Returns to schooling in Palestine: a Bayesian approach

Abstract

This paper presents an empirical method to find more efficient estimates of returns to schooling using Bayesian linear regression instead of OLS method. The private returns to schooling in Palestine using the Palestinian labour force survey (PLFS) for the year 2017 have been estimated, where on average, males earn 40.7% more than females. Separate regressions have been performed for males and females, in which the returns to schooling for females are found higher than their males' counterparts. Bayesian inference has also been applied into Heckman two-step procedure with logit and probit models to correct self-selection bias for females' sample. It is found that logit Heckman correction yields positive and higher coefficient of years of schooling than probit and OLS. The wage disparities in Palestine have been found influenced by various factors like age, sex, and occupational groups. These findings are useful for policymakers to plan for future investment in higher education.