NCJ Number
191738
Journal
Journal of Drug Issues Volume: 31 Issue: 3 Dated: Summer 2001 Pages: 695-716
Date Published
2001
Length
22 pages
Annotation
This article details the substantial differences in the reported results of three national surveys.
Abstract
The surveys are the Youth Risk Behavior Surveillance System (YRBS), the Monitoring the Future studies (MTF), and the National Household Survey on Drug Abuse (NHSDA). The differences in survey-based estimates are worth studying because the surveys provide crucial information needed by health policy makers for the development and monitoring of programs to reduce youth substance abuse. Also, because these three sources all use different methods to ask what are perceived by respondents as sensitive questions, the comparison of the different surveys provides a natural experiment on method effects of threatening questions. There are important differences in the administration, anonymity, sample design, weighting, and results of the three surveys. This study concentrated on the differences in methods of administration, questionnaire context in sample design and execution and weighting. Although the differences in sample design, execution and weighting are significant, the most likely effect of these differences would be to increase the reported levels of drug use in the household surveys relative to the school surveys. Weighting methods could be one reason why the two school-based studies differ. Two aspects of the questionnaires might cause differences among the surveys: the specific wording on questions about drug use, and the contexts and introductions to the questionnaires. The most important source of differences among the surveys are the methods of administration. The different methods of administration are critical because they relate to respondents’ perceptions of anonymity. The household interview created the least sense of perceived anonymity as compared to the other two group-administered surveys. 2 notes, 4 tables, 19 references