• TABLE OF CONTENTS
HIDE
 Front Cover
 Title Page
 Copyright
 Table of Contents
 List of Tables
 Acronyms and abbreviations
 Acknowledgement
 Executive summary
 Introduction
 The study area
 Methodology
 Maize production technology...
 Demographic and socioeconomic...
 Maize in the farming system and...
 Farmers adoption and disadoption...
 Soil fertility management
 Access to rural support servic...
 Logistic model estimates
 Conclusion
 Reference
 Annex






Title: Adoption of maize seed and fertilizer technologies in Embu District, Kenya
CITATION PDF VIEWER PAGE IMAGE ZOOMABLE PAGE TEXT
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00077471/00001
 Material Information
Title: Adoption of maize seed and fertilizer technologies in Embu District, Kenya
Physical Description: Book
Language: English
Creator: Ouma, James O.
Publisher: International Maize and Wheat Improvement Center (CIMMYT)
Publication Date: 2002
 Subjects
Subject: Africa   ( lcsh )
Farming   ( lcsh )
Spatial Coverage: Africa -- Kenya
Africa
 Record Information
Bibliographic ID: UF00077471
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: African Studies Collections in the Department of Special Collections and Area Studies, George A. Smathers Libraries, University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: isbn - 970-648-093-5

Downloads

This item has the following downloads:

embu ( PDF )


Table of Contents
    Front Cover
        Front cover
    Title Page
        Page i
    Copyright
        Page ii
    Table of Contents
        Page iii
    List of Tables
        Page iv
    Acronyms and abbreviations
        Page v
    Acknowledgement
        Page vi
    Executive summary
        Page vii
        Page viii
    Introduction
        Page 1
        Page 2
    The study area
        Page 3
        Page 4
    Methodology
        Page 5
        Page 6
        Page 7
    Maize production technology recommendations
        Page 8
    Demographic and socioeconomic characteristics
        Page 9
    Maize in the farming system and adoption of varieties
        Page 10
    Farmers adoption and disadoption of improved maize seed
        Page 11
        Page 12
        Page 13
        Page 14
    Soil fertility management
        Page 15
        Page 16
    Access to rural support services
        Page 17
    Logistic model estimates
        Page 18
        Page 19
    Conclusion
        Page 20
        Page 21
    Reference
        Page 22
    Annex
        Page 23
        Page 24
        Page 25
Full Text




Adoption of Maize Seed and

Fertilizer Technologies in

Embu District, Kenya





James 0. Ouma
Festus M. Murithi
Wilfred Mwangi
Hugo Verkuiji
Macharia Gethi
Hugo De Groote



October 2002






CIMMYTM.
INTERNATIONAL MAIZE AND Funded by the
WHEAT IMPROVEMENT CENTER European Union









Adoption of Maize Seed and


Fertilizer Technologies


in


Embu District, Kenya











James 0. Ouma
Festus M. Murithi
Wilfred Mwangi
Hugo Verkuiji
Macharia Gethi
Hugo De Groote*


October 2002






J. 0. Ouma and FM. Murithis are Agricultural Economists at the Kenya International Research Institute (KARl), Embu, and KARl headquarters in Kenya.
W. Mwangi is an Agricultural Economist with the International Maize and Wheat Improvement Center (CIMMYT) Economics Program, currently on leave of
absence. H. Verkuijl is an Agricultural Economist with the Royal Tropical Institute (KIT) in the Netherlands, and previously worked with the CIMMYT
Economics Program in Addis Ababa. M. Gethi is an Enthomologist with KARI-Embu, Kenya. H. De Groote is an Agricultural Economist with the CIMMYT
Economics Program in Nairobi, Kenya.



















CIMMYT (www.cimmyt.org) is an internationally funded, nonprofit, scientific research and training
organization. Headquartered in Mexico, CIMMYT works with agricultural research institutions worldwide
to improve the productivity, profitability, and sustainability of maize and wheat systems for poor farmers in
developing countries. It is one of 16 food and environmental organizations known as the Future Harvest
Centers. Located around the world, the Future Harvest Centers conduct research in partnership with
farmers, scientists, and policymakers to help alleviate poverty and increase food security while protecting
natural resources. The centers are supported by the Consultative Group on International Agricultural
Research (CGIAR) (www.cgiar.org), whose members include nearly 60 countries, private foundations, and
regional and international organizations. Financial support for CIMMYT's research agenda also comes
from many other sources, including foundations, development banks, and public and private agencies.

F U T U R E" Future Harvest builds awareness and support for food and environmental research for a
HARV/EST world with less poverty, a healthier human family, well-nourished children, and a better
environment. It supports research, promotes partnerships, and sponsors projects that bring the results of
research to rural communities, farmers, and families in Africa, Asia, and Latin America
(www.futureharvest.org).

I C IM M T International Maize and Wheat Improvement Center (CIMMYT)
2002. All rights reserved. The opinions expressed in this publication
are the sole responsibility of the authors. The designations employed in the presentation of materials in this
publication do not imply the expression of any opinion whatsoever on the part of CIMMYT or its
contributory organizations concerning the legal status of any country, territory, city, or area, or of its
authorities, or concerning the delimitation of its frontiers or boundaries. CIMMYT encourages fair use of
this material. Proper citation is requested.

Correct citation: Ouma, J., F. Murithi, W. Mwangi, H. Verkujl, M. Gethi, and H. De Groote. 2002.
Adoption of Maize Seed and Fertilizer Technologies in Embu District, Kenya. Mexico, D.F.: CIMMYT

Abstract: This study reviews socioeconomic and technical factors that affect the adoption of improved
maize and fertilizer in the Embu District, Kenya and the role of credit in both. A total of 127 farmers (82
adopters and 45 non-adopters) were interviewed for the study during the long and short rainy seasons in
1998 in the Nembure, Runyenjes, and Kieni Divisions in Embu District. Most farmers in the study area
used basal fertilizer. However, the use of fertilizer was below recommended levels. More adopters used
hired labor and had greater access to credit and extension services than non-adopters. The Pioneer
H3253 variety and 2-kg seed packages were found to be most popular among adopters. Agroecological
zones, gender, manure use, hiring of labor, and extension services were found to be statistically significant
in explaining adoption of improved varieties. Similarly, agroecological zone, gender, manure use, hiring
of labor, and extension services were important in explaining the amounts of basal fertilizers farmers used.

ISBN: 970-648-093-5
AGROVOC descriptors: Maize; Varieties; Innovation adoption; Technology transfer; Plant production;
Plant breeding; Soil fertility; Seeds; Fertilizers; manures; Farming systems;
Kenya
Additional keywords: CIMMYT
AGRIS category codes: El 4 Development Economics and Policies
F04 Fertilizing
Dewey decimal classification: 338.166762


Printed in Mexico.











Contents


iv Tables
iv Figures
v Acronyms and Abbreviation
vi Acknowledgements
vii Executive Summary

1 Introduction
3 The Study Area
5 Methodology
5 Sampling procedure
5 Model specification
8 Maize Production Technology Recommendations
8 Maize varieties
8 Fertilizer and manure recommendations
9 Demographic and Socioeconomic Characteristics
10 Maize in the Farming System and Adoption of Varieties
11 Farmers' Adoption and Disadoption of Improved Maize Seed
11 Maize varieties (1996-1998)
12 Popular maize varieties and seed package size
13 Rate of adoption of improved maize varieites
15 Soil Fertility Management
15 Fertilizer use and sources
16 Preferred fertilizer packages and use of organic manure
16 Rate of fertilizer adoption
17 Access to Rural Support Services
17 Credit use and access
17 Extension services and membership in an organization
18 Logistic Model Estimates
20 Conclusion
22 References










