A hierarchical linear model for estimating gender-based earnings differentials

In this article, the authors demonstrate how a multilevel model (the hierarchical linear model, HLM) can be applied to the study of gender inequality in the labor market and help us to better understand its sources. The authors estimate a two-level model containing individual- and occupation-level variables to examine the effect of gender on earnings inequality. The results are compared to those derived from the conventional ordinary least squares (OLS) model. The comparison reveals that the estimated gender effect on earnings is robust regardless of the model by which it is derived. Gender exerts a significant impact on earnings both in the OLS as well as in the HLM equations. The HLM method, however, enables us to estimate the extent to which the rate of gender-based earnings inequality is dependent on occupational characteristics.