was particularly clear. The preference
expressed for earlier-maturing maize varieties cut across
all three recommendation domains. Only in Otavalo, the area
with the best soil, flattest lands, and highest moisture,
was there a lower preference for earlier-maturing varieties.
However, even in Otavalo 65 percent of the respondents were
interested in obtaining short-season varieties. For the
three domains, 80 percent of the respondents signaled a
strong preference for shorter-season-varieties suitable for
a maize-climbing bean association and which could allow for
the introduction of an additional short-season crop (such as
peas or chickpeas) into the cropping pattern.

Of the respondents who named early-maturity as a
desired varietal characteristics, 60 percent also said that
they would be willing to sacrifice some yield in a variety
that could shorten the growing cycle by four to five weeks.
Maize breeders used this information as a selection criteria
for maize varieties to be tested in the on-farm trials.
Breeders sought to identify improved varieties that were
30-40 days earlier-to-maturity than traditional varieties
and that yielded within 80-85 percent of the available
long-season varieties.

Insect control was a second potential element
considered by the research team as they pre-screesed

Table 1. Important circumstances identified in the Imbabura province farm-level survey,



Area in






Sfor Earlier

--p-------- pe recent of f a rmers ------

Ibarra 53 1.17 93 75 2 10 60 90

Cotacachi .33 1.46 70_ 40 -- 11 35 -85 -

Otavalo 26 2.40 100 35 8 40 38 65

possible technological components to be included in the
on-farm trials. Although most of the survey respondents said
that they had problems with insects (mainly corn ear worm),
most did not consider insect damage a major problem. Using
the survey data along with the information from the
department of entomology at Santa Catalina station, it was
estimated that the use of an insecticide to control corn ear
worm could increase yields by at least 15 percent. With the
existing average yield of 1.5 t/ha, such control would
result in a yield increase of 225 kg/ha of grain. Although
this estimated benefit from insecticide use was conservative
(e.g., with the simultaneous use of other inputs such as
fertilizer and improved varieties, the impact on grain
production of using insecticides would probably be higher),
the research team calculated a criterion to evaluate the
different insecticide treatments in multifactorial trials
included in the on-farm experimentation phase. Assuming a
need for a 25 percent return to investment to cover capital
costs and risk factors associated with chemical control of
insects, it followed that acceptable insecticide treatments
should have an equivalent cost per hectare of 180 kg of
grain or less to be attractive to representative farmers in
the region. This finding guided selection among alternative
insect control strategies.

A third element considered during the pre-screening
process was fertility. With many farmers using manure and
few applying chemical fertilizer to maize (although many
were applying it to potatoes), the possibility for
increasing productivity through the use of chemical
fertilizer was evident.

Weed control was a fourth element considered. Many
farmers used weeds to feed "cuyes", a small animal raised
for sale and for meat. The question was, how did weed
control affect yields and what would be implied for meat

In addition to the examples described above on how
survey data were used to prescreen priority technological
components for on-farm trials, the survey also provided the
research team with information on the most common
(representative) cultural practices employed by farmers in
the three recommendation domains. Through the survey
sequence, researchers knew how farmers planted their maize,
how they fertilized their crop, and how they controlled
their weed populations. They knew about planting dates,
rotations, the equipment available, and common uses of
maize. These findings identified representative farmers, set
the levels for non-experimental variables, and influenced
the levels over which experimental variables ranged.

Selection of On-farm Trial Cooperators

Data from the survey were next used by INIAP
researchers to identify a number of on-farm trial sites
which were representative of the agro-economic conditions in
each recommendation domain. Survey respondents had been
asked whether they were willing to collaborate in on-farm
trials through the contribution of a small amount of land
and labor for experimental purposes. A surprising number had
indicated an interest in participating. Consequently,
researchers already had a list of interested cooperators
from which to choose. The survey had indicated that
representative farmers in Imbabura grew their maize in
association with beans. Beyond this, representative farmers
prepared their fields with oxen, used manure if they
fertilized at all, typically sold only a small portion of
their product at the market place, and were involved to a
considerable extent in off-farm employment. These factors
were explicitly considered in the selection of collaborating
farmers for the subsequent on-farm trials and established
the levels at which non-experimental variables were fixed.

