Robust diagnostics in regression models for the evaluation of the university teaching effectiveness: the case of graduates of Florence.

The aim of this study is to validate the robust diagnostics approach as a tool for the analysis of the degree programs activities effectiveness. This approach allows identifying singles or groups of students with particular characteristics, which could help to implement universities programmes on teaching activities aimed at improving the quality of this service. Some distinct analyses are developed, within the classical and the robust diagnostic approaches, using a new indicator defined as skills acquired during the university studies, used in the job. In particular, logit models, which take into account the graduates' characteristics and the effects pertaining to degree programs and jobs, are used on the dataset from the survey on job opportunities of all the 1999’s graduates at University of Florence.