Evaluation of Small-Area Estimates of Substance Use in the National Survey on Drug Use and Health

• Activity Number: 414
• Type: Contributed
• Date/Time: Tuesday, July 31, 2012 : 2:00 PM to 3:50 PM
• Sponsor: Section on Survey Research Methods
• Author(s): Art Hughes*+ and Akhil K Vaish and Neeraja Sathe and Kathryn Spagnola
• Companies: SAMHSA and RTI International and RTI International and RTI International
• Keywords: substance use ; small area estimation ; validation ; substate estimates

Abstract

Small area estimation (SAE) methods are used to produce state and substate estimates of substance use and mental disorders using data from a major U.S. behavioral health survey. State-level policymakers use these estimates to help justify funding for substance abuse prevention and treatment programs in their jurisdictions. Thus, it is important to determine how the SAEs compare to their design-based counterparts (where design-based estimates are less expensive and can be produced in less time) and if the SAE models can be improved to increase the precision of change over time.

In 2001, a validation study was conducted to judge the quality of the state SAEs and found that they were generally more precise than design-based estimates while exhibiting only small levels of bias (Wright, 2001). We will extend this analysis using more recent data to validate the substate SAEs in a similar f