Initiating On-farm Trials

With the survey data analyzed and discussed by
biological scientists and economists, and with the priority
research issues related to early-maturing varieties,
fertilization, and insect and weed control identified
through the "pre-screening" activity, the INIAP on-farm
researchers initiated their first series of exploratory
trials on farmer's fields.

Since farm-level trials (e.g., regional trials) were
not new at INIAP, the director general of the institute
called a meeting of interested staff to explain that one
objective of the first round of on-farm trials in the
Imbabura on-farm research project would be to validate,
economically and agronomically and under representative
farmer conditions, the technology already available and
recommended by the Santa Catalina station staff for the area.
Consequently, validation of available INIAP technology, as
well as actual technology generation, per se, was set as one
goal of the first cycle of trials.

Four different types of trials were designed for
the Imbabura on-farm experimentation phase, initiated during
1977-78, and carried forward in subsequent years. These
included variety trials originally with maize, beans, and
later peas and lima beans; multiple factor interaction
trails to identify critical interactions among management
factors associated with maize production; fertilizer levels
trials to identify appropriate fertilization recommendation;
and verification trials to evaluate potential technology

recommendations identified through the on-farm research.
Because of the interest in evaluating existing INIAP
production recommendations, technology verification
experiments, in addition to exploratory research trials,
were undertaken from the outset. Descriptions of these
experiments follow.

Variety Trials

Maize variety trials were carried out in the three
recommendation domains during the first two years of
experimentation. Five maize varieties plus the farmer's
variety were included in these trials. Of the improved INIAP
materials, two were long-season varieties and three were
short-season varieties. In all cases, maize and beans were
grown in association. Table 2 shows the results of the
variety trials grown during the first cycle of
experimentation in two of the three recommendation domains.
The three short-season varieties included in the trials had
acceptable grain types (both white and yellow) were 45-55
days earlier-to-maturity than the farmers' traditional
long-season varieties, but within the range of yield
acceptability previously determined by the survey, e.g.,
within 15 percent of the yield level of the local variety.
It was also apparent that the local climbing bean varieties
tested were too aggressive for the earlier-maturing maize
varieties, knocking them down with resulting yield losses.
The long-season improved maize varieties (INIAP 125, INIAP
126) were not significantly higher in yield performance than
the farmers' local varieties.

In the second cycle of experimentation, INIAP's grain
legume program staff provided eight short and long-season
bean varieties that were less aggressive in their vegetative
and climbing characteristics than the farmers local
varieties. These, then, were included in the variety trials
of that year. An important feedback from the cooperating
farmers was that yield potential (largely a function of
disease resistance) was much more important to them than
grain type and color.

The major conclusions from these varietal trials was
that early-maturity was a major requirement of many farmers
in the various recommendation domains. Further, the
earlier-maturing white floury variety, by then named INIAP
101, had been enthusiastically received by farmers. While
some grain quality problems had been identified with this
variety in the preparation of mote (a type of hominy), one
of the uses of white grain maize in the area, and in the
greater susceptibility (compared to local varieties) of
INIAP 101 to stored-grain pests, it was apparent that the
agronomic qualities of early-maturing maize were highly
desired by local farmers. This fact meant that INIAP should
continue research to develop such varieties.

Table 2. Yields obtained in the 1977-78 variety trials in
two recommendation domains

Domain Long-Season Varieties Short-Season Varieties

1/ Comp. Am. Comp
Local INIAP 125 INIAP 126 INIAP 1011 Harinoso Ccacahuazintle.


Ibarra 1.95 1.45 2.20 1.50 1.91 1.60
(two trials)

Otavalao 3.50 1.65 2.00 2.85 2.25 3.70
(one trial)

1/ Reselection made by INIAP from a Cacahuazintle composite
variety originally developed in Mexico and later named

Among the long-season varieties, INIAP 126 showed
promise for the Cotacachi recommendation domain. The maize
variety, Varios x Chillos, an intermediate maturity, yellow
grain material included for the first time in the second
year of trials, performed favorably at a number of
locations. However, the lack of uniformity, particularly
with respect to maturity, resulted in feedback to experiment
station researchers on the need for further improvement
before it was ready for release to farmers. On the other
hand, the performance of the variety INIAP 125, a
long-season yellow grain variety, was unsatisfactory for two
years in a row. This information was communicated to
scientists at the Santa Catalina experiment station, and
resulted in the removal of this material from INIAP's list
of recommended varieties.

