Dataset and Supporting Information for: Evaluating an Open-Source R Programming Training Program for Agricultural Researchers in Ethiopia

### Abstract Open-source software like R is critical for global knowledge production, yet researchers in low- and middle-income countries (LMICs) often encounter digital inequities and a lack of context-specific training. This study evaluates an intensive R programming capacity-building initiative delivered to 255 agricultural researchers and university instructors in Ethiopia between 2022 and 2024. Using a mixed-methods approach across 11 workshops, the program employed instructor-led live coding grounded in domain-specific applications. Quantitative analysis of 201 participants (79% response rate) demonstrated high perceived satisfaction (Median = 4.0; IQR = 1.0). Exploratory Factor Analysis confirmed a robust three-factor structure (Technical Proficiency, Instructional Delivery, and Outcomes) accounting for 61% of the total variance. Results indicate equitable perceived impact across gender, institutional affiliation, and educational levels (p > .05); conversely, prior programming experience significantly mediated perceived mastery (d = -0.63), suggesting that existing schemas substantially reduce the cognitive burden of R syntax. Qualitative co-occurrence mapping identified an infrastructural ceiling, where pedagogical success was frequently constrained by environmental barriers such as power instability and connectivity volatility. While the lack of objective performance-based metrics, longitudinal tracking, or a control group limits causal claims regarding long-term skill acquisition, these self-reported findings demonstrate strong participant acceptance and underscore the necessity of resilient instructional design in LMICs. Future initiatives should shift from universal models toward a structural, Three-Tiered Scaffolding Framework incorporating distinct tiered learning trajectories, sustained follow-up mechanisms, and offline-compatible resources to ensure high learner satisfaction translates into verified, long-term technical autonomy within national research systems.