Secondary schools efficiencyand non-controllable inputs: sensibility of the results to different DEA model specifications

The aim of this paper is to study how much sensitive efficiency scores are when different approaches proposed in the literature are used to include exogenous factors in efficiency analysis. Specifically, we will concentrate on two crucial issues. One the one hand, although we have an extensive data set on these variables derived from school surveys, it is not possible to use all these data, so we consider the possibility of either including the most relevant variables in the efficiency analysis or using Principal Components Analysis to summarise information contained in such variables. On the other hand, two alternative methodologies are considered to include these factors in efficiency analyisis, we will compare the approach according to which such factors must be taken into account to calculate the final efficiency indices and alternative multi-stage approaches. The analysis covers 79 public high schools in the region of Extremadura (Spain) for the 2001-02 school year.