Based on the trial results with respect to climbing
beans, the Santa Catalina legume improvement staff began to
intensify their work to develop a broader range of
earlier-maturing bean varieties with adequate levels of
disease resistance.

Multiple Factor Trials

These trials served to identify critical management
factors in maize production, their order ?f priority, and
their interactions. A factorial design (2 ) was used in
these trials and the interactions among the following four
factors were studied: varieties, weed control, fertilizer,
and insect control. In such trials, two levels (low level -
high level) are used for each factor (e.g., the farmer's
level of fertilization, INIAP's previously recommended
dosage; the farmer's variety, and INIAP improved varieties).
The results of the multiple factor trials prompted
considerable discussion between the on-farm research team
and scientists at the Santa Catalina station, since
recommended technologies only showed small positive
biological responses over the farmers' technology. For
example, the "complete" recommended technological package
showed responses up to 1 t/ha over the farmers' technology
and the "non chemical control" recommended technology only
showed reponses of up to 0.5 t/ha over the farmers'
technology. However, when the various treatment combinations
were submitted to economic analyses, they were generally
found to be only minimally profitable in each of the three
recommendation domains. Only in Otavalo, the most favored
environment of the three recommendation domains, was there a
substantial interaction and economic return from the
combined use of the recommended varieties, fertilizer rates,
and cultural practices. Interactions among the factors in
the factorial trials grown in the other recommendation
domains did not appear to be economically significant.

The weed control treatments included in these trials
showed that the chemical herbicide mixture used in the
maize/bean association gave control for less than four
months before an additional hand weeding was necessary. The
need for additional weeding to achieve adequate control made
the use of chemicals unprofitable, given the cost of
herbicides and the existing wage rate for hand weeding. An
economic sensitivity analysis applied to these data
indicated that the cost of hand weeding would have to
increase by 80 percent before chemical control could be
recommendable. In response to this analysis, scientists from
the department of weed control began studies to improve the
effectiveness of herbicide use. A number of off-station
trials were conducted to evaluate different application
methods, such as banding, in order to reduce the costs and
improve the benefits associated with the use of herbicides
for weed control.

A similar situation was observed in the -case of insect
control, especially for corn ear worm, the major insect pest
in the research area. The recommendation for corn ear worm
control (which was used as a treatment in the factorial
trials) was the application of an insecticide four times to
the ear silks using a backpack sprayer. However, the yield
advantage of this practice was not substantiated in the
factorial trials. Follow-up discussions with technicians who
carried out the recommended treatment in the on-farm trials
revealed that it was difficult to carry out insecticide
applications with the backpack sprayer when the maize/bean
crop stands were as tall and dense as they were at the
recommended period for treatments. The result was that an
insufficient amount of the active chemical agent reached the
stigmas of the ears. As a consequence of this observation,
the entomology department began to test other application
methods in order to determine whether a more simple
insecticide application method could be developed which
could give more effective control with smaller doses of the
active agent than was possible with the previous method.

Fertilizer-Levels Experiments

These experiments serve to identify economic levels of
fertilizer use. The trials included various varieties and
levels and combinations of fertilizer nutrients and
recommended application methods. Other management practices
were set at the levels normally used by the farmer. Only in
the Otavalo recommendation domain was an experimental
treatment (20 kg/ha N, 20kg/ha P 05 plus organic fertilizer)
more profitable than the farmers check treatment which only
included the application of organic fertilizer (manure). By
estimating the nutrients contained in the organic fertilizer
dosage, profitable application rates were determined to be
in the range of 50-80 kg/ha N, 20-40 kg/ha PO Based on
this series of trials, it was decided to set tne general

fertilizer recommendation when using improved varieties
(local varieties did not show an economic response) at
80-40-0 for subsequent verification trials.

Verification Trials

The purpose of these experiments was to validate the
technology presently recommended by INIAP as opposed to
current management technology and varieties of the farmer.
Results from these verification experiments, as well as
information on farmers reaction to the different
technologies, provided feedback to INIAP's experiment
station researchers on currently-available technologies.
Since verification trials are designed to evaluate a
relatively few number of treatments (technologies) for their
suitability for commercial production, larger plot sizes are
used than in factorial and fertilizer levels trials and the
farmer provides almost all of the management.

