Association for Public Policy Analysis & Management

Service members face significant challenges transitioning to and succeeding in the civilian workforce. Multiple federal agencies, including the Department of Defense, Department of Labor, Department of Health and Human Services, and the Department of Veterans Affairs, collect data to track service members through the transition process. Despite this wealth of data, it can be difficult to form a complete picture of transition assistance efforts and their impact on civilian employment outcomes because these databases are compiled and maintained by separate federal agencies.

The Department of Labor tasked ICF with analyzing transition assistance participation and employment data to better understand the impact of the Transition Assistance Program on employment outcomes for veterans. ICF is working with the West Point Office of Economic and Manpower Analysis and the Department of Health and Human Services’ National Directory of New Hires to compile a data set that includes TAP participation and employment information after separation for all U.S. Army enlisted soldier separations from 2013 through 2015 (over 300,000 individual records). This unique data set will provide an unprecedented level of precision for the analysis, which will utilize Propensity Score Matching to compare the impact of the re-designed Transition GPS TAP curriculum to the prior “Legacy” TAP curriculum on employment and wage outcomes.

This presentation will consist of three parts. First, we will describe, based on experience from this evaluation, the process to obtain access to federal administrative data through the Office of Economic and Manpower Analysis and the National Directory of New Hires. This description will include advice and recommendations related to developing a common understanding between researchers and the federal agencies who maintain these databases, with particular attention to issues of data security, data elements and coding, codebooks and data descriptions, data formats, and data transfer procedures. Second, we will discuss statistical methods and the approach to the analysis, which is designed to take advantage of this large data set to draw precise comparisons over time and track the impact of curriculum change on employment outcomes after separation. Third, the presentation will conclude with a recap of recommendations for using federal administrative data to track employment and training outcomes, as well as observations about the future use of federal administrative data for improving employment and training programs.