Evaluating the University System Controlling for Potential Confounding Factors

SULIS, ISABELLA;PORCU, MARIANO

2011-01-01

---

Abstract

The aim of this research is to advance a statistical methodology suitable for building up adjusted indicators of efficiency of the degree programs taking into account of the internal sources of heterogeneity across the objects under comparison (for instance the heterogeneity across students’ socio-cultural characteristics). This aim will be pursued throughout the use of direct statistical standardization methods which allow to calibrate some of the indicators of efficiency currently used by the government with respect to the composition/structure of a standard population. Zero-Augmented models will be use to assess the influence of the Potential Confounding Factors on the credits accumulation process. The estimates of coefficient parameters which characterize the influence of individual factors on the value of the efficiency indicators will be used to adjust the outcome measures with respect to a standard population