An evaluation of the effects of information and technology characteristics on technology choice and adoption The case of Striga and Stemborer control technologies in maize production in western Kenya

Abstract

Kenya’s population is increasing, thus leading to increased demand for food, especially maize which is a staple food for most of the Kenyan population. Demand for maize in Sub Saharan Africa is projected to double to 52 million tons by 2020 (Pingali and Pandey). The increasing population has also caused the per capita farm holdings in Kenya rural areas to become smaller due to continuous sub-division. Besides the reduced farm holding, maize production is constrained by both biotic and abiotic factors. Two of the most important biotic constraints in maize production are Striga and stemborer infestations.Maize yield loss due to stemborers has been estimated to range from 20-80% depending on the severity of the infestation by the pest and the growth stage of the crop (Khan et al, 1997a). Striga is a highly invasive parasitic weed and it infests more than 400,000ha of Kenyan farmland (Kanampiu,2003). In western Kenya, 50-100 percent yield losses due to Striga have been reported in both on on-farm and on-station experiments (Hassan et al, 1994).Since the opening of new land is not a viable option in contributing to increased maize production in Kenya, increasing maize production is dependent on two factors; the ability of the research and development agents to supply constraints mitigating technologies and the ability of the farmers to access and utilize such technologies. The purpose of this study is to evaluate the factors that influence the farmers’ knowledge, choice and adoption of technologies for the control of Striga/stemborer in Vihiga and Suba Districts. Data was collected from 476 randomly selected farm households through a cross-sectional household survey, using a pre-designed questionnaire, in 2006. Principal Component Analysis (PCA) was used to generate and classify two proxy indices that were used to represent the type of agricultural information as obtained by farmers from various sources. Multivariate regression was used to evaluate the factors that influence the agricultural information indices. Conjoint analysis was applied to assess the technology characteristics that the households consider important in influencing choice of Striga control technology. A bi-variate probit and Tobit analyses were used to assess the factors that influence awareness, adoption and use of the Push-Pull Technology (PPT).Results show that the farmers accumulate two distinct types of agricultural knowledge (i.e. type I and type II) from 15 information sources, which significantly influence their decision to adopt a technology to control Striga or stemborer. Technology messages are better disseminated using a host of different communication channels. Different farmers’ access agricultural information through different channels, depending on their education level, association with groups, relationships with neighbours, and opportunities to travel and visit research institutions, among others. The most popular information channels, based on those accessed by over 30% of the farmers in both districts, are farmer groups, contact farmers, pamphlets and brochures, agricultural shows, PPT farmer teachers and posters.Farmers also identified seven important characteristics that influence their choice when adopting a technology to control Striga in their cereal crops, including: (i) change in costs per season, (ii) if a fallow period is required, (iii) yield increase attained, (iv) possibility of intercropping food legumes, (v) additional benefits like animal feed supplements or firewood obtained from the technology, (vi) requirements for crop rotation and (vii) the amount of labour required. Except for crop rotation, the influence of the other six characteristics was validated as being statistically significant in influencing the probability of a farmer choosing a Striga control technology for adoption. Farmers’ assessment of technologies in terms of these characteristics should be integrated into the ex-ante diagnostic surveys so that the farmers’ preferences are included in technology design. The adoption decision for Push-pull technology (PPT) was modelled as a three-phase decision model: (i) factors influencing awareness of existence of PPT, (ii) factors influencing adoption of PPT, and (iii) factors influencing the intensity of use of PPT. The results show that 30% of the respondents knew of the existence of the PPT and this was significantly influenced by age, travel time to the nearest market, and being in a village used for PPT demonstration. Of those who were aware of the PPT’s existence, 19% had decided to adopt the PPT and this was significantly influenced by whether they had access to farm labour, the area under maize, farmers’ assessment of the severity of the Striga infestation, farm income and farming experience. The intensity of use (proportion of their maize farm put under PPT) was significantly influenced by farming experience,farm income and land tenure security index. The residual effects from the awareness and adoption phases were also significant in influencing the intensity of use of the PPT. Push-pull technology is not yet widely known by the farmers. This study recommends that strategic information dissemination programmes, in collaboration with the Ministries of Agriculture and Livestock Development, NGO’s and CBO’s partner organizations, would increase the number of farmers reached and made aware of the existence of PPT and its benefits. This is critical in order to enhance its adoption and thus ensure food security in the Striga infested areas of western Kenya.