Abstract
This paper proposes two original measures of socioeconomic polarization, in order to quantify phenomena that are not always taken into account by social inequality measures. Our approach is inspired by the literature on bivariate inequality (the concentration index) and univariate polarization. Like the concentration index, our social polarization measures can be easily computed thanks to a “convenient” regression and decomposed into their determinants. Moreover changes in polarization can also be decomposed into their causes. The paper also provides an empirical illustration of our methods for the probability of reporting excellent or very good health, using cross‐sectional data on French women. The findings suggest that after 65 years of age, social polarization in this probability decreases whereas social inequality remains stable. Consequently social polarization conveys additional information to that contained in the concentration index.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call