NLS user since 1980
Much of my work using various NLS data sets has been concerned with racial disparities. My more recent papers establish the conditions under which we can make meaningful statements about the evolution of the test-score gap in elementary and middle school. One paper shows that this is sensitive to scaling, while the more recent approach argues for a particular approach to scaling and reestablishes that the gap is large and constant from kindergarten through grade 7. My very early work suggested that conditional on background characteristics, blacks and whites made similar decisions regarding when to end schooling, while my more recent work extends the analysis and establishes that conditional on a number of possible confounding factors, blacks complete more education than otherwise similar whites. I also have quite separate papers that find that the adverse effects of growing up with one parent absent have been greatly exaggerated due to inadequate controls for other differences, and a paper that uses the NLSY79 to show that the pattern of constraints on the choice of working hours is inconsistent with Lazear's agency model.
The longitudinal information combined with the breadth of the questions asked makes the NLS suitable for a wide range of investigations. The ability to span generations through the CNLSY has been particularly useful for some of my recent research.