Assessing the quality of institutions’ rankings obtained through multilevel linear regression models
Year: 2009
Type: report
Abstract: The aim of this paper is to assess the quality of the ranking of institutions obtained with\nmultilevel techniques in presence of different model misspecifications and data structures.\nThrough a Monte Carlo simulation study, we find that it is quite hard to obtain a reliable\nranking of the whole effectiveness distribution while, under various experimental conditions,\nit is possible to identify institutions with extreme performances. Ranking quality increases\nwith increasing intra class correlation coefficient and/or overall sample size. Furthermore,\nmultilevel models where the between and within cluster components of first-level covariates\nare distinguished perform significantly better than both multilevel models where the two\neffects are set to be equal and the fixed effect models.
Authors: Bruno Arpino, Roberta Varriale
Institutions: Bocconi University, University of Florence
Language: English
Subfield: Statistics and Probability
Field: Mathematics
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W2171632615
