"background": "Understanding the dynamics of off-grid energy adoption is critical for rural development, yet methodological limitations in longitudinal analysis persist. Existing approaches often rely on cross-sectional data, which fail to capture temporal dynamics and household-level heterogeneity in adoption decisions, particularly in agricultural communities. ", "purpose and objectives": "This article presents a methodological framework for constructing and analysing a household-level panel dataset to estimate the determinants and rates of off-grid system uptake in rural settings. The objective is to provide a replicable model for measuring adoption trajectories and causal drivers. ", "methodology": "The framework details a stratified random sampling procedure for a rotating panel of agricultural households, with biennial surveys. The core analytical model is a random-effects probit specification: Pr (y{it=1|Xit, \) =\ (\0+\1Xit+\+), where y₈ₓ indicates adoption. Estimation employs maximum likelihood with robust standard errors clustered at the village level to account for intra-cluster correlation. ", "findings": "As a methodology article, it presents no empirical results. However, simulation exercises based on pilot data indicate the model's capacity to identify significant positive effects of seasonal agricultural income on adoption likelihood, with a hypothesised coefficient of 0. 15 (95% CI: 0. 08, 0. 22) for a one-standard-deviation increase. ", "conclusion": "The proposed framework provides a rigorous, statistically sound methodology for generating nuanced evidence on technology adoption pathways. It addresses key limitations of prior static analyses by modelling unobserved household heterogeneity and temporal change. ", "recommendations": "Researchers should employ panel designs with at least three waves to control for time-invariant unobservables. We recommend the collection of high-frequency data on agricultural yield and energy expenditure to refine the model's predictive power for policy simulation. ", "key words": "energy access, panel data, random-effects probit, technology adoption, rural households,
Nkurunziza et al. (Sat,) studied this question.