Activity Number:
|
318
- Analyzing Government Data with Missing Item Values: A WSS Invited Session
|
Type:
|
Invited
|
Date/Time:
|
Tuesday, August 1, 2017 : 10:30 AM to 12:20 PM
|
Sponsor:
|
Host Chapter
|
Abstract #322122
|
|
Title:
|
Recommended Methods for Handling Missing Item Values in Regression Analyzes of the National Survey on Drug Use and Health
|
Author(s):
|
Peter Frechtel* and Phillip Kott and Rachel Harter and Susan Edwards and Jeff Laufenberg and Stephen Tueller and Jiantong Wang and Sarra Hedden and Jonaki Bose and Rebecca Ahrnsbrak and Rachel Lipari and Matthew Williams
|
Companies:
|
RTI International and RTI and RTI International and RTI International and RTI International and RTI International and RTI International and Substance Abuse and Mental Health Services Administration and Substance Abuse and Mental Health Services Administration and Substance Abuse and Mental Health Services Administration and Substance Abuse and Mental Health Services Administration and Substance Abuse and Mental Health Services Administration
|
Keywords:
|
item nonresponse ;
missing data ;
imputation ;
regression ;
National Survey on Drug Use and Health
|
Abstract:
|
The National Survey on Drug Use and Health (NSDUH) is sponsored by the Substance Abuse and Mental Health Services Administration (SAMHSA) of the U.S. Department of Health and Human Services (DHHS) and has been conducted by RTI International since 1988. It provides data on the use of tobacco, alcohol, illicit drugs (including non-medical use of prescription drugs) and mental health in the U.S. at both the national and state-level. Analysts often fit regression and other models to data from this complex survey. We provide a guide to analysts interested in fitting regression models using data from the NSDUH by providing them with scientifically defensible methods for handling missing item values in regression analyses (MIVRA). To this end, a simulation experiment was performed that evaluated several MIVRA methods using NSDUH data. Combining the results of a literature review and the simulation experiment, we offer advice to analysts on which methods are best to use in which situations.
|
Authors who are presenting talks have a * after their name.