Association for Public Policy Analysis & Management

This paper provides an algorithm to predict which students are going to drop out of high schools relying only on information from 9th grade. It verifies that using a naïve model - as implemented in many schools - leads to poor results. It shows that schools can obtain more precise predictions by exploiting the available high-dimensional data jointly with machine learning tools such as Support Vector Machine, Boosted Regression and LASSO. Goodness-of-fit criteria are carefully selected based on the context and the underlying theoretical framework: model parameters are calibrated by taking into account policy goals and budget constraints. Finally, unsupervised machine learning is used to divide students at risk of dropping out into clusters.