Finally some policy implications and considerations. First, in spite of the limits due to the data constraints (parents’ education cannot be taken into account), if dropout is due more to personal factors (family background and pre-enrollment individual characteristics) than to a real scarcity of education supplied, then evaluating universities on indicators such as the number of students who persist between the first and the second year might not guarantee objectivity in assessing the quality of the universities. Second, according to the empirical findings once the entrance test score have been taken into account, university can raise access standard based on this indicator in order to reduce dropout rates.