E-registration, spatial referencing and tracking of farmers for innovation scaling in Burundi

This report presents a dataset of the e-registration of farmers in Burundi for assessing the adoption of innovations and the diffusion of new technologies. Data was collected from actors after a census conducted in three steps. First, main crops production regions and value chain actors were identified. In the second step, the list of actors was updated based on membership of their associations. Finally, a census of all individual actors was conducted as well as the geo localization of all farmers’ fields and villages using GPS device. Data were collected for the 2022 growing seasons and the dataset contains 1,751 observations with 159 variables divided into six sections: (i) preliminary information on the respondents; (ii) socio-economic characteristics; (iii) information on the rice plots; (iv) knowledge, use and access to rice varieties; (v) knowledge, use and access to agricultural equipment and methods; and (vi) information on post-harvest activities. Five categories of actors were identified: seed producers (127), crops producers (1,394), millers (35), traders (231) and service providers (8). On average, a farmer grew three crops. The main crops of farmers were rice (745) followed by cassava (254), banana (202), potato (181) and beans (131). The dataset is valuable for the diffusion at large scale of improved technologies and an effective monitoring of the dissemination. Data can be used by scientists to have better understanding of crops value chains, production systems, the level of knowledge, accessibility and adoption of improved rice varieties and agricultural technologies, for further research in the field of rice value chain development, technologies testing and socioeconomic studies of rice value chain actors and other crops such as cassava, banana, potato, and beans. Because of the large number of observations (1,751 actors), data can be used as sampling frame for further experiments or surveys based on random samples. Moreover, the dataset has the potential of generating descriptive statistics at the most disaggregated level of administrative units or villages for different equipment, methods and varieties adopted by gender and country.