Effect of participation in commercial production of medicinal plants through community-based conservation groups on farm income at Kakamega forest, Kenya
Nyangau, Paul et al. (2019), Effect of participation in commercial production of medicinal plants through community-based conservation groups on farm income at Kakamega forest, Kenya, Dryad, Dataset, https://doi.org/10.5061/dryad.nzs7h44mq
Forest conservation by peripheral communities is one of the most advocated strategies for ensuring sustainable supply of forest goods and services. In Kenya, although conservation activities generate considerable environmental improvements, studies evaluating the economic benefits are very limited. This study aimed at assessing the impact of participating in the commercialization training program by forest adjacent communities on farm income. Utilizing matching techniques, the analysis is based on the data collected from a randomly selected sample of farm households around Kakamega forest in Kenya. The results show that age, forest distance, forest benefit, market distance and farming experience are the main factors that influence the participation in the commercialization program. Also, participating in commercial production of medicinal plants has a positive and significant effect on farmers’ income. The average treatment effect on the treated (ATT), based on three estimation algorithms, ranges from KSH. 80,047 to KSH. 176,788 per hectare, per year, implying that policy efforts that focus on the participation in the commercialization program of medicinal plants can enhance incomes among farm-dependent households. The findings from this study suggest that upscaling the participation in the commercialization program to other areas.
A two-stage stratified random sampling technique was employed to identify the household level sample units. First, five major medicinal plant producer groups involved in the icipe project were selected purposively with the help of agricultural experts in the area. Second, 5 farmers from each group were randomly selected from the list of members summing up to 25 households. An additional 25 households were randomly selected among households that were not participating in the commercial production of MPs program but living in the same locality with those who were participating in the project. Data was collected from the selected households using pre-tested semi-structured questionnaires by trained and experienced enumerators who understood the local language. Data collected included information on household demographic characteristics, household monthly expenditure, access to training, the perceived value of the forest including source of medicinal plants, availability of timber and fire wood and livestock grazing, production and marketing of MPs and other farm crops and returns generated from sales.
Given the cross-sectional nature of our dataset, we employed matching methods in estimating the observed mean outcome differences between the treatment and control groups.
No missing values
European Union for Switch Africa Green project
United Nation Office for Projects Services (UNOPS)