School-level inequality measurement based categorical data: a novel approach applied to PISA

Abstract This paper introduces a new method to measure school-level inequality based on Item Response Theory (IRT) models. Categorical data collected by large-scale assessments poses diverse methodological challenges hinder measuring due truncation and asymmetric intervals between categories. I use family possessions from PISA 2015 exemplify the process of computing measurement develop set country-level mixed-effects linear regression models comparing predictive performance novel with Gini coefficients. find is negatively associated learning outcomes across many non-European countries.