Examining Prevalence Differences in Three National Surveys of Youth: Impact of Consent Procedures, Mode, and Editing Rules

• NCJ Number: 191735
• Journal: Journal of Drug Issues Volume: 31 Issue: 3 Dated: Summer 2001 Pages: 615-642
• Author(s): Michael Fendrich; Timothy P. Johnson
• Date Published: 2001
• Length: 28 pages
• Publication Format: Article
• Publication Type: Report (Study/Research)
• Language: English
• Country: United States of America

Annotation

This article compares the impact of mode of administration, consent, and editing procedures on substance use prevalence estimates derived from three studies.

Abstract

The three studies are Monitoring the Future (MTF), the Youth Risk Behavior Survey (YRBS), and the National Household Survey on Drug Abuse (NHSDA). Two aspects that clearly vary across the studies are consent procedures and survey mode. The potential impact that editing procedures may have on prevalence estimates is studied. The weight of the evidence from the review of the three main areas investigated suggests a potentially important role for consent and mode effects in depressing prevalence rates for NHSDA relative to the other studies. Two main variables can be used to summarize potential consent differences across the studies and potential effects on prevalence. The first is the level of parental involvement, and the second is the overall volume of consent-related information provided to subjects recruited in the study. The available documentation suggests that by far, NHSDA invests the greatest amount of time and resources on data editing procedures. In contrast, the approach to editing taken by the two school-based studies would most likely increase the chances that drug users are excluded from prevalence calculations. Between-study differences extend beyond the numerous administrative procedures to include variation in sampling, question wording, and questionnaire format. There are three main directions for future research. First, epidemiological research on youth drug abuse would strongly benefit from a well-designed study of school vs. household effects employing random assignments of subjects, standardized instruments, and standard editing procedures. Second, there is a continued need to provide more information about the validity of school-based drug surveys, through the implementation of school-based methodological studies. Lastly, extended secondary analysis is recommended, employing multivariate methods on each of the data sets. 5 tables, 52 references