FoodLAND - Datasets of randomised controlled trials of farm innovations implemented in Morocco, Tanzania, and Uganda

This submission derives from Task 5.9 "Innovation validation across the network" of the H2020 project FoodLAND "Food and Local, Agricultural and Nutritional Diversity" (2020-2025). T5.9 had two objectives: (1) assessing the impact of different dissemination strategies (information provision or financial incentivisation) on the adoption of selected FoodLAND innovations by smallholder farmers; (2) assessing the impact of these innovations, if adopted and used, on various indicators of farm performance and household wellbeing. To achieve these closely related goals, T5.9 implemented Randomised Controlled Trials (RCTs) in three so-called Food Hubs: Meknès (Morocco), where a phone application for precision irrigation and protection was considered; Mvomero (Tanzania), where the focus was on two new bean varieties characterised by higher zinc and iron content; and Kajjansi-Masaka (Uganda), where fish farmers were introduced to the integration between aquaculture and vegetable farming. The RCTs were conducted between May 2023 (initial survey) and early January 2025 (post-implementation/final survey), depending on the country and specific local conditions. A detailed description of the methodological approach and the data collection tools is provided in deliverable D5.20 "Protocol to implement the randomised controlled trials of the innovations." This submission consists of three datasets, one per Food Hub, plus the report D5.21 "Dataset of RCTs of innovation results, linked to socioeconomic and demographic factors." The datasets are provided in Excel Workbook format. Each dataset consists of eight (Morocco and Tanzania) or 13 (Uganda) worksheets, plus the metadata and the sheet descriptions. The sheets are the following: baseline survey (base_survey), baseline crop production (base_crop_prod), baseline crop inputs (base_crop_input), planned crop production (plan_crop), baseline fish production – Uganda only (base_fish_prod), baseline fish inputs – Uganda only (base_fish_input), planned fish production – Uganda only (plan_fish), post-training survey (post_survey), final survey (final_survey), final crop production (final_crop_prod), final crop inputs (final_crop_input), final fish production – Uganda only (final_fish_prod), final fish inputs – Uganda only (final_fish_input). Each variable is assigned a short name, e.g., “treatment,” and a long name that conveys information about the country and the worksheet, e.g., “u_post_surv_treatment.” The unique ID of the farmer, “unique_id,” appears with the same name in all the worksheets to facilitate merging during the analysis; all the variable names include 32 characters or less.

Update 2025-05-23: The Ugandan dataset was reviewed based on new observations, some errors were corrected in the Moroccan dataset, and the report D5.21 was updated accordingly.