Introduction--the validity of self-reported drug use: improving the accuracy of survey estimates.

Measuring levels and patterns of illicit drug use, their correlates, and related behaviors requires the use of self-report methods. However, the validity of self-reported data on sensitive and highly stigmatized behaviors such as drug use has been questioned. The goal of this monograph is to review current and cutting-edge research on the validity of self-reported drug use and to describe methodological advances designed to reduce total error in estimates of drug use and quantify sources of nonsampling error. This monograph reviews a number of studies that use some presumably more accurate measure of drug use to validate self-reported use. In addition, evolving methods to improve a wide variety of procedures used in survey designs are explored, including computer-assisted interviewing, predictors of response propensity, measurement error models, and improved prevalence estimation techniques. Experimental manipulations of various survey conditions and situational factors also show promise in improving the validity of drug prevalence estimates in self-report surveys.