Un esperimento di misurazione del valore aggiunto delle scuole sulla base dei dati PISA 2006 del Veneto

A school value added measure by using OECD-PISA 2006 data of Veneto Objectives The principal aim of this paper is to find a school value added measure by considering OECD-PISA 2006 data of Veneto and individual school evaluations of sampled students. The rankings of 45 upper secondary schools based on the science scale scores are compared by using row cognitive data and value added measures. The work analyzes to what extent they differ and finds out the schools that show results significantly below or above the average. Methods and Results To realize the objective of the research a two level hierarchical linear model has been estimated. The measure of the initial cognitive level of pupils is based on the marks given to them by teachers in the first year of upper secondary education and on the final evaluation (expressed on a four level scale) got by the same students in the leaving examination of lower secondary education about two years before the PISA survey. These evaluations are summarized in one indicator calculated by using the Rasch Analysis. The comparison between row data and value added indicators shows a relevant difference among the rankings. It is remarkable that the value added ranking puts in evidence better results are obtained not by academic schools but by technical schools and sometimes even by vocational schools. Conclusions This work suggests that the evaluation of school outcomes obtained by comparing row scores leads to wrong conclusions. The paper shows explicitly that initial cognitive level of pupils attending a school is relevant and it has to be taken seriously into account in order to evaluate appropriately students results.

Jel codes: I21

Angela Martini, Roberto Ricci, Un esperimento di misurazione del valore aggiunto delle scuole sulla base dei dati PISA 2006 del Veneto in "RIVISTA DI ECONOMIA E STATISTICA DEL TERRITORIO" 3/2010, pp 80-107, DOI: 10.3280/REST2010-003004