Tables


3 Table 1. Administrative divisions and population of Embu District, Kenya
8 Table 2. Recommended maize varieties, year of release, expected yields and maturity period,
RRC-mandate zones
9 Table 3. Demographic and socioeconomic characteristics of adopters and non-adopters of
improved maize varieties, Embu District, Kenya
9 Table 4. Demographic and socioeconomic characteristics of adopters and non-adopters of
improved maize, Embu District, Kenya
10 Table 5. Adopters and non-adopters ranking of crops grown according to subsistence and
commercial needs, Embu District, Kenya
11 Table 6. Most important maize varieties grown (1996-1998), Embu District, Kenya
12 Table 7. Maize varieties farmers value, Embu District, Kenya
12 Table 8. Attributes that farmers value in maize varieties, Embu District, Kenya
13 Table 9. Maize seed package preferred by adopters and non-adopters in Embu District, Kenya
14 Table 10. Change in adoption rates of improved maize varieties over time (coefficients of linear
and logistic regression)
15 Table 11. Fertilizer used by adopters and non-adopters, 1998, Embu District, Kenya
16 Table 12. Nitrogen and phosphorus application, adopters and non-adopters, 1998, Embu
District, Kenya
17 Table 13. Farmers' access to rural support services, Embu District, Kenya
18 Table 14. Logit analysis for adoption of improved maize variety, Embu District, Kenya
19 Table 15. Linear regression model for fertilizer use, Embu District, Kenya (dependent variable
is total fertilizer use in kg/ha and phosphorus in kg/ha)



Figures


3 Figure 1. Embu District Divisions, Kenya
13 Figure 2. Evolution of adoption of improved maize varieties, Embu District, Kenya, 1965 to
1999
16 Figure 3. Evolution of fertilizer adoption for adopters and non-adopters of improved maize,
Embu District, Kenya










Acronyms and Abbreviations



ASN Ammonium Sulphate Nitrate
CAN Calcium Ammonium Nitrate
CIMMYT Centro Internacional de Mejoramiento de Mafz y Trigo
(International Maize and Wheat Improvement Center)
DAP Di-Ammonium Phosphate
EU European Union
KIT Koninklijk Instituut voor de Tropen (Royal Tropical Institute)
KSh Kenyan shilling
masl Meters above sea level
MOA Ministry of Agriculture
NGO Non-governmental organization
OPV Open-pollinated variety
PHB Pioneer hybrid
RRC Regional Research Center
SPSS Statistical Package for the Social Sciences
TSP Triple Super Phosphate










Acknowledgements



We are grateful to the staff at the Regional Research Centre (RRC), Embu, who assisted in
data collection and report writing. Special thanks go to Sally Muhoro, Administrative
Assistant, National Agroforestry Research Project, Joseph Njagi for typesetting, and Fred
Manyara and Mugo Rimui for data collection. We are also thankful to the staff of the
Ministry of Agriculture in Nembure, Kieni and Runyenjes Divisions for their cooperation
and assistance during data collection.

The financial support provided by the International Maize and Wheat Improvement
Center/European Union Project on Strengthening Economics and Policy Research in
National Agricultural Research Systems in Eastern Africa and invaluable facilitation by
the Kenya Agricultural Research Institute is immensely appreciated.

We recognize with many thanks the input and support given by S.P. Gachanja, Centre
Director of the RRC, Embu. We also appreciate the editorial assistance provided by
Satwant Kaur and Eliot SAnchez Pineda for layout, production, and printing.

Finally we thank the farmers in Nembure, Runyenjes and Kieni Divisions for availing
their time for the survey.











Executive Summary



Maize is a major food crop in Kenya and dominates all food security considerations in the country. It accounts
for more than 20% of all agricultural production and 25% of agricultural employment. Smallholder farmers
contribute more than 70% of the country's maize. Commercial farmers also contribute a significant portion of
commercial maize. Because of the importance of maize to food security, the Kenyan Government has supported
a maize breeding program that has released more than 20 modern maize varieties since 1955. The adoption of
these improved maize varieties and use of inorganic fertilizer by large-scale farmers in high potential areas
were major factors for increased yield growth between 1963 and 1974. While adoption rates by small-scale
farmers in high potential areas equaled those of large-scale farmers between 1975-1983, yield gains were
smaller many small-scale farmers adopted new varieties but not fertilizer technologies. Between 1985 and
1991, improved seed was adopted in low potential areas, but fertilizer use remained low.

This study was undertaken to identify socioeconomic and technical factors affecting the adoption of improved
maize seed and fertilizer use, and also to determine the role of credit in both. The information will assist maize
research and extension specialists to develop suitable technologies for farmers in different environments and
from different socioeconomic backgrounds. The analysis will also enable policy makers to identify policy and
institutional factors that can contribute to increased adoption of high-yielding maize technologies.

Three maize growing divisions in the Embu District, Kenya-Nembure, Runyenjes, and Kieni-were
purposively selected for the study. Embu District was selected because it is representative of maize growing
areas in the region. The survey was conducted in Upper Midland 2 and 3 zones (Jaetzold and Schmidt 1983),
where land use is dominated by coffee/dairy land use systems.

The study covered both the long rains lasting from March through June and short rains from October to
December. One hundred and twenty-seven farmers were randomly selected and interviewed. Using a
structured questionnaire, data was collected on farmer and farm attributes and institutional structure. Adopters
of certified maize were defined as farmers who planted certified maize for two consecutive seasons in 1998 and
non-adopters were defined as farmers who planted local recycled hybrid seeds or recycled seeds of open-
pollinated varieties for more than three seasons. A comparative analysis was done between adopters and non-
adopters of improved maize seed. Since productivity gains in maize seed depend on simultaneous use of other
inputs, particularly fertilizer, the level of fertilizer used by adopters and non-adopters was also determined.
Logit and linear models were used to analyze factors affecting adoption of improved maize seed and quantity
of fertilizer, respectively.

Nearly all farmers (about 98% of adopters and non-adopters in the long rains and 96% and 88% of adopters and
non-adopters in the short rains, respectively) used basal fertilizer irrespective of whether they planted certified
maize seed. The use of top dress fertilizer in the long rains was more common among adopters (44%) than non-
adopters (17%). Compound fertilizer (23:23:0) was the main basal fertilizer adopters (52% and 55% of responses
in the long and short rains, respectively) and non-adopters used (56% and 64% in the long and short rains,
respectively). Calcium ammonium nitrate was the main top dress fertilizer adopters (100% and 96% in the long
rains and short rains, respectively) and non-adopters used (88% in both seasons). Most maize farmers applied
much less than the optimal level of fertilizer (50 kg/ha N). Adopters applied 35 kg/ha N in the long and short
rains, while non-adopters applied 31- and 32 kg/ha N in the long and short rains, respectively.











More adopters (83%) than non-adopters (69%) hired labor for farm operations. Significantly more adopters
(78%) than non-adopters (64%) obtained formal credit from co-operative societies in the form of seed and
fertilizer (p<0.1). About 66% and 49% of adopters used credit to purchase fertilizer for coffee or maize,
respectively, while 56% and 44% of non-adopters did so. The average amount of credit was Ksh 6,802
(US$1=Ksh 60,263 (1998)) for adopters and Ksh 3,613 for non-adopters. To get credit from the cooperative
society, farmers had to deliver the coffee crop. The main reason cited by adopters for not using credit was
availability of capital (29%). Twenty-one percent of adopters said that it did not pay to use credit in maize
production. For non-adopters, the main constraint was lack of collateral (71.4%) (in the form of coffee delivered
to coffee co-operative societies).

More adopters (90%) than non-adopters (76%) had access to extension. The main source of extension was the
Ministry of Agriculture (83% of 64 responses by adopters and 74% of 28 responses by non-adopters). More
adopters (81%) than non-adopters (69%) received advice on improved maize production. However, this
difference was not significant. The main type of extension advice was on fertilizer use (52% of 82 responses for
adopters and 50% of 44 responses for non-adopters).

Agroecological zones, gender, manure use, hiring of labor, and extension were statistically significant in
explaining adoption of improved maize variety. Other variables-age and education of household head, farm
size, credit, years of formal education, area under coffee, and farmer group membership- which were
expected to influence adoption and fertilizer use were not significant (at 10% or lower probability level) in
explaining adoption decisions. Likewise, agroecological zone, gender, manure use, hiring of labor, and
extension were important variables in explaining the amounts of basal fertilizers farmers applied.

Most adopters of certified seed preferred Pioneer H3253 in 1998. While there was a marked increase in the use
of Pioneer H3253 between 1996 and 1998, the proportion of farmers using older hybrid varieties, H511 and
H512, decreased in the same period. Adopters preferred the Pioneer hybrid for its high yield and early
maturity, despite its high price, poor storage, and root lodging. Other valued traits were pest and drought
tolerance, large grains, taste, good threshing quality, and ease of cooking. Most adopters preferred the smaller
2-kg seed package because it was affordable and sufficient for their plots of maize.

