Title: Comparative Analysis of Ordinal Social Indicators: The Latent Copula Approach

Comparative social indicator research often involves analyzing discrete ordered data. This type of data presents challenges when trying to identify statistical relationships in a multivariate context without any contamination from univariate marginal distributions. The problem is widespread. For example, in comparative social mobility research using categorical data, this issue is known as the problem of margin-free association. This paper discusses the limitations of current approaches and introduces the Latent Copula (LACOP) approach to address the demand for margin-free measures of association. The LACOP approach enables new analytic questions to be addressed while delivering quantitative results that are comparable across different time periods, birth cohorts, social groups, or societies. We discuss software implementations and extensions to multi-level modeling and present numerical examples to illustrate the methodological framework.