Cluster Analysis of School Performance in Mexico City

Leon Palafox, Daniel Villegas

TL;DR: We analyze for the first time the correlation between quality of education and living quarters of students in Mexico City

Keywords: clusters, education

Abstract:

Like many others around the world, Mexico's education system presents a problem that afflicts the performance of its students as well as its overall quality. However, the educational problem obeys to particular functions of each country, state, and locality.

Many factors could be accounted to have an influence on the evaluation of a school inside its district, including students performance, the number of students evaluated, the municipality in which it is located, the overall budget it possess to invest in quality education and hiring of teachers as well as socioeconomic factors such as the cost of living in the area where it is located, etc.

Having a way to visualize and identify this problem becomes a major point in trying to mitigate an education deficit amongst regional government and policymakers. This becomes a complex socio-cultural problem that involves many dimensions of the schools of a region as well as the timeframe in which the study is made with many dynamic factors playing a key role in which schools are afflicted and can help map out and identify them.

In this work we test different Machine Learning techniques to try to identify what are the primary factors in the performance of schools in Mexico City, the applications of this works may help in the future to create new public policy in the education and further studies that try to model a solution for this problem.