Thanks to the effort of two local educational authorities, in two regions of North Italy (Valle d'Aosta and the autonomous province of Trento) the PISA 2009 test was re-administered to the same students one year later. This paper is the first to analyse in the OECD-PISA context the potential advantages of re-testing the same students in order to provide better measures of schools' contributions to student achievement. We show that while cross-sectional measures of school value added based on PISA student literacy, which measures "knowledge for life", tend to be very volatile over time whenever there is a high year-to-year attrition in the student population, longitudinal measures of school value added are very robust to student attrition (even without controlling for sample selection). Moreover, persistence in individual test scores tends to be higher in highly "selective" (i.e. high drop-out) school environments.