Abstract.
In the paper we present a comprehensive methodology for the analysis of student ratings of university courses. First, simple descriptive measures, which take into account the ordinal nature of the ratings, are discussed. Then net measures, which adjust for the characteristics of the students, are obtained through multilevel modelling. Finally, the measures relative to the various aspects of the course are synthesized through a weighted mean, building gross or net multidimensional indicators of course quality. The different indicators are then contrasted with respect to the rankings of courses they induce.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Author information
Authors and Affiliations
Corresponding author
Additional information
Correspondende to: Carla Rampichini
The current study has been financed with the MURST (40%) grant “Production and experimentation of computer-assisted systems to survey the quality of teaching at the university level and the success of university graduates in the job market” and with the MURST (40%) grant “Evaluation of quality and effectiveness in human services, with particular attention to health and education”.
Rights and permissions
About this article
Cite this article
Rampichini, C., Grilli, L. & Petrucci, A. Analysis of university course evaluations: from descriptive measures to multilevel models. Statistical Methods & Applications 13, 357–373 (2004). https://doi.org/10.1007/s10260-004-0087-1
Issue Date:
DOI: https://doi.org/10.1007/s10260-004-0087-1