The research activities in the third and fourth cycles
of on-farm experimentation were primarily focused on the
verification of the results obtained from experiments the
first two cycles. A series of verification trials were
conducted in the three recommendation domains. The results
of the first two years of on-farm experimentation had led
researchers to believe that at least for the Otavalo
recommendation domain, the use of the short-season variety
INIAP 101 fertilized at 80-40-0 offered farmers a
significant economic return. Consequently, farmer field days
were organized to view the verification trial results and
the diffusion process for this recommended technology began.
In the Cotachachi recommendation domain, characterized by
sandy soils, a longer-season maize variety, INIAP 126,
showed good yield potential, especially when nitrogen was

For the Ibarra recommendation domain, no ciear
recommendation had emerged from the first three years of
experimentation. Therefore, the majority of the experiments
in the fourth on-farm experimentation cycle were centered in
Ibarra. The most promising variety for the Ibarra area was
the short-season yellow grain variety Varios x Chillos.
Although this variety had a lower yield performance
(3.15 t/ha) than the local check, it was in the same
maturity range of INIAP 101, and thus offered the potential
for double cropping. Figure 1 shows a net benefit curve
derived from the 1980-81 verification trails conducted in
Ibarra. When looking at the full farming system, the net
benefit to the farmer of using short-season maize and bean
varieties becomes significant because it brings the option
of intensifying the cropping cycle. When an additional crop
is added to the annual cycle, in this case peas, the annual
returns to labor and capital are significantly higher than
when the returns to the short-season maize variety are

Figure 1. Curve of net benefits in verification
trials for Ibarra recommendation domain, with
and without peas 1980-81




Vi F

(S. / ha )

= local variety
= INIAP 101
= farmer's fertilizer practices
= INIAP recommended fertilizer application (80-40-0)






j -



0 -




R ___ ______ ______________


- b I I I


Vo F

viewed in isolation. (Benefits for peas were confirmed
during the 1981-82 cycle.)

Continuing On-farm Experimentation and Technology Transfer

For the 1981-82 crop production cycle in Imbabura,
on-farm researchers directed their efforts toward three sets
of activities: demonstrations and technology transfer
associated with the use of the short-season maize variety
INIAP 101, a continuation of on-farm trials emphasizing new
activities associated with legumes and various other crops
planted in rotation with maize, and special studies to
monitor the acceptance and derived benefits from the
recommended technologies extended to date.

A promotional plan to distribute small quantities of
seed of the short-season maize variety INIAP 101 was
developed by the PIP research staff. INIAP's director
general then contacted the national seed production
organization (Empresa Mixta de Semillas) to provide two tons
of INIAP 101 seed packaged in 10 kg bags. Eventually,
150/10- kg bags were provided and these were sold to
interested farmers through Ministry of Agriculture offices,
mostly in Otavalo. A limit of two bags per farmer was set
and approximately 80 farmers bought the available seed. A
simple promotional brochure was prepared to explain INIAP
recommendations for using INIAP 101 in association with
beans. Farmers were advised to use the least aggressive
climbing beans varieties they could obtain, their own
planting density, and fertilizer if they could get it. In
order to monitor how these participating farmers took
advantage of the earlinesss" of INIAP 101 (e.g., what they
did with the extra days available for other crops and farm
activities), a sample of 30 farmers was selected for
follow-up surveys (results not available in mid-82).

In addition to these technology transfer activities, a
new cycle of on-farm experiments was planned for the three
recommendation domains in Imbabura. A total of 21 trials,
equally distributed across the three recommendation domains,
were planted, including new variety trials to evaluate
recently developed short-season maize and bean materials
emanating from the Santa Catalina station crop improvement
programs. The increasing availability of these short-season
maize and bean materials has resulted from new emphasis in
the breeding priorities of the maize and grain legume
improvement programs at the Santa Catalina station.

A number of new special studies were also included in
the 1981-82 on-farm research cycle:

1. A study was initiated on production problems of
broad beans, a crop of secondary importance often

qrown in association with maize, and which is
heavily afflicted in Imbabura by disease and
insect damage. This crop has a maturity period
that also allows it to be used in rotation with
the short-season, maize-bean crop association. A
series of verification trials were also planted in
conjunction with the department of entomology and
pathology to evaluate the effects of the pest and
disease problems in broad beans.

