Methods for quantifying ordinal variables: a comparative study

摘要: The solution to the problem of ‘quantification’ or scoring, i.e., assigning real numbers qualitative modalities (categories) an ordinal variable, is primary relevance in data analysis. literature offers a wide variety quantification methods, all with their pros and cons. In this work, we present comparison between univariate multivariate approach. approach allows estimate category values variable from observed frequencies on basis distributional assumption. simultaneously transforms set variables into interval scales through process called optimal scaling. As example application, consider Bank Italy coming “Survey Household Income Wealth” order ‘quantify’ self-rating item happiness. A simulation study compare performance two approaches also presented.