The Cumulative Distribution Function (CDF) Adjustment Method Applied to Small-Area Estimates of Serious Psychological Distress in the National Survey on Drug Use and Health (NSDUH)

• Activity Number: 70
• Type: Contributed
• Date/Time: Sunday, July 29, 2007 : 4:00 PM to 5:50 PM
• Sponsor: Section on Survey Research Methods
• Author(s): Jeremy Aldworth*+ and Neeraja Sathe and Misty Foster
• Companies: RTI International and RTI International and RTI International
• Keywords: Design-based ; Small area estimation ; Respondent-level adjustments

Abstract

The NSDUH is an annual national survey with a sample size of approximately 67,500, and it also provides state-level estimates using small area estimation (SAE) methods. Serious psychological distress (SPD) is measured by the K6 scale in the NSDUH. In 2003, a broad array of mental health questions preceded the K6 items. In 2005, only the K6 questions were asked. In 2004, the sample was evenly split between the long and short modules used in 2003 and 2005, respectively. Results from 2004 showed large differences in SPD prevalence rates between the two modules. Since NSDUH SAE methods require two years of data to be combined, a reversible CDF adjustment method was devised so that SPD estimates from one module in 2004 could be adjusted to be similar to those from the other module, and vice versa. This allowed us to combine 2003/2004 and 2004/2005 data to obtain state-level estimates of SPD.