2. Long-term fertilizer evaluation trials were
initiated to study the effects on soil fertility
of the crop rotation pattern of a maize/bean
association followed by peas. The objectives are
to develop a fertilizer recommendation for this
crop rotation sequence, and to quantify more
explicitly the economic returns of the
maize/bean-pea rotational pattern.

3. New verification trials were planted at the
request of the Santa Catalina entomology staff to
re-evaluate the economic utility of the ear worm
control methods formulated by the department of

4. Several experiments on methods to reduce grain
losses from stored-grain pests were also under
way, again at the request of the departments of
entomology and agricultural engineering. In these
storage experiments, the effectiveness of the
department's recommended grain treatment (1%
Malathion) is being evaluated under farm-level
storage conditions.

5. A more systematic method of market data collection
in Imbabura was also added to the work as
responsibilities of the on-farm research team.
This information will now be collected on a
regular basis for use in production cost studies
conducted by INIAP's department of agricultural

Additional Comments on the On-farm Research Experience in

Several additional points need to be made about the
on-farm trials conducted in Imbabura Province during
1977-82. The first is that, throughout the process,
station-based and off-station researchers collaborated in
designing trials and in interpreting results. Specialists
from the maize improvement program and the economics
department at Santa Catalina oriented the early on-farm
research work in Imbabura. They were soon joined by station

specialists from other departments. Always, of course, the
guiding principle was the farmer's requirements, and these
were the points of departure for interaction the INIAP
researchers. Later, with the formation of the PIP (see
below) this collaboration took in a somewhat different form.

For the most part, the trials were managed by farmers.
Only those activities directly related to the experimental
variables included in the initial exploratory trials were
managed by the researchers. A second point is that the
researchers, themselves, learned by doing, refining their
methods over the years. The case of planting densities is a
good example. At the outset, planting densities for variety
trials were set at the level of the "average" for the
recommendation domain. As the work progressed, researchers
noticed that, while the average stand densities of
cooperating farmers approximated the original averages
estimated during the survey phases, there seemed to be
systematic differences among the cooperating farmers. More
careful observation showed that plant stands were closely
related to fertility, even within a recommendation domain.
While this variation was not large enough to warrant the
delineation of new recommendation domains, it did induce
researchers to adopt the practice of working alongside each
trial cooperator during the planting stage so that the plant
stand in each trial varied as did that of the farmer in his
normal production fields.

From year-to-year, about 40 percent of the trials were
lost due to a number of factors. Drought, in particular, was
a serious problem in a number of years. Supplemental
irrigation systems also turned out to be an unexpected
problem. The 1976 survey sequence had revealed that 40
percent of the farmers interviewed had access to some form
of supplemental irrigation. Consequently, about 40 percent
of the on-farm experiments were situated on similar fields.
Many of these trials, however, were eventually abandoned.
Upon closer examination, the research team found that the
supplemental irrigation systems on these lands were so
deficient that they were almost totally inadequate for
irrigation purposes in seasons of unusually low rainfall.
Trials were also lost due to causes other than drought. Such
losses occurred because of theft of early-maturing maize at
the green cob stage, problems of land tenure of the
cooperator, and bird and livestock damage.

The high number of trials which were "lost" during the
on-farm experimentation cycles was a matter of great concern
to some researchers at the Santa Catalina experiment
station. Further, there was considerable concern about the
high coefficients of variability in the data obtained from
those trials that were harvested. Very few trials had
statistical confidence intervals at the 95 percent level.
Rather, the confidence intervals in most trials were in the

75 to 90 range, considered very low by experiment station
standards. These factors led to strong doubts among some
INIAP researchers about the utility of cooperating with
representative farmers, about the cost-effectiveness of this
type of on-farm research, and about the confidence one could
place in the results.

The on-farm researchers contended that the loss of
trials and the lower statistical confidence intervals were
the real costs implicit in any attempt to obtain valid
information at the farm-level on the performance of
alternative technologies. With respect to statistical
confidence levels, they held the position that in research
to formulate production recommendations, the critical
determination to be made was whether the various
technologies under investigation increased the net benefits
to the farmer within acceptable risk levels. Such
determinations, they contended, could be made using on-farm
trial data even with higher coefficients of variability. No
consensus was reached among the INIAP researchers engaged
in this debate regarding these diverging points of view.


