Examining Substance Abuse Data Collection Methodologies

INTRODUCTION The U.S. Department of Health and Human Services produces three major sources of information about substance abuse by high school-aged youth. These sources are the Youth Risk Behavior Surveillance System (YRBS) conducted by the Centers for Disease Control and Prevention, the Monitoring the Future studies (MTF) conducted by the National Institutes on Drug Abuse, and the National Household Survey on Drug Abuse (NHSDA) conducted by the Substance Abuse and Mental Health Services Administration (SAMHSA). Comparisons of the results indicate, as will be seen below, that there are substantial differences in the reported results of these three surveys. In this paper, I speculate about the possible reasons for these differences. The discussion is not based on any new experiments or analysis, but on previous research on response effects. In general, this research indicates that survey respondents tend to under-report iUegal or socially undesirable behavior. The differences in the survey-based estimates are worth studying for two major reasons. First, the surveys all provide crucial information needed by health policy makers for the development and monitoring ofprograms to reduce youth substance abuse. Significantly different results from surveys are certainly confusing to policy makers. Second, because these three sources all use somewhat different methods to ask what are likely perceived by respondents as being highly sensitive questions, the comparison ofthe different surveys provides a natural experiment on method effects of threatening questions. Thus, a comparative analysis may contribute new insights into the literature on response effects in surveys.