Analysing SET over time using multilevel multidimensional explanatory IRT models

This contribution makes an attempt to analyze students' ratings of university teaching on a broad prospective, trying to adjust the final assessment from a wide range of factors which jointly may influence the process under evaluation: academic year peculiarities, course characteristics, students' characteristics and item dimensionality. From a methodological point of view, by setting complex Item Response models as special case of Generalized Linear or Mixed Models a large flexibility is introduced in the specification of ad hoc modelling approaches for the analysis of students' ratings.