INIAP's current national production research program,
formally established in 1979 (known by its Spanish acronym,
PIP), can trace its origins to the Imbabura project and to
an Inter-American Development Bank (IDB) loan made to
Ecuador for INIAP in mid 1977. This US$11 million dollar IDB
loan, along with US$5.9 million from the Ecuadorian
government, and additional funds from the Swiss government in
1978-79, provided INIAP with funds to reinforce the
institute's on-going research programs as well as to
establish a production research program focused on the small
farmers of the country. Few details about the form this
production research program should take were spelled out in
the original loan document, except that a "system of
technology transfer will be introduced through specialists
in production."

In 1977, during the initial phase of the new
Ecuadorian effort in on-farm research, it was considered
desirable to obtain experience with several production
systems for important food crops in which Ecuadorian
production was deficient or had the potential for expansion.
The existence of previous research and the availability of
results that might be used in the program were also taken
into consideration in selecting research priorities.
Originally, maize, wheat, rice, potatoes, and dairy
production systems were selected for on-farm research
attention. Similar research projects for other important

Ecuadorian crops and livestock systems were soon launched in
other small-farmer areas of the country.

Creation of the National Production Research Program (PTP)

As results from the on-farm research emerged, INIAP
leaders were thinking about alternative ways to
institutionalize the activity. Although TNIAP's existing
organization structure worked well in the development of
some technologies, it was not at its best when research
activities around of complex production systems had to be
integrated, as in the case of Imbabura with various
associated and multiple croping patterns.

In 1979, INIAP decided to establish the national
Production Investigation Program (PIP) with its own
personnel especially trained in the on-farm research
procedures previously described (see figure 2). An early
document about the newly created PIP defined its objectives
in the following way:

Definition: PIP is a technology transfer program for
the investigation of production
constraints and production opportunities
on farmers' fields and focuses on
farming systems.

Objectives: 1. Screen and test on the farmer's own
field those technological components
that are being generated in the research
support departments and crop programs of
the experiment stations, for immediate
adaptation or adjustment to the
principal farming systems of a region.

2. Provide farm-level feedback which
will orient and guide the research which
is being carried out at the experiment
stations. This feedback will give rise
to the development of new technological
components which respond to problems and
limiting factors detected among the
farmers of a region.

3. Formulate alternative technologies,
subject to economic validation, which
can be made available for subsequent
technology transfer by the extension
service and agricultural credit

Figure 2. Structure of the Production Investigation Program (PIP) within the
overall INIAP structure

- Wheat Soils
- Maize Weed Control
- Potatoes Nutrition
- Legumes Entomology
- Etc. Agricultural Economics
I Etc.

- Cattle
- Swine
- Poultry
- Etc.


By 1980, PIP production research programs were under
way in 10 major production environments and ecological
regions of the country with 18 on-farm researchers directly
assigned to the program (see table 3). These included PIPs
for the lowland Pacific coastal areas,the Andean highland
valleys, and the piedmont areas of the Amazon basin.

A key manpower deployment characteristics of the PIP
program has been that the production researchers live in the
region selected for their research so that they can
establish close contacts with local farmers and with the
community, and above all, so that they can obtain a better
perception of the most important production problems and
research needs facing area farmers. Researchers assigned to
each PIP are provided with vehicles and fuel for travel in
the research area. Other inputs needed for the trials are
supplied by the various INIAP experiment stations which
support each PIP project area. A special incentive system
also has been established for the PIP researchers, with the
salary depending on geographic location, stipends for living
expenses, and the same opportunity as experiment station
researchers for postgraduate studies after two or three
years of service. The interest of young INIAP personnel in
the PIP has been extremely high.

