Challenges in Collecting Military Family Data on a Nationally Representative Survey

• Activity Number: 238 - Instrumentation and Data Quality
• Type: Contributed
• Date/Time: Monday, July 31, 2017 : 2:00 PM to 3:50 PM
• Sponsor: Survey Research Methods Section
• Author(s): Struther Van Horn* and Rachel Lipari
• Companies: Kent State University and Substance Abuse and Mental Health Services Administration
• Keywords: Survey ; Instrumentation ; Military

Abstract

This presentation will summarize efforts to expand military family data by leveraging an existing large national survey of the noninstitutionalized civilian population. Other national surveys could model this effort; however, addressing the data quality concerns is required. In 2015, the National Survey on Drug Use and Health (NSDUH) added questions to identify whether respondents had immediate family members who were currently serving in the U.S. military, regardless of where those family members were currently living. Although the military questions were pretested, these items had about 9% missing data. The NSDUH military population estimates were compared to Department of Defense personnel record population data to assess coverage. Based on this comparison, the 2015 data were analyzed while the 2016 NSDUH questions were redesigned to reduce missing data. Changes were made to the military questions to address the weaknesses in the questions regarding the definition of military and the definition of immediate family.