Shulin Shen

graduate student

Department of Economics, Syracuse University

NLS user since 2015

What I learned from NLS data

By constructing a panel data set of four years observations from NLSY, I learned that there is no significant gender differential in returns to schooling after controlling for correlation of time-variant and time-invariance explanatory variables with the unobserved individual fixed effects.

Why I chose NLS data

The panel data format of NLS is the reason that I chose to use it. For the study of returns to education, it's a well-know problem that unobserved ability could confound the effect of education on earning. With the panel data which follow individuals across years, one can control for the unobserved individual fixed effect easily and thus get more reliable results.