Volume 33, Number 2 (Spring) 1998: Attrition in Longitudinal Surveys

MaCurdy, Thomas, Mroz, Thomas, and R. Mark Gritz. 1998. "An Evaluation of the National Longitudinal Survey of Youth." Journal of Human Resources 33(2):345-436.

The National Longitudinal Survey of Youth (NLSY) has become one of the most widely used data sources for investigating many of the economic and demographic circumstances faced by young adults during the 1980s. The usefulness of the NLSY for empirical analyses relies on the presumption that these data are representative of the population of U.S. youths throughout the 1980s. As the NLSY approaches its third decade, researchers may be concerned about the ongoing representativeness of the NLSY due to the possibility of nonrandom attrition. Also, in contrast to most other data sources, sample members attriting from the NLSY are often recruited back into the sample at a later date, and this "returning" group may also be nonrandom, possibly either offsetting or exacerbating the effects of attrition.

There has been surprisingly little attrition from the NLSY, compared with other longitudinal data sets that have been plagued with such problems. Yet concerns about nonrandomness of attrition in the NLSY may still be justified, since there has little research to date on this issue.

Thomas MaCurdy is a Professor of economics and Senior Fellow at the Hoover Institution, Stanford University. Thomas Mroz is a professor of economics at the University of North Carolina, Chapel Hill. Mark Gritz is a research scientist at Battelle Meynorial Institute in Seattle. The National Longitudinal Survey on Youth was funded by the U.S. Department of Labor Bureau of Labor Statistics. Opinions stated in this document do not necessarily represent the official position or policy of the U.S. Department of Labor. The authors gratefully acknowledge many useful comments from Ed Johnson and coments and expert research assistance from Bret Dickey, Hoon Soh, Ya-Chen Shih, and Chris Timmins.