For most farmers the high price of improved maize seed was the major constraint for adoption (96% of
adopters and 88% of non-adopters). Other constraints mentioned were the low selling price of maize (20% of
adopters and 18% of non-adopters) and lack of credit (12% of adopters and 18% of non-adopters).

The study shows that many attributes are taken into consideration in variety selection. In addition, farmers
also mentioned that they used recycled seed because of the high price of improved maize and because they
find that there is little difference between yields in improved and recycled seed. A greater focus on farmer
participatory breeding will help incorporate farmers' assessments of maize varieties in the research process.
Further studies on the economics of seed recycling will help in greater understanding of farmers' use of
recycled seed. The packaging of maize seed in small and more affordable packages such as the 2-kg bags will
also help increase adoption of certified maize.









Adoption of Maize Seed and


Fertilizer Technologies in Embu


District, Kenya
James O. Ouma, Festus M. Murithi, Hugo Verkuijl, Wilfred Mwangi,
Macharia Gethi, and Hugo De Groote



Introduction


Agriculture is the main economic sector in Kenya and contributes significantly to national
development. It is a sector that receives high priority and attention from the Kenyan Government.
In its Sessional Policy Paper No. 1 on Economic Management for Renewed Growth, the Kenyan
Government stated that self-sufficiency in the production of agricultural products such as maize,
beans, and vegetables is a key strategy for sound economic management and renewed growth for
the country. For the agricultural sector to play this central role in a sustainable way, rapid growth
in output and productivity is critical. It is widely recognized that the sustained flow and use of
improved agricultural technologies is key to increased growth and agricultural productivity.

Maize is a major food crop and dominates all national food security considerations in Kenya. The
area under maize is estimated to be approximately 1.5 million hectares, and per capital
consumption averages 103 kg/yr (Pingali 2001). Maize accounts for more than 20% of all
agricultural production and 25% of agricultural employment. Smallholders produce about 70% of
the nation's maize, although large-scale farmers also contribute a significant proportion of
commercial maize production (GOK 1983). Because of the importance of maize to national food
security, the government of Kenya has supported maize research since 1955, when a maize research
program was started to increase maize production.

More than 20 modern maize hybrids and varieties have been released in the four decades since the
program began. Large-scale farmers in high potential areas adopted the new hybrids (Gerhart
1975). About half of these farmers adopted improved seed and inorganic fertilizer use, which were
major factors in the growth of maize yields between 1963 and 1974. Between 1975 and 1984, many
smallholder farmers in high potential areas also adopted improved seed and eventually equaled
adoption rates of large-scale farmers (Hassan and Karanja 1997). However, yield gains were
smaller during this period for a number of reasons: many smallholders adopted improved seed but
not fertilizer use, an unfavorable policy environment, and severe drought in 1979-80 and 1983-84.
Between 1985 and 1991, improved seed was adopted in low potential areas, but fertilizer use
remained low.










The average yield for maize in Kenya is about 1.5 t/ha (Pingali 2001). Most smallholders produced
under 1 t/ha, well below the potential average of 4.7 t/ha (Hassan et al. 1998). In Embu District, the
study area for this report, maize yield is estimated at 0.5 t/ha (Ministry of Agriculture 1998). Other
studies (Murithi et al. 1994) estimated yields at 1.3 t/ha and 0.9 t/ha in the major (March to June)
and minor (October to December) seasons, respectively. Studies (Matiri et al. 1996) conducted in the
neighboring Meru District, concluded that many farmers adopted the recommended agricultural
practices (time of planting, time and number of weeding, type of planting arrangement and
spacing). Even though there were indications from the study that many farmers were also using
hybrid seeds recommended by the national maize research program, the issue remained contentious
because farmers regarded recycled seed from the previous harvest as improved seed as well.
Fertilizer use was also far below recommended levels.

This study identifies constraints and potentials for higher yield gains in maize in the Embu District.
More specifically, it examines maize farmers' circumstances and farming practices, identifies major
socioeconomic and technical factors that influence adoption of improved maize seed and fertilizer
technologies, and determines the role of credit in the study area. In particular, it analyses
determinants of recent patterns of maize technology adoption and sources of maize productivity
gains. This information will assist maize research and extension to develop suitable technologies
and target them appropriately to farmers in different environments and from different
socioeconomic domains. The analysis will also enable policy makers to identify policy and
institutional factors that can contribute to increased adoption of high yielding maize technologies.











The Study Area


Embu District has a diversity of agroecological conditions ranging from high altitude dairy or
temperate vegetables zone (UM1) to very dry lowland livestock-millet zone (L5). Ten major
agroecological zones covering 81% of agricultural land have been identified in Embu. The District
has five major soil types, nitosols, andosols, vertisols,ferrosols, and cambisols. The soils and agroecology
of the area are greatly influenced by Mount Kenya and Nyandarua ranges.


Embu District covers an area of 819
km2 and has a population of more
than 278,000 (Table 1). Excluding 210
km2 of Mount Kenya, this translates
into a high population density of 456
people per km2. The district is divided
into five administrative divisions:
Central, Kyeni, Nembure, Runyenjes,
and Manyatta (Figure 1). These five
divisions are sub-divided into 15
locations and 52 sub-locations with
about 45,000 farm families.


Table 1. Administrative division and population, Embu District, Kenya.
Population
Total Households Area density
Division population (no.) (km) (persons/km2)
Central 52,466 14,726 70.6 743
Kyeni 48,385 10,441 104.9 461
Manyatta 71,332 15,523 197.1 666
Nembure 41,590 8,976 88.1 472
Runyenjes 64,111 13,981 149.0 432
Mt Kenya Forest 332 246 210.2 2
Total 278.216 63,893 819.0 456t


Source: Central Bureau of Statistics (2001)
Note: tExcluding Mount Kenya Forest


Kyeni, Nembure, and Runyenjes are the three major
maize producing Divisions with a mixed farming
system. Maize, beans, mangoes, pawpaw, pigeon
peas, cowpeas, coffee, cotton, avocado, bananas,
greengrams, and sunflower are the main crops.
Maize and beans are grown as either intercrops or
monocrops. The average farm size ranges between
2-2.8 ha per household.


Kyeni Division was carved out of Runyenjes's \i .
Division in 1996. It borders the Meru South District ,
to the east, Runyenyes Division to the west, Mbeere .,,.
District to the south and Mount Kenya forest to the Embu Town .
Embu Town'f, -- -
north. It lies between 1,000-1,700 masl and covers Embu
104.9 km2, of which 78.62 km2 is arable land. Kyeni r Embu D
has three administrative locations Kyeni North,
Kyeni, and Karurumo. Each administrative location Figure 1. Embu District Divisions, Kenya.
has 6-10 sub-locations. The average family size is six
and the average farm size is 0.8-2.4 ha in the upper
and middle zones, respectively. The rainfall is bimodal: the October-November short
rains provide between 1,200 to 1,850 mm and April to May long rains between 850 to
1,850 mm. Soils are well drained and mostly nitrosols.










The Runyenje Division lies between 1,200-2,070 masl. It covers 149 km2, out of which 96.26 km2 is
arable land. The Division has three administrative locations and sub-locations. The estimated
population is 64,111 and average farmholding is 0.4-0.8 ha. The annual average rainfall is between
1,000-2,000 mm and distribution is bimodal. The long rains fall in March to April and October to
December.

Nembure lies between 1,000-1,500 masl. It covers an area of 88 km2, of which 65 km2 is arable land.
Nembure has three administrative locations: Gaturi South, Kithimu, and Makengi. The estimated
population is 41,590 and population density is 472 persons per km2. The average annual rainfall
ranges from 1,200 to 1,500 mm. Rainfall is bimodal and distributed in March/April (long rains) and
October/November (short rains). Soils are fertile and well drained.










Methodology



Sampling Procedure
Embu District was chosen because it is representative of other Districts in the region where maize is
a major crop. The District's proximity to the Regional Research Center in Embu also ensured quality
data and made data collection easy. Three of the five Divisions mentioned above Nembure,
Runyenjes, and Kyeni were purposively selected. The survey was conducted in the main maize
growing areas of the three Divisions, classified as Upper Midland 2 (UM2) and Upper Midland 3
(UM3) (Jaetzold and Schmidt 1983), that together are dominated by the coffee/dairy land use
systems.

