To allow analyses of its personnel practices, the Department of Defense maintains historical administrative data files and administers surveys of military personnel. Military manpower analyses also make use of civilian cross-sectional and longitudinal data. Klerman provides an overview of these data sources and discusses how they can be analyzed with currently underutilized data-matching strategies. These data-matching strategies involve matching DoD administrative data files to (1) civilian administrative data (such as Social Security Administration earnings data); (2) DoD survey data; and (3) civilian survey data. These strategies have the potential for large payoffs in terms of better analysis-and therefore better policy-for DoD. Klerman also discusses the degree to which DoD should help fund a future National Longitudinal Study of Youth, and whether DoD should initiate a new military panel survey. Data for DoD Manpower Policy Analysis maintains that the research questions that these proposed surveys would help answer can instead be explored through data matching. Moreover, streamlining procedures for data matching-that is, making it easier for researchers to analyze the data DoD already has-is likely to be much less expensive than engaging in major new data-collection efforts.

"Prepared for the Office of the Secretary of Defense"--T.p.

Introduction -- Military administrative data -- Current military cross-sectional survey programs -- Matching DoD administrative data to DoD cross-sectional surveys -- Matching DoD administrative data to civilian administrative data -- Matching DoD administrative data to civilian cross-sectional surveys -- Civilian panel surveys and choice-based sampling -- An alternative model for military cross-sectional surveys -- A military panel survey -- Discussion -- Appendix A: formal discussion of some technical issues -- Appendix B: U.S. Census Bureau residence rules -- Appendix C: survey veteran questions.