Learning from experience: Estimating teen use of alcohol, cigarettes, and marijuana from three survey protocols

Estimates of teen substance use from 3 federally funded surveys were examined. Results from the NHSDA, a household survey, and the MTF and YRBS, both school-based surveys, were compared by gender and ethnic group. Trends in reported use from 1993-1997 were also analyzed. Many discrepancies exist among the results from the 3 surveys and are likely due to several factors, including different sample sizes and frames, policies regarding sample substitution, question wording, and data collection protocols. Discussion of these factors underscores the need for more research dedicated to methodology. Routinely investing a portion of survey budgets in studies of the methods that are used and how they affect comparability with other surveys would build a knowledge base that would make comparisons of results easier and would inform design decisions to increase comparability across surveys.