Villages were randomly selected from four sub-locations selected from six maize growing areas (two
from each Division). One hundred and twenty-seven farmers were interviewed 82 adopters and 45
non-adopters. Data was collected on farmer and farm characteristics and institutional factors using a
structured questionnaire. Adopters of certified maize seed were defined as farmers who planted
certified maize seed for two consecutive seasons in 1998, and non-adopters were defined as farmers
who planted local seed, recycled seed of hybrids or, in the case of open-pollinated varieties (OPVs),
recycled seed for more than three seasons.




Model Specification
In adoption studies, responses to a question such as whether a technology was adopted could be yes
or no, a typical case of a qualitative dichotomous variable. However, factors (independent variables)
that affect a given maize technology adoption can be expressed both qualitatively and
quantitatively. When the dependent variable is continuous, linear models such as ordinary least
squares (OLS) are used to show the effects of independent variables on the dependent variable.
However, when the dependent variable is dichotomous, the use of linear probability models is a
major problem: the predicted value can fall outside the relevant range of zero-to-one probability
values. To overcome this problem, logit or probit models have been recommended (Gujarati 1988).

Logit and probit models have been widely used in different adoption studies (for example, Yahanse
et al. 1990; Polson and Spencer 1991; D'Souza, Cyphers, and Phipps 1993; Hussain, Byerlee, and
Heisey 1998; Salalsya et al. 1996; Chilot, Shapiro, and Demeke 1996). These models not only help
assess various factors that affect adoption of a given technology, but also provide predicted
probabilities of adoption. For example, they can be used to indicate how the likelihood of a farmer
adopting a particular technology changes according to his or her level of education, keeping all
other factors constant.

Generally, adoption studies consider many factors to explain farmers' adoption decisions. In Embu
District, it was hypothesized that a farmer's decision to use or not use a given maize technology is
influenced by the characteristics of the household head (gender, age, and formal education), farm
size, manure use, use of credit, extension advice, labor, cash crop, and farmer group membership.










Detailed discussion of how some of these factors might influence technology adoption is found in
CIMMYT (1993). The empirical model for the two technologies is specified as follows:

TECH = B0 + B1X1 + B2X2 + BX +.... BloXo1 + U

Where: TECH = adoption of improved maize varieties over the last two seasons, or quantity of
fertilizer applied

X, = gender of household head
X, = farm uses organic manure (dummy variable)
X3 = agroecological zone (UM2=1, UM3=0)
X4 = farm uses hired labor (dummy variable)
X5 = farmer received extension services (dummy variable)
X, = farmer uses credit (dummy variable)
X7 = formal education of household head (years of schooling)
X8 = age of household head (yr)
X9 = farm size (acres)t
Xlo = area under coffee (main cash crop) in acres
X,, = membership in farmer organization (dummy variable)
U = disturbance term; B0is the intercept and Bis are the coefficients of the independent
variables



The working hypotheses for this report are:

* Use of organic manure: It is hypothesized that farmers who use cattle manure are more aware of
the effect of fertilizers on crops and therefore better adopters of fertilizer technologies. This is a
dummy variable (0=farmer did not use manure, 1=farmer used manure).
* Hired labor: It is hypothesized that there is a positive relation between adoption of improved
maize technologies and hiring labor. This is a dummy variable (0=no use of hired labor, 1=use
of hired labor).
* Contact with extension agent: Agricultural extension services are a major source of information
in the study area and contact with extension agents increases a farmer's likelihood of adopting
improved maize technologies. This is a dummy variable (0=farmer did not have extension
contact, 1=farmer had extension contact).
* Credit: Access to credit increases the probability of adopting improved maize technologies. This
is a dummy variable (0=no use of credit, 1=use of credit).
* Educated household head: A higher level of education increases a farmer's ability to obtain,
process, and use adoption information of an improved maize variety or fertilizer. Education
thus increases the probability of adopting improved maize technology.
* Farmer's age: It is hypothesized that with increasing age a farmer will be less likely to be aware
of new maize varieties or fertilizer use. Younger farmers may have greater access to information
because they have greater access to education, and thus will be more aware of technologies.
Older farmers might not have access to this information.

S 1 acre = 0.405 ha










* Farm size: Farm size (acres) is an indicator of wealth (and perhaps a proxy for social status and
influence within a community) and expected to be positively associated with the decision to
adopt improved maize technologies.
SArea planted to coffee: Farmers with larger areas planted to coffee are better adopters of
improved maize technologies. Coffee provides farmers with cash to buy inputs.
* Membership in an organization: Members of organizations (farmer groups, non-governmental
organizations (NGOs)) have better access to information on improved maize technologies.
Being a member of an organization is hypothesized to be positively associated with adoption of
improved maize technologies. This is a dummy variable (O=farmer is not a member of an
organization, 1=farmer is a member).










Maize Production Technology Recommendations



Maize Varieties
Table 2 shows different maize varieties that are recommended in the study area and their
characteristics. As a significant amount of maize is planted in the coffee/dairy-based land use zone,
it was selected as the study site. Even though maize is also planted in small amounts in the tea/
dairy (lower highland 1 (LH1) UM1) and maize/sunflower zones (upper midland 4 (UM4)), lower
midland 3 (LM3) (UM4) and lower midland 4 (LM4), they were not included in the study.


Table 2. Recommended maize varieties, year of release, expected yields, and maturity period, RRC-Mandate Zone.
Maturity Yield potential
Land use system Altitude Variety Year released (days) (t/ha)
Tea Dairy Zone 1,500-2,100 masl H627 1996 180-240 3.6
(LH1-UM1) H626 1989 180-240 3.4
H625 1981 180-240 2.8
H614D 1986 165-210 2.7
Coffee Dairy Zone 1,000-1,800 masl H513 1996 120-150 1.8
(UM2-UM3) C5222 1996 120-150 1.8
PAN 5195 1996 120-150 1.8
PHB 3253 1996 120-150 1.8
CG 4141 1996 105-130 1.4
H512 1970 120-150 1.8
H511 1968 120-140 1.5
EMCO92SR 105-130 1.5
Maize Sunflower Zone <1,800 masl DH1 drylandd hybrid) 1996 90-120 1.2
(UM4/LM3/LM4) DH2 1996 90-120 1.2
DLC1(Makueni) 1989 90-120 1.1
KCB (Katumani) 1968 90-120 1.1
CG4141 1996 110-120 1.2
Source: Recommendation Guidelines for Crops, Soil Fertility and Livestock Management (1998), RRC-Embu


Fertilizer and Manure Recommendations
The RRC-Embu guidelines recommend the use of compound fertilizers 20:20:20 or 23:23:0 at a rate
of 50 kg N and 50 kg P205 per ha at sowing in the coffee/dairy zones, top dress fertilizer applied at a
maize height of 45 cm and at a rate of 50 kg N/ha, and farmyard manure at a rate of 5 t/ha at
planting or before planting.











Demographic and Socioeconomic Characteristics



Tables 3 and 4 show demographic and socioeconomic characteristics of sampled households for
adopters and non-adopters. About 4% of adopters and 11% of non-adopters were female-headed
households. At least 58% of all farmers had primary education and above. Forty-six percent of
farmers in UM2 were adopters and 20% were non-adopters. In UM3, 54% were adopters and
80% were non-adopters (Table 4). Farm households in UM2 are probably better resource
endowed than those in UM3. More adopters (82.9%) hired labor compared to non-adopters
(68.9%) (p<0.1).



Table 3. Demographic and socioeconomic characteristics of adopters and non-adopters of improved maize varieties,
Embu District, Kenya.
Adopters Non-adopters All farmers
Characteristic Mean Standard deviation Mean Standard deviation t-statistic Mean
Age of household head (yrs.) 50.0 12.4 49.3 12.00 0.27 (NS) 49.8
Farm size (acres)t 3.7 2.3 3.8 3.6 0.18 (NS) 3.7
Area under coffee (acres)t 1.0 0.8 0.8 0.8 1.08 (NS) 0.9
Area under maize (acres)t 1.1 0.7 0.9 0.6 1.62 (NS) 1.0
Household size (no.) 8.2 3.3 7.0 2.8 2.10 (NS) 7.8
Note: NS= not significant. t 1 acre = 0.405 ha.


