Academic performance on urban-rural gradients

Joaquín Gairín, together with Fabián Santos-García, Karina Delgado Valdivieso and Andreas Rienow, have published the article “Urban–Rural Gradients Predict Educational Gaps: Evidence from a Machine Learning Approach Involving Academic Performance and Impervious Surfaces in Ecuador”. The publication focuses on using a spatial perspective to explain academic performance in secondary studies in urban-rural settings in Ecuador.

Academic performance is explained by a multitude of factors, mainly those related to the socioeconomic, cultural, and educational environment. However, academic performance is less understood from a spatial perspective. The aim of this study was to investigate a learning-focused methodology to determine which responses to a questionnaire-based survey were relevant in explaining the high academic performance of high school students on urban-rural gradients in Ecuador. High school locations were used to construct individual data sets and stratify them based on academic achievement scores. The Boruta algorithm was used, the best predictors were identified, it was classified randomly and the probabilities of academic performance classification were mapped. These results were summarized as observed frequent responses for each natural region in Ecuador and their probability outputs were used to formulate hypotheses regarding the urban-rural gradient derived from maps of impervious surfaces.

Read the article here.