The introduction of a new level of agricultural
research into an established institute, especially an
adaptive research program with cuts across different crops,
research disciplines, and agricultural organizations, is not
without its institutional difficulties and frictions.
Frequent coordination meetings are necessary with the staff
of the commodity programs and research support departments
to establish and review goals, objectives, and strategies,
and to determine the respective responsibilities of the
production research teams and the experiment station
The steady interest and participation of the experiment
station research leaders is also of great importance for the
success of a program such as the PIP, since considerable
administrative support is required from both the central
offices of INIAP and the regional experiment stations for
the on-farm researchers to carry out their work effectively.
Considerable attention has been paid to the delineation of
responsibilities among the various research programs and
departments and the PIP. The feedback mechanism between the
PIP and the crop programs and research departments of the
experiment stations has been considered as a fundamental
dimension in the institutionalization process. The PIP has
been given the primary responsibility of identifying farmer
requirements for new technology and the experiment stations
have been given the responsibility of generating new
technological components in response to those requirements.

Table 3. INIAP's production research programs, 1977-80

Name and









Puerto Ila-chone



Maize, Beans



Maize, Cassava,

Maize, Cassava


Maize, Beans

Maize, Castor
Bean, Pumpkin

Coffee, Cocoa,

Coffee, Cocoa,

Exp. Station

Sta. Catalina

Sta. Catalina




Sta. Catalina

Sta. Catalina



Sto. Domingo

Exp. Station (s)

Sta. Catalina

Sta. Catalina




Sta. Catalina

Sta. Catalina















The established research programs and departments
concentrate most of their research work on INIAPs experiment
stations. These researchers also continue to carry on
regional trials under varying ecological conditions to
determine the yield potential of the technological components
developed on experiment stations. The PIP has a more
explicit responsibility for the actual formulation of
production recommendations for defined recommendation

The complementary nature of the PIP to the experiment
station research programs and departments has been strongly
emphasized. Through a production research program, a
substantial number of professionals achieve a better
understanding of the problems and needs of farmers, and
become more effective in generating and disseminating
alternative technologies for the improvement of the welfare
of rural families. In order to strengthen communication and
coordination between the PIP and INIAP's experiment station
programs and departments, the plans of various PIP trials in
each region have been reviewed by the technical committees
of the experiment stations which support a particular PIP
project. These technical committees, in existence for a
number of years, meet once a week to consider and approve
specific research proposals prepared by INIAP's various
programs and departments. Annual research plans of work for
the PIPs also are reviewed and approved by these committees.
A system of quarterly reporting by the PIP field staff has
also been formalized so that feedback information from the
on-farm trials is regularly sent to the experiment station.

The PIP is also seeking to strengthen the respect and
trust between researchers and extension agents. Through the
PIP, a more integrated and stable working relationship,
based upon collaborative field work, has emerged between
research and extension. Extension and researchers cooperate
in production surveys, selection of on-farm cooperators and
trial sites, evaluation of the research data, and
dissemination activities associated with recommended
technologies. Additional efforts to strengthen and
consolidate the relationship of research and extension are
also being made through the establishment of common training
and technology evaluation activities.


From the first cycle of on-farm experimentation in the
Province of Imbabura in 1977, it was noted that the
station-based INIAP scientists tended to favor experiments
with many variables and levels for each component in the
study, following traditional station research methodologies.
It was also evident that few of INIAP's researchers were
trained to identify non-biological factors influencing the

small farmer. For these reasons, training in on-farm
research procedures has been a central activity since the
inception of this research program.

Through 1982, two types of training have been offered
to INIAP's production researchers. Each year a few INIAP
scientists have attended CIMMYT's in-service production
agronomy training courses in Mexico. In addition, a strong
and on-going national in-service training program was
developed to provide the growing number of production
researchers with the necessary skills to undertake programs
of on-farm research. This training goes on each year and
features a unique system of bringing together participants
for short periods during critical stages of the research

The key skills taught in the current version of the
training program are:

1. Training in the identification (through a sequence
of surveys) of the biological and economic
circumstances affecting the small farmer.

2. Training in basic production systems, with the
trainees carrying out all phases of the
cultivation practices.

3. Training in basic agricultural research
methodology, principally in experimental design
and on-farm execution for verification of

4. Training in methods for economic analysis of
experimental data to formulate production

INIAP plans to continue to utilize different individual
PIP research projects each year as in-service training sites

for newly hired on-farm researchers, as well as for
extension agents engaged in technology transfer activities.
With the system of training used, on-going production
research and training can be combined into an effective
system for production workers to gain practical experience
in the PIP research methods.

