Document Type



Master of Arts


Criminology and Criminal Justice

Date of Defense


Graduate Advisor

Richard Rosenfeld, Ph.D.


Lee Ann Slocum

Richard Wright


Determining the accuracy of self-reported drug use is important for criminal justice professionals so that they are better able to provide proper treatment referrals to those in the criminal justice system who may need substance abuse help (Rosay et al., 2007). However, self-reports, especially those of drug users, are not always accurate (Harrison, 1997). Drug use is a highly sensitive topic and disclosure of such behavior could lead to negative repercussions for the individual within the criminal justice system as well as lead to further stigmatization of the individual outside the system (Golub et al., 2002; Harrell, 1997). The current study uses data from the 2003 Arrestee Drug Abuse Monitoring (ADAM) survey to examine the accuracy of self-reported drug use across seven different types of drugs to determine if the anticipated strain of admitting to the use of drugs, compounded by respondents’ current levels of strain, are strong enough to inhibit individuals from accurately reporting drug use. Binomial conditional logistic regression models with fixed effects and robust standard errors were used to conclude that experiencing strain reduces the likelihood of accurately reporting drug use. The current study expands the current literature on Agnew’s general strain theory to include purposeful deception as a deviant coping mechanism used in response to strain. The results of the current study may help criminal justice professionals more accurately identify active substance abusers who may be less than truthful about their drug use. Implications from this study suggest that it may be useful to incorporate strain-related variables into the risk and needs assessment measures that criminal justice professionals use to better guide treatment referrals.

OCLC Number