Table 4. Demographic and socioeconomic characteristics of adopters and non-adopters of improved
maize, Embu District, Kenya.
Characteristic Adopters (%) Non-adopters (%) X2 All (%)
Gender 2.7*
Male 96.3 88.9 93.7
Female 3.7 11.1 6.3
Education 0.6 (NS)
None 9 13.2 10.5
Primary 57.7 57.9 57.8
Secondary 26 28.9 27.0
Division 1.4 (NS)
Kyeni 48.8 44.4 47.2
Runyenjes 28.0 37.8 31.5
Nembure 23.2 17.8 21.3
Zone 8.6***
UM2 46.3 20 37.0
UM3 53.7 80 63.0
Use hired of labor 3.3***
No 17.1 31.1 22.1
Yes 82.9 68.9 77.9
Note: NS= not significant; = significant at p<0.1; ** = significant at p<0.05; *** = significant at p<0.01











Maize in the Farming System and Adoption of Varieties



As mentioned earlier, maize is a major food staple in the study area. A household without
maize is termed food insecure. All sampled farmers grew maize. Other major food crops, in
order of importance are beans, coffee, bananas, and Irish potatoes (see table in Annex).

Table 5 shows the main crops grown in the study area and their ranking by adopters and non-
adopters according to their ability to meet subsistence or commercial needs. Overall, maize was
the most important food crop for both groups of farmers, followed by beans, bananas, and Irish
potatoes. Coffee was the main cash crop.




Table 5. Adopters and non-adopters ranking of crops grown according to subsistence and commercial needs, Embu
District, Kenya.


Adopters ranking of crops (%)
Rank 2 Rank 3 Rank 4
61.0 11.0 1.2
26.8 57.3 4.9
4.9 14.6 3.7
3.7 9.8 42.7
1.2 1.2
1.2 -
3.7 13.4

2.4
1.2
3.7
2.4


Non-adopters ranking of crops (%)
Rank 2 Rank 3 Rank 4
51.1 11.1 6.7
35.6 55.6 2.2
11.1 17.8 2.2
2.2 2.2 31.1

2.2
8.9 20.0
4.4

2.2


Overall percentage
Rank 5 of first 5 crops
98.8
93.9
91.5
11.1 68.3
2.4
1.2 2.4
8.6 25.6

4.9 7.3
1.2
3.7
1.2 3.7
1.2 3.7


Rank 5
2.2


11.1


4.4

2.2

2.2

2.2


Overall percentage
of first 5 crops
100
97.8
95.6
46.7

2.2
33.3
4.4
2.2
2.2
2.2

2.2


Rank 1
25.6
4.9
68.3
1.2


Rank 1
33.3
4.4
62.2


Crops
Maize
Beans
Coffee
Banana
Mango
Avocado
Irish potato
Kale
Sweet potato
Arrow root
Cabbage
Macadamia
Tomato


Crops
Maize
Beans
Coffee
Banana
Mango
Avocado
Irish potato
Kale
Sweet potato
Arrow root
Cabbage
Macadamia
Tomato










Farmers Adoption and Disadoption of

Improved Maize Seed



Maize Varieties (1996-1998)
Farmers were asked what was the most important maize variety they grew. Table 6 shows that
Pioneer H3253 was the most important maize variety during the survey year. It was grown by
34% of farmers in both the short and long rains of 1998. While the use of Pioneer H3253
increased sharply between 1996 and 1998, the proportion of farmers using old hybrid varieties
H511 and H512 decreased. This disadoption was attributed to low yields associated with old
varieties, the availability of other promising new hybrids, and unavailability of seed. A
considerable number of farmers also used recycled maize seed between 1996 and 1998 and
cited high price of certified seed, coupled with lack of credit and low price of maize as the main
limitation to using improved maize seed on a continuous basis.

Farmers indicated that the high price of improved maize seed was the major constraint for
adoption (86% of adopters and 79% of non-adopters). Other important constraints were the low
selling price of maize (20% and 18% of adopters and non-adopters, respectively) and lack of
credit (12% of adopters and 18% of non-adopters).


Table 6. Most important maize varieties grown (1996 -1998), Embu District, Kenya.
Farmers (%)
Variety LR 1996 SR 1996 LR 1997 SR 1997 LR 1998 SR 1998
H513 0 0 0 1.6 0 4.1
H511 34.1 38.9 17.5 17.9 13.0 14.6
H512 14.3 12.7 9.5 7.3 7.3 4.9
C5222 0 0 0 0 0 0.8
Pan 5195 0.8 1.6 0 0.8 0 0.8
CG 4141 0 0 0 0 0 1.6
PHB 3253 0.8 2.4 25.4 33.3 34.1 34.1
H614 3.2 1.6 2.4 1.6 2.4 0
H611 0 0 0 0 0 0
H626 0.8 0 0 0 0 0
H625 5.6 1.6 3.2 0.8 3.3 0
Katumani 0.8 1.6 0.8 0 0.8 1.6
Makueni 0 0.8 0 1.6 1.6 1.6
Recycled seed 40.5 41.3 43.7 39.0 43.1 44.7
Note: LR=long rains; SR=short rains.










Popular maize varieties and seed package size
Farmers were asked which varieties they favored (a yes or no answer, but each farmer could opt
for more than one variety). Pioneer 3253 was the most popular maize variety (Table 7) followed
by H511, Makueni, and H512. Pioneer 3253 was popular because of its high yield and early
maturity despite its high price, poor storage, and root lodging (Table 8). Other valued traits were
pest and drought tolerance, large grains, good threshing quality, taste, and ease of cooking. Both
H511 and H512 were valued for their high yields and early maturity. H512 was also valued for its
good taste.




Table 7. Maize varieties farmers value, Embu District, Kenya.
Variety Adopters (%) Non-adopters (%) %2 All
PHB 3253 76.5 32.4 21.1*** 60.9
H511 69.1 43.2 7.2*** 59.9
H512 28.4 18.9 1.2 (NS) 25.0
H614 4.9 2.7 1 4.1
H625 7.4 1 4.8
Pan 5195 3.7 1 2.4
CG4141 1.2 1 0.8
Katumani 6.2 1 4.0
Makueni 32.1 56.8 6.4** 40.9
Note: NS = not significant; ** = p<0.05;*** = p<0.01; 1 = p not calculated.


Table 8. Attributes that farmers value in maize varieties, Embu District, Kenya.
Phb 3253 (%) H511 (%) H512 (%)
Attributes Adopters Non-adopters Adopters Non-adopters Adopters Non-adopters
High yield 93.2 90.0 52.7 73.3 81.0 100.0
Easy to cook 3.4 5.5 13.3 14.3 -
Early maturing 44.1 90.0 63.6 73.3 38.1 28.6
Pest tolerance 10.2 14.5 6.7 14.3 -
Stores longer 5.1 14.5 13.3 9.5 -
Good taste 6.8 20.0 18.2 19.0 42.9
Low price of seed 1.7 1.8 -
Large grains 10.2 5.5 9.5 14.3
Drought tolerance 8.5 3.6 -
Good anchoring 1.7 -
Threshing quality 1.8 -



Maize seed is available in 2-, 10-, and 25-kg bags. Most adopters (88%) preferred the 2-kg package,
which is consistent with the small farm sizes reported in the study area (Table 9). Forty-nine
percent preferred the 2-kg package because it was sufficient for the area allocated to maize, while
32% preferred the 10-kg package because it was affordable. Although the Kenya Seed Company
(KSC) tried to accommodate smallholders by making seed available in 2-kg packages during the
1980s, this trend was reversed in the early 1990s in favor of 10-kg packages. The 10-kg packages
accounted for more than 80% of all packages sold by KSC (Hassan et al. 1998). However, the










present study revealed discrepancies between farmers' preferences and seed package sizes
available at KSC-only 7% of adopters favored the 10-kg package, yet the most common seed
package sold by KSC was 10 kg.

The price of a 2-kg improved maize variety package rose from 144 Kenya Shilling (Ksh)1 in 1996,
189 Ksh in 1997 to 240 Ksh in 1998, a 67% increase in two years.


Table 9. Maize seed package preferred by adopters and non-adopters
in Embu District, Kenya.
Adopters (%) Non-adopters (%) All farmers
Package size
Do not purchase 3.7 65.1 25.5
2 kg 87.8 30.2 67.4
10 kg 7.3 4.7 6.4
25 kg 1.2 0.8


Rate of Adoption of Improved
Maize Varieties
Figure 2 shows the evolution of adoption of
improved maize varieties in Embu District. The
rates were derived by asking farmers the year
they first planted improved maize seed, and
then running cumulative frequencies to
determine the proportion of farmers who
adopted improved maize seed. About 75%
adopted improved maize by 1997.