In 1981, USAID committed funds to Ecuador to help INIAP
consolidate and support the PIP during 1982-85. Five of the
ten PIP project areas are funded under this agreement.
Included in the grant are funds for sending PIP researchers
for graduate training at the M.S. degree level.


The process of institutionalization of the PIP is still
under way. A variety of coordination problems are still
unsolved. The PIP staff are responsible in technical and
administrative matters to heads of the various INIAP
experiment stations. As well, their activities are
coordinated by the national PIP coordinator based in the
central INIAP offices in Quito. The current national PIP
coordinator is also head of the Agricultural Economics
Department. Because of his dual responsibilities as well as
budget constraints, the PIP coordinator is not able to visit
each of the PIP field programs more than three times a year.

Although the PIP researchers are administratively
attached to an INIAP experiment station, their work is
carried out away from the station and they operate with a
considerable degree of independence in scheduling
activities, a situation which has been criticized by some
experiment station heads who feel that inadequate
supervision is being exercised over PIP personnel. Closer
supervision, however, is occurring in those PIP field
programs that are integrated into the national integrated
rural development projects. In these cases, the PIP field
staff are administratively responsible to the project leader
of each integrated rural development project.

Another problem has been related to the PIP budgeting
situation. Much of the financing of the PIP up to 1982 has
been through special grants from international donor
agencies, and many of the.PIP personnel are still not
permanent INIAP staff members. Some vehicles and supplies
have been provided through these special grants. For most
logistical support (materials for experiments, spare parts
for vehicles reimbursements for gasoline, payment of salary)
the PIP field staff rely on their respective experiment
stations, where their requests are handled by the station
director. The PIP national coordinator, in addition to his
technical support responsibilities to the PIP field staff,
has had to devote a considerable portion of his time in
order to expedite administrative matters, particularly
related to payment of official expenses, rents, input
purchases, and vehicle repair associated with the field
research programs.

The PIP research staff tends to be young with most in
the beginning INIAP professional salary grades. Although
they receive the same basic pay as other INIAP staff of the
same grade, their rents are paid because they live in the
field. In addition, they receive a cost of living allowance.
This allowance was granted because the work of the PIP field
staff involved more risks than station work; PIP technicians
spend long hours on roads and paths that are in poor
condition, often work in isolated areas, and generally have

more irregular hours of work. Further the PIP field staff
generally cannot easily avail themselves of station
facilities and fringe benefits such as libraries, clinics,
and subsidized food that are readily available to the
regular station-based INIAP staff. Finally, PIP field staff,
because of the nature of their work, do not have the option
of supplementing their income through teaching as do some
experiment station researchers. Nevertheless, the PIP
allowances have been questioned by other INIAP staff.

The final point of contention is that of the role of
the PIP vis-a-vis experiment station research programs and
departments in the generation of technology. The question
basically centers around who should be responsible for the
final shaping of INIAP's production recommendations to
farmers--the PIP or the various crop improvement programs
and research departments. Some view the PIP as the final
step in the research process which leads to the development
of recommendations. Others see the PIP solely as an
off-station testing service unit for experiment station
research programs and departments. It should, however, be
clearly understood that the PIP is not a disciplinary
research activity of either biological or social scientists,
but rather a collaborative, multidisciplinary research
approach in search of alternative technologies that are
valid for Ecuadorian farmers.

After five years of experience, there is an emerging
consensus among INIAP scientists about the potential
contribution of the PIPs to national production-oriented
research. In the course of a few years, the work in Imbabura
has given evidence of the need for improvement in existing
recommendations and has pointed out the need for
earlier-season varieties, careful tailoring of fertilizer
recommendations to the agro-climatic circumstances of
defined areas, more effective insect control, improvements
in pea and broad bean varieties, effective weed control
technology, and research on water management. Equally useful
results are emerging from the work of other PIPs, e.g., work
on wheat technologies has pointed the way to substantially
different fertilizer recommendations and has opened new
issues for the relationship between soil tests and
fertilizer applications.

It was this flow of results and opportunities that led
INIAP leadership to develop the PIP program and to expand
its scope. Clearly, the contribution of the PIPs, through
on-farm research, points to a growing partnership for the
program within INIAP in the forging of more appropriate
production technologies for Ecuadorian farmers.


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