Spatial data analysis of Mahatma Gandhi national rural employment guarantee scheme and its influencing factors

As a part of the Mahatma Gandhi National Rural Employment Guarantee Act (MGNREGA) program in India, durable assets under various work categories are created in rural areas by employing adult members belonging to the marginal communities. Taking the Prakasam district in Andhra Pradesh as an example, we analyzed the spatial clustering of the various work implemented under MGNREGA using kernel density estimation (KDE). We also analyzed the spatial clustering of villages in terms of overall assets and their statistical significance using hotspot analysis (Getis-Ord). The socio-economic factors influencing village-level clustering are analyzed using a machine learning model. Results show significant spatial variations in kernel density and village hotspots, thus, indicating the demand-driven nature of the program as influenced mainly by the marginal worker population female followed by the main other worker population. The study thus offers a methodological framework that may help improve and complement the empirical indicators currently employed in mapping the performance of the MGNREGA and similar programs.