The Validity of Self-Reported Data on Drug Use

Abstract

Surveys of drug use are continually criticized on the premise that respondents underreport the extent of their drug use. Validation studies conducted prior to the mid-1980s involving known samples of drug users or urinalysis techniques showed that drug use was fairly accurately reported in self-report surveys. However, more recent validation studies conducted with criminal justice clients using improved urinalysis techniques suggest less concordance between urinalysis and self-report. This paper reviews these studies and their implications for the validity of self-report in epidemiological drug surveys. Some general conclusions can be drawn from various validation studies. Valid self-reporting of drug use is a function of: 1) the recency of the event, 2) the social desirability of the drug, and 3) nuances of data collection methodology. The paper discusses methods used to improve the validity and quality of self-report data on drug use.

Partner naming and forgetting: Recall of network members

“…For drug use, underreporting is more severe depending on the stigmatization of the drug. Marijuana use appears to be reported more validly than cocaine use in the general population (Harrison 1995) and in arrestee populations (Fendrich & Xu 1994;Mieczkowski et al 1991).…”

Price and purity analysis for illicit drug: Data and conceptual issues

“…There has been a great deal of research on the validity of self-reported drug use (e.g., Harrison, 1995), but less on the validity of self-reported price data.…”

Detection and Prevalence of Substance Use Among Juvenile Detainees

“…Studies of detained adults (Golub et al, 2002;Gray and Wish, 1999;Harrison, 1995;Mieczkowski, 1990) have found self-report veracity levels for cannabis and cocaine that are similar to those found for juveniles. As with juveniles, self-reports of use decline when subjects are asked about more recent use, and veracity levels are markedly lower for cocaine than for cannabis.…”

“…◆ Self-report veracity is related to the type of drug. Cannabis use is more likely to be reported than cocaine use, probably because the former is more socially acceptable (Golub et al, 2002;Harrison, 1995;Mieczkowski et al, 1991;Fendrich and Xu, 1994) and because it carries less severe legal consequences. …”