This dataset is part of a larger research project that the Chalmers University of Technology conducts in east Africa with respect to solar PV powered mini-grids, rural electrification, and its impact on sustainable development goals. The aim of the data collection was thus to gather relevant social, economic, electrification, geographic, demographic and energy use data in order to investigate the impact of solar mini-grids on the economic development, income generation and biomass fuel consumption of rural off-grid households in two remote towns in Ethiopia. Since, the data analysis approach to be used was initially determined to be Propensity Score Method (PSM), the data collection was made in two separate groups: Intervention (mini-grid electrified) groups, and control (non-electrified) groups.As part of efforts to increase access to electricity and promote socio-economic development in rural areas, many sub-Saharan African countries are deploying decentralized solar mini-grids. However, little empirical research has been done to examine the impact of these technologies. Using propensity score matching, this paper examines the effects of rural electrification through distributed solar mini-grids on household income, wood-fuels use and kerosene expenditures. Data were collected from a survey of 464 sample households (212 electrified and 252 non-electrified) in two remote towns in Ethiopia.

The data collection method used is Survey, administered through through face-to-face interviews with a total of 464 sample rural households and small businesses in two remote towns in southern Ethiopia. Of the total 464 HHs, 212 are electrified (intervention group– IG), and 252 are non-electrified (control group–CG). A multi-stage purposive random sampling technique is used to select sample HHs for the survey in each town. First, all MG-electrified HHs in each town were identified from the EEU’s complete list of legally connected customers. The district EEU office keeps track record of its customers including the customer's name, address, date of electrification, electricity consumed, etc. In the second stage, all HHs that had been electrified during the first two months of the MGs’ operation were identified and a list of HHs for the IG was created. The “initial two-month electrification” criterion was deliberately set to ensure that the treatment (electrification) is a point-treatment rather than a time-varying treatment where HHs receive MG connection at various time points. This is because, in a time-varying treatment situation, the difference in the treatment period can confound the estimated average treatment effect on the treated HHs – our outcome variable (VanderWeele and Hernan, 2013).