A reexamination of marital status and the earnings of men

TR Number



Journal Title

Journal ISSN

Volume Title


Virginia Tech


Within the field of labor economics, there has been a lot of empirical work which has shown that married men earn more than unmarried men. Theoretically, this makes some sense in that men which earn more money than other men may be more attractive partners for marriage. Also, married men may work harder and longer in order to increase their earnings to support their family. Empirical evidence has shown that marital status is a statistically significant variable when performing a simple regression on earnings using Ordinary Least Squares (OLS). However, OLS can produce biased estimates if one or more of the independent variables are determined endogenously, which is the contention made by Robert A. Nakosteen and Michael A. Zimmer. They found that after adjusting for the endogenous nature of the marital status dummy variable the variable was no longer significant in explaining the variation in earnings. However, two of their shortcomings were a narrowly defined age group for the data set and less than richly specified regression equations. But, using a different data set and additional explanatory variables in order to address the problems of the Nakosteen and Zimmer research I was able to show that even when adjusting for the endogenous nature of the marital status variable its effect on earnings was still significant



earnings, regression analysis, marital status