University admission test and students’ careers: evidence from the School of Economics in Florence

In the academic year 2008/2009, the School of Economics of the University of Florence introduced a compulsory test to evaluate the background of the students wishing to enrol in a degree program. In this paper, we assess the predictive power of the test score in terms of number of gained credits, making comparisons with the predictive power of variables recorded in administrative data, such as the type of high school and the high school final grade. To disentangle direct and indirect effects, the result of the admission test is treated as an intermediate variable in a regression chain graph. About 20% of the enrolled student did not gain any credit at the end of the first year, thus we consider a two-part (hurdle) model, in order to deal correctly with excess zeros.