Type of Document Thesis Author Ragan, Daniel T. Author's Email Address email@example.com URN etd-07152009-035659 Title The Role of Selection Effects in the Drug-Crime Relationship: A Propensity Score Matching Approach Degree Master of Science Department Criminology and Criminal Justice, School of Advisory Committee
Advisor Name Title Kevin M. Beaver Committee Chair Bruce Bullington Committee Member Daniel P. Mears Committee Member Keywords
- Substance Use
- Propensity Score Matching
- Selection Effects
Date of Defense 2009-06-08 Availability unrestricted AbstractResearch within the criminological literature has consistently found an association between drug use and crime. Despite this strong empirical association, however, debate still remains regarding the mechanisms responsible for the relationship between these behaviors. An obstacle to the study of this association has been that observational data are highly susceptible to confounding factors. Thus, relationships hypothesized to be causal based on observational data may in fact be spurious.
One type of confounding factor that is especially important to understanding the drug–crime relationship is that of selection effects. Because substance use is not a randomly assigned event, it is likely that substance users and non-users differ on a number of important factors. These factors likely influence not only the likelihood of using drugs, but of participating in criminal and delinquent behavior as well. Those who make the decision to use drugs, then, may be more likely to engage in other types of antisocial behavior as well, regardless of whether substance use occurs.
Using data from the National Longitudinal Study of Adolescent Health (Add Health), this thesis employs propensity score matching to explore the role of selection effects in the relationship between drug use and crime. Possible explanations for, and past research addressing, the drug–crime relationship will be reviewed. Following this, the sample, measures, and statistical analyses employed within the thesis will be introduced, and the results of the propensity score analyses will be explained. Finally, the conclusion of this thesis discusses key findings, the implications of these findings, and the study’s limitations.
Filename Size Approximate Download Time (Hours:Minutes:Seconds)
28.8 Modem 56K Modem ISDN (64 Kb) ISDN (128 Kb) Higher-speed Access RaganDThesis2009.pdf 536.73 