Cumulative percentage
90 -----------


The evolution of adoption of new technologies .... ... .. .. .... ....
1965 70 75 80 85 90 95 2000
can be quantified by estimating the trend using
Figure 2. Evolution of adoption of improved maize varieties
time as a dependent variable in a regression on in Embu District, Kenya, 1965 to 1999.
the adoption rate. Such a regression captures the
historical trend of adoption over a given time
and can be used to assess the effectiveness of agricultural institutions that have served the
farming system over time. A linear regression is the most convenient, even though it has the
disadvantage of having the dependent variable (adoption rate) fall partly outside the possible
range (between 0 and 100%). Mathematically, the linear trend is given by the formula:


Y, = a+bt





1 US$ 1 = Ksh 60.263 (1998).










where:
Y, = the cumulative percentage of adopters at time t
t = time (0=1965, 1=1966 ...... 34=1999)
a = the intercept
b = the slope, indicating the change of adoption rate over time

An alternative measure is the use of a logistic curve, which has an S-shape and keeps the
dependent variable between a minimum and maximum limit. The basic assumption is that
adoption increases gradually and then accelerates to a maximum level (CIMMYT 1993). The
functional form of the logistic curve is given by:

Y, = Kl(l+e-"-bt)

Where:
Y, = the cumulative percentage of adopters at time t;
t = time (0=1965, 1=1966, ...... 34=1999)
K = the upper boundary of adoption
a = a constant, related to the time when adoption begins
b = a coefficient related to the rate of adoption

Both functional forms were estimated (Table 10), but the linear model provides a better fit (R2=0.98)
than the logistic model (R2=0.93). The coefficient b of the linear regression indicates that from 1965
to 1999, the adoption rate of improved maize varieties increased by 2.25% per year.


Table 10. Change in adoption rates of improved maize varieties
over time (coefficients of linear and logistic regression).
Linear Logistic
a -5.058 *** 0.346 ***
(1.080) (0.048)
b 2.253 *** 0.864 ***
(0.055) (0.006)
RSquare 0.98 0.93
F 1701 *** 432 ***
Standard Error 3.26 0.42
N 34 34
Note: Standard errors of the coefficients are in brackets; *** = significant at the 1% level.











Soil Fertility Management



Fertilizer Use and Sources
Most surveyed farmers used basal fertilizer (98% of both types of farmers in the long rains, and
96% and 88% of adopters and non-adopters in the short rains, respectively) (Table 11). The use of
top dress fertilizer in the long rains was more common among adopters of improved maize (44%)
than non-adopters (17%) (2=9.4; p<0.05). Compound fertilizer (23:23:23) was the main basal
fertilizer used by adopters (52% and 55% in the long and short rains, respectively) and non-
adopters (57.5% and 64% in the long and short rains, respectively). Calcium ammonium nitrate
was the main top dress fertilizer used by both categories of farmers.

During the long rains, basal fertilizer was applied at an average rate of 2.9 and 2.2 bags/ha (50 kg
bags/ha) by adopters and non-adopters, respectively. The average amount of basal fertilizer
applied by both groups was 2.9 bags/ha. The difference of the means was significant (p<0.05). In
the short rains, adopters applied 2.8 bags/ha, while non-adopters applied 2 bags/ha (Table 12).
The difference was again significant (p<0.05). These rates were lower than the recommended rates.



Table 11. Fertilizer used by adopters and non-adopters, 1998, Embu District, Kenya.
Fertilizer Adopters (%) Non-adopters (%) All
Basal fertilizer (LR) 97.5 97.6 97.5
Basal fertilizer (SR) 96.3 88.1 93.4
Top dress fertilizer (LR) 44.4 16.7 34.6
Top dress fertilizer (SR) 34.6 19.0 29.1
Fertilizer type (Top dress: LR)
Calcium ammonium nitrate 100.0 87.5 95.6
Ammonium sulphur nitrate 12.5 4.4
Fertilizer type (Top dress: SR)
Calcium ammonium nitrate 96.1 87.5 93.1
Ammonium sulphate nitrate 3.8 12.5 6.9
Fertilizer type (Basal: LR)
Di-ammonium phosphate 27.7 17.5 24.1
20/20/0 19.3 25.0 21.3
23/23/23 51.8 57.5 53.8
Triple super phosphate 1.2 -0.8
Fertilizer type (Basal: SR)
Di-ammonium phosphate 22.9 17.9 21.1
20/20/0 19.3 17.9 18.8
23/23/23 55.4 64.1 58.5
Triple super phosphate 2.4 -1.5
Note: LR = long rains; SR = short rains










Table 12. Nitrogen and phosphorus application, adopters and non-adopters, 1998, Embu District, Kenya.
Amount (kg/ha) Adopters Non-adopters All farmers
Nitrogen (LR) 35.0 31.4 33.7
Nitrogen (SR) 35.2 32.8 34.3
Phosphorus (LR)** 46.8 34.5 42.4
Phosphorus (SR) 49.1 36.1 44.5
Basal fertilizer (50-kg bags/ha)
LR 2.9 2.2 2.7
SR 2.8 2.0 2.5
Top dress (50 kg bags/ha):
LR 2.3 2.7 2.4
SR 2.0 3.0 2.4
Note: ** = significant at p<0.05; LR = long rains; SR = short rains



The main sources of fertilizer were coffee cooperative societies (72% of 59 responses by adopters and
69% of 31 responses by non-adopters), local seed dealers (50% of adopters and 51% of non-
adopters). Most farmers used fertilizer. Those who did not cited high price and unavailability.




Preferred Fertilizer Packages and Use of Organic Manure
The 50-kg fertilizer package was most preferred among adopters (76% of 62 responses) and non-
adopters (72% of 31 responses). Fertilizer was used for maize and other crops, in particular, coffee.
The 25-kg package was the second most preferred package among adopters (27%) and the 10- and
25-kg packages among non-adopters (16% and 19%, respectively). Most farmers (95% and 96% of
adopters and non-adopters, respectively) reported that they had easy access to the packages they
preferred. About 96% of adopters and 81% of non-adopters reported that they applied manure in
their maize fields. This difference was significant at p<0.01. Cattle manure was the main organic
fertilizer (98% of 77 responses for adopters and 97% of 36 responses for non-adopters).


Rate of Fertilizer Adoption
Figure 3 indicates that non-adopters of
improved maize were a few years behind
adopters in fertilizer use. In 1998, at least 90% of
adopters and non-adopters used fertilizer.


Cumulative percentage


8Adopters of improved maize
80
60--- ----- ------ ---- ---------

60-------------------------- -----------------------


40----------------- ---------------------------------


n------------ --- --------------------------------


Non-adopters of improved maize

A-- -- ---


1965 70 75 80 85 90 95
Figure 3. Evolution of fertilizer adoption for adopters and
non-adopters of improved maize, Embu District,Kenya.


I..


I










Access to Rural Support Services



Credit Use and Access
More adopters (78%) than non-adopters obtained credit from cooperative societies in the form of
seed and fertilizer (p<0.1) (Table 13). About 66% of adopters used credit to purchase fertilizer for
coffee and 49% for maize, while 56% of non-adopters used credit to buy fertilizer for coffee and 44%
for maize. About 28% of both groups used credit for school fees. Credit was also used to purchase
maize seed, animal feed, and to construct small maize mills. The average amount of credit was Ksh
6,802 for adopters and Ksh 3,613 for non-adopters. The coffee crop was used to obtain credit from
the cooperative society. More than 28% of adopters cited availability of their own working capital
and lack of interest as the reason for non-use of credit. Adopters (21.4 %) also thought it was
uneconomical to use credit for maize production. For non-adopters, the main constraint was lack of
collateral (71.4%).

Extension Services and Membership in an Organization
Adopters of improved maize (90%) had better access to extension services than non-adopters (76%)
(p<0.05). The main source of extension services was the Ministry of Agriculture (83% of 64 responses
by adopters and 74% of 28 responses by non-adopters), NGOs (12% of adopters and 21% of non-
adopters), and other farmers (18% of adopters and 16% of non-adopters). There was no significant
difference between the numbers of adopters and non-adopters receiving advice on improved maize
production, that is, 75%. The main extension advice was on fertilizer use (52% of 82 responses for
adopters and 50% of 44 responses for non-adopters). Other important extension services were on
seed (44%) and spacing .133. ) for adopters, and for non-adopters (27%) on spacing. Thirty-five
percent of adopters and 55.6% of non-adopters were visited twice in the last two years. More than
90% of both groups were members of a cooperative or farmers' group. The main service provided
by these groups was credit to purchase inputs (68% and 62% of adopters and non-adopters,
respectively).


Table 13. Farmers' access to rural support services, Embu District, Kenya.
Rural support services Adopters (%) Non-adopters (%) X2 All
Credit 78.0 64.4 2.7* 73.3
Extension service: 90.2 75.6 4.9** 85.1
No. of visits in past two years:
<5 55.0 74.1 61.6
5 -10 20.0 7.4 15.6
>10 25.0 18.5 22.7
Extension services on maize 80.5 68.9 2.2 (NS) 76.5
Membership in an organization 95.1 91.1 0.8 (NS) 93.7
Note: NS = not significant, = significant at p<0.1; ** = significant at p<0.05.











Logistic Model Estimates



Factors influencing the adoption of improved maize were analyzed using maximum likelihood
estimation of a logistic regression model. These results are presented in Table 14. The model has a
correct prediction rate of 75%, correctly predicting adopters at 89% and non-adopters at 44%.
Factors that influenced adoption were agroecological zone, gender, use of manure, and hiring labor
(all with significantly different coefficients from zero to 10%).

Interpretation of the coefficient with the logistic regression model is not as straightforward as the
linear probability (LP) model where coefficients estimate the change in probability to adopt.
However, dividing the logit coefficients by a factor of 4 gives an approximation of the linear
probability coefficients (Maddala 1983:23). Thus, the coefficient on gender is 2.21 and can be
interpreted as an LP coefficient of 0.55. The interpretation is that men, keeping all other factors
constant, have a 55% higher probability than women of adopting an improved maize variety.
Comparing the coefficient with that of the other factors, gender is clearly the most important.

The second most important factor is the use of manure, with a coefficient of 1.82. Using the same
reasoning, this indicates that farmers who use manure have a 45% higher probability to be
improved maize adopters than those who do not. Farmers from the UM2 zone have a 34% higher
probability (1.37/4) than those from the UM3 zone, and those who hire labor have a 32% higher
probability. The smallest significant factor was extension farmers with access to extension had a
24% higher probability to be improved maize adopters. Other factors hypothesized to influence
adoption did not have significant coefficients. They included age of household head, education,
farm size, credit, extension, area under coffee, and farmer group membership.



Table 14. Logit analysis for adoption of improved maize variety in Embu District, Kenya.
Variable Coefficient estimate (B) Standard error P-value
Gender (0=women, 1=men) 2.21 1.06 0.04**
Farm uses manure (1=yes, O=no) 1.82 0.82 0.03**
Agroecological zone (UM2=1, UM3=0) 1.37 0.60 0.02**
Farm hires labor (1=yes, 0= no) 1.29 0.55 0.02**
Access to extension (1=yes, O=no) 0.95 0.57 0.10*
Use of credit (1=yes, 0= no) 0.95 0.61 0.12
Education of household head (yr) 0.04 0.08 0.59
Age household head (yr) 0.05 0.02 0.56
Farm size (acres)t -1.25 1.08 0.25
Area under coffee (acres)t -0.02 0.10 0.85
Member of farmer group (1=yes, 0= no) 0.05 0.33 0.87
Intercept -5.59 2.15 0.01**
Model X2 30.59 0.00***
Overall cases correctly predicted (%) 72.97
Correctly predicted adopters (%) 88.0
Correctly predicted non-adopters (%) 41.67
Sample size (no.) 111
Note:* = significant at 10% level, **= significant at 5% level, *** = significant at 1% level. t 1 acre = 0.405 ha.











Factors that influence fertilizer use were analyzed by estimating a linear model since the
dependent variable is quantitative (kg/fertilizer/ha) and almost all farmers use some fertilizer.
Two different models are estimated: the first model uses total fertilizer use, expressed in kg/ha
(first two columns of Table 15): the second model uses total phosphorus in kg/ha. Only three
factors, hiring labor, credit, and education of household head, had a significant effect on the
quantity of fertilizer used. Hiring labor increased the amount of fertilizer used by 45 kg/ha,
while using credit increased it by 53 kg/ha. Education of the household head had an unexpected
negative impact for each year of schooling, fertilizer use decreased by 7 kg.




Table 15. Linear regression model for fertilizer use, Embu district, Kenya (dependent variable is total fertilizer use in
kg/ha and phosphorus in kg/ha)


Explanatory variable
Gender (0=Women, 1=Men)
Farm uses manure (1=yes, 0= no)
AEZ (UM2=1, UM3=0)
Farm hires labour (1=yes, 0= no)
Access to extension (1=yes, 0= no)
Access to credit (1=yes, 0= no)
Formal education of household head (yr)
Age household head (yr)
Farm size (acres)'
Area under coffee (acres)'
Member of farmer group (1=yes, 0= no)
Intercept
R
Sample size (no.)
Note: *= significant at 10% level, ** = significant at 5% level, ***:
t1 acre = 0.405 ha.


Total fertilizer (kg/ha)
Coefficient Standard
estimate (B) error
10.5 39.9
-44.5 34.7
32.4 21.5
45.4 23.6**
-1.2 24.2
52.8 23.1*
-7.3 3.4**
-1.7 0.97
-1.3 3.92
12.2 14.2
-35.4 43.4
260.2 78.9**
0.409
111
= significant at 1% level.


The variable 'total fertilizer used' indicates a mix of fertilizer with different composition of
nutrients. To avoid this accumulation problem, the nutritional components of each type were
summed. The second model uses the amount of phosphorus per hectare in each farm. Hiring of
labor is again significant (it increases P use by 6.4 kg), but credit is not. Formal education is again
significant, but the opposite of the first model: for each year the P use increases by 2.3 kg. Age is
also significant and increases P use by 0.56 kg for each year.


P (kg/ha)


Coefficient
estimate (B)
4.7
1.6
0.9
6.4
5.83
0.5
2.3
0.56
0.65
0.4
10.8
9.0
0.138
111


Standard
error
12.3
10.9
6.7
7.8**
7.9
7.3
1.1**
0.3*
1.2
4.4
13.4
24.0***


P(kg/ha)










Conclusion



This study was undertaken to identify socioeconomic and technical factors affecting the adoption of
improved maize seed and fertilizer use and the role of credit in both. The survey was conducted in
Upper Midland 2 and 3 (UM2 and UM3) (Jaetzold and Schmidt 1983) zones in the Nembure,
Runyenjes, and Kieni Divisions of Embu, which represents the main maize growing zones in the
area. These zones are dominated by the coffee/dairy land use systems.

The study covered both the long and short rainy seasons. One hundred and twenty-seven farmers
were randomly selected and interviewed. Using a structured questionnaire, data was collected on
farmer and farm attributes and institutional structure. Adopters of certified maize were defined as
farmers who planted certified maize for two consecutive seasons in 1998 and non-adopters were
defined as farmers who planted local seeds, recycled hybrid seeds or recycled seeds of open-
pollinated varieties for more than three seasons. A comparative analysis between adopters and non-
adopters of improved maize seed was done. Since productivity gains from maize seed depend on
simultaneous use of other inputs and particularly fertilizer, the level of fertilizer used by adopters
and non-adopters was also determined. Logit and linear models were used to analyze factors
affecting adoption of improved maize seed and quantity of fertilizer, respectively.

The study found that most adopters of certified seed preferred Pioneer H3253. While there was a
marked increase in the use of Pioneer H3253 between 1996 and 1998, the proportion of farmers
using older hybrid varieties, H511 and H512, decreased in the same period. Adopters preferred
Pioneer hybrid for its high yield and early maturity, despite its high price, poor storage, and root
lodging abilities. Other valued traits mentioned were pest and drought tolerance, large grains, taste,
good threshing quality, and ease of cooking. Most adopters preferred the smaller 2-kg seed package
because it was affordable and sufficient for their plots of maize.

The high price of improved maize seed was the main constraint for adoption for most farmers.
Other constraints were the low selling price of maize and lack of credit.

Nearly all farmers used basal fertilizer. More adopters used top dress fertilizers than non-adopters.
Compound fertilizer 23:23:0 was the main basal fertilizer, and calcium ammonium nitrate was the
main top dress fertilizer used by both categories of farmers in the long and short rains. Most maize
farmers in the study area applied much less than the optimal level of fertilizer.

Improved maize seed adopters used more labor for farm operations than non-adopters. More
adopters obtained formal credit from co-operative societies in the form of seed and fertilizer (p<0.1).
About half of adopters and non-adopters used credit to purchase fertilizer for coffee or maize. The
average amount of credit was Ksh 6,802 for adopters and Ksh 3,613 for non-adopters. To get credit
from the cooperative society, farmers had to deliver the coffee crop. The main reason cited by
adopters for not using credit was availability of capital and that it did not pay to use credit in maize
production. For non-adopters, the main constraint to credit use was lack of collateral.










Generally, adopters had greater access to extension than non-adopters. The main source of extension
was the Ministry of Agriculture. More adopters than non-adopters received advice on improved
maize production. However, this difference was not significant. The main type of extension advice
was on fertilizer use.

Agroecological zones, gender, manure use, hiring of labour, and extension were statistically
significant in explaining adoption of improved maize variety. Other variables, such as age and
education of household head, farm size, credit, years of formal education, area under coffee, and
farmer group membership, which were expected to influence adoption and fertilizer use were not
significant (at 10% or lower probability level). Likewise, agroecological zone, gender, manure use,
hiring of labour, and extension were important variables in explaining the amount of basal
fertilizers farmers applied.

In light of the many attributes considered in variety selection, the use of recycled seed by some
farmers on grounds of high price and unnoticed differences in yield of improved and recycled seed,
it is necessary to focus on farmer participatory breeding to incorporate farmers' assessment of maize
varieties in the research process. There is also need to undertake studies on the economics of seed
recycling to establish whether there is justification for farmers using recycled seed. The packaging of
maize seed in small and more affordable packages such as the 2-kg bags should be encouraged to
increase adoption of certified maize.











References


Central Bureau of Statistics.2001.1999 Population and Housing Census.
Volume I: Population Distribution by Administrative Areas and
Urban Centers. Nairobi: Ministry of Finance and Planning.
Chilot Y., B.I. Shapiro, M. Demeke.1996. Factors affecting adoption of new
wheat technologies in Wolmera and Addis Alem areas of Ethiopia.
Ethiopian Agric. Econ.1 (1): 63-84.
CIMMYT Economics Program. 1993. The Adoption of Agricultural Technology:
A Guide for Survey Design. Mexico, D.E: CIMMYT.
D'souza, G., D. Cyphers, and T. Phipps. 1993. Factors affecting the adoption
of sustainable agricultural practices. Agric.Res.Econ.Rev. 159-165.
Feder, G., E.R. Just, and D. Zilberman. 1985. Adoption of agricultural
innovations in developing countries: A survey. Economic
Development and Cultural Change 33: 255-298.
Gerhart, J. 1975. The Diffusion of Hybrid Maize in Western Kenya. Research
report for CIMMYT. Mexico, D.F: CIMMYI. Mimeo.
Gujarati, D.N. 1988. Basic Econometrics. 2nd edition. New York: McGraw-Hill.
Hassan, R.M, K. Njoroge, M. Njore, R. Otsyula, and A. Laboso.1998.
Adoption Patterns and Performance of Improved Maize in Kenya in
Maize Technology Development and Transfer A GIS Application for
Research Planning in Kenya. Oxford, U.K.: CABI.
Hassan, R. M., and D. Karanja.1997. Increasing Maize Production in Kenya:
Technology, Institutions, and Policy. In: D.Byerlee and C.K. Eicher,
Africa's Emerging Maize Revolution. Boulder, Colorado/London, UK:
Lynne Rienner Publishers. Pp. 82-93.
Heisey, P.W, and W. Mwangi. 1997. Fertilizer and maize production. In:
D.Byerlee and C.K. Eicher, Africa's Emerging Maize Revolution.
Boulder, Colorado/London, UK: Lynne Rienner Publishers.
Pp. 193-211.
Hussain, S.S., D. Bylerlee, and P.W. Heisey. 1994. Impacts of training and
extension system on farmer's knowledge and adoption of
technology: Evidence from Pakistan. Agricultural Economics 10:
39-47.
Jaetzold, R. and H.Schmidt. 1983. Farm Management Handbook of Kenya.
Vol. Il/C. Nairobi,Kenya: Ministry of Agriculture.
Maddala, G.S. 1983. Limited Dependent and Qualitative Variable in
Econometrics. Econometric Society Monographs, Cambridge
University Press.
Matiri, F.M, J.O. Ouma, F.M. Murithi, and J.M. Ndubi. 1996. Assessment of
the level of adoption of maize production technologies: A case study
of Meru district, Kenya. Paper Presented at the 1st KARl
Socioeconomics Conference, November 20- 24, 1996, Nairobi,
Kenya.


Ministry of Agriculture. 1998. Embu District Annual Report for the
Department of Agriculture 1998. Embu, Kenya: Ministry of
Agriculture.
Murithi, EM, G. Macharia, R.M. Hassan, and N. Macharia. 1994. Maize
Farming in the Mid-Altitude Areas of Kenya: Farmers Practices,
Research, and Extension Challenges and the Potential for Increased
Productivity Nairobi, Kenya: KARI/CIMMYT Maize database Project
Report.
Ntega-Nanyeeya, W., M. Mugisa-Mutetikka, W.Mwangi, and H.Verkuijl.
1997. An Assessment of Factors Affecting Adoption of Production
Technologies in Iganga District, Uganga. Addis Ababa, Ethiopia:
CIMMYT and National Agricultural Research Organization (NARO).
Pingali, P.L. 2001. CIMMYT 1999-2000 World Maize Facts and Trends.
Meeting World Maize Needs: Technological Opportunities and
Priorities for the Public Sector. Mexico, D.E: CIMMYT.
Poison, A., and D.S. Spencer. 1991. The technology adoption process in
subsistence agriculture: The case of cassava in Southern Nigeria.
Agricultural Systems.36: 65-78.
Rashid, M.H, K. Njoroge, M. Njore, R. Otsyula, and A. Laboso. 1998.
Adoption Patterns and Performance of Improved Maize in Kenya.
In: R.M Hassan (ed.). Maize Technology and Transfer: A GIS
Application for Research Planning in Kenya. CIMMYT and
KARI:CABI.
Republic of Kenya. 1983. National Development Plan for the Period 1984-
1988. Nairobi, Kenya: Government Printers.
Republic of Kenya. 1986. Economic Management for Renewed Growth.
Sessional Paper No. 1 of 1986. Nairobi, Kenya: Government
Printers.
Central Bureau of Statistics. 2001. 1999 Population and Housing Census.
Volume 1. Population Distribution by Administrative Areas and
Urban Centres. Nairobi, Kenya: Ministry of Planning and National
Development, Central Bureau of Statistics (CBS).
RRC-Embu. 1998. Recommendation Guidelines for Crops, Soil fertility, and
Livestock Management (up to 1998). Embu, Kenya: RRC-Embu.
Salasya, B.D, W. Mwangi, H. Verkuijl, M.A. Odendo and J.O. Odenya. 1998.
An Assessment of the Adoption of Seed and Fertilizer Packages and
the Role of Credit in Smallholder Maize Production in Kakamega
and Vihiga Districts, Kenya. Mexico, D.E:CIMMYT and KARl.
Yahanse Kebede, K. Gunjal, and G. Coffin. 1990. Adoption of new
technologies in Ethiopian agriculture: The case of Tegulet-Bulga
District, Shoa Province. Agric.Econ. 4: 27- 43.











ANNEX


Table Al. Food and cash crops grown in Embu District Kenya.
Farmers growing (%)
Crops Adopters Non-adopters
Maize 100.0 100.0
Beans 96.3 97.8
Coffee 93.9 95.6
Napier 95.1 88.9
Bananas 68.3 53.3
Irish potatoes 24.4 41.2
Sweet potatoes 9.8 6.7
Mangoes 4.9 2.2
Macadamia 6.1 2.2
Cabbages 6.1 2.2
Tomatoes 1.2 2.2
Kale 4.9 6.7
Arrow roots 1.2 2.2
Avocado 1.2 2.2
Cassava 12.2 8.9














































































24








ISBN: 970-648-093-5




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