rizao_rice_value_chain_analytical_data
Description
The dataset contains 2,660 observations collected through face-to-face surveys conducted between October and November 2024 in Côte d’Ivoire, Senegal, and Togo under the Rice in West Africa (RIZAO) programme. It includes rice value-chain actors such as producers, traders, retailers/distributors, input sellers, processors, mechanized-service providers, and other economic actors. The analytical dataset contains variables on demographic and socioeconomic characteristics, digital access and use, agricultural application awareness and use, organizational and institutional linkages, information access and coordination, productive practices and innovation, climate information, agricultural credit and insurance, household assets, and resilience-related capacities. It also includes constructed indices used in the analysis, namely the digital access index, digital use index, social capital index, information-coordination index, productive innovation index, and resilience-capacity index, as well as standardized variables and model-specific analytical sample indicators. The dataset used for the main digital-use model includes 2,314 economic actors, while producer-focused analyses use up to 2,003 producers. The information-coordination, productive-innovation, and final resilience models use 1,288, 1,192, and 1,189 observations, respectively.
Files
Steps to reproduce
Steps to reproduce Open the supplied Stata dataset rizao_rice_value_chain_analytical_data.dta in Stata. The file contains the cleaned analytical variables, constructed indices, standardized variables, sample indicators, and clustering identifier used in the study. Restrict each analysis to the relevant analytical sample. The digital-use model uses 2,314 economic actors; the information-coordination model uses 1,288 actors; the productive-innovation model uses 1,192 producers; and the resilience-capacity model uses 1,189 producers. Estimate the four sequential OLS models using the standardized outcome and explanatory variables. The analytical sequence is: (i) digital access → digital use; (ii) digital use and social capital → information coordination; (iii) information coordination and social capital → productive innovation; and (iv) productive innovation, digital use, information coordination, and social capital → resilience capacity. Include the controls specified in the manuscript: sex, age, age squared, education level, household size, rice-activity experience, country, and position in the rice value chain. Producer-specific models additionally control for seed-producer status and the logarithm of cultivated area. Cluster standard errors at the village level using the variable village_cluster. Include the digital access × social capital interaction in the digital-use model and the digital use × social capital interaction in the information-coordination model. Report standardized coefficients, clustered standard errors, p-values, number of observations, and R². These specifications reproduce the main results reported in Table 3. For robustness analyses, restrict the resilience model to paddy producers, replace the resilience index with the equally weighted alternative, replace the productive-innovation index with the broader innovation measure, and replace the composite social-capital index with current organizational membership. Gender and youth heterogeneity can be reproduced using the corresponding interaction specifications described in the manuscript.