dc.description.abstract | Agriculture is the main economic sector in Kenya and contributes significantly to national
development. For the sector to play this centra! role in a sustainable way, rapid growth in
output and productivity is critical. One of .the major factors that continue to constrain
agriculture is the low and declining fertility of land; the problem is even more pronounced in
the semi-arid areas.
This study focused on developing strategies for improving adoption of soil fertility and water
management technologies in the semi-arid areas of Machakos, Makueni and Kitui districts.
Following the low adoption of soil fertility and water management technologies and
consequent fall in yields in the semi-arid areas of Eastern province, there is need for
technological recommendations that are specific to farm types. This is expected to take care
of the differences between farm types.
A total of 228 farmers were interviewed during the period January/February 2004 using a
single-visit survey approach. Geographical Information System (GIS) guided random
sampling methodology was used to select farmers to be interviewed and the data obtained
using semi-structured questionnaires.
The logistic regression model was applied and the results showed that off-farm employment,
hired labour, maize output, agricultural extension and agro-ecological zone positively
influenced fertilizer adoption, while the distance to the nearest market was negatively related
to fertilizer adoption. Off-farm employment, livestock ownership, distance to the nearest
market and agricultural extension positively influenced animal manure adoption, while
education negatively influenced the adoption of animal manure. Hired labour use positively
influenced compost manure adoption and the distance to the nearest market negatively
influenced compost manure adoption. Maize output positively influenced the adoption of soil
and water conservation structures, while the distance to the nearest market and agroecological
zone were negatively related to the adoption of this practice. These factors should
be incorporated in the design of policies and strategies for soil fertility improvement.
As a result of the need to design specific soil management strategies, three major farm types
were identified in this study, using k-rnean cluster analysis. Farmers/farms were classified as
socio-economically unconstrained, resource and information constrained and socioeconomically
constrained. The identified farm types had varying technology adoption
abilities that decreased with an increase in the group socio-economic constraints. To increase
the adoption of improved soil fertility practices, short-term and long-term strategies were
developed for each farm type. The short-term strategy was to improve on the use of what is
adoptable and the long-term strategy was to relax the constraints associated with the
respective farm types. These strategies are expected to ensure better soil fertility technology
adoption and higher crop yields. The study also recommended that the strategies be
implemented hierarchically, starting with the socio-economically constrained group, who
were only able to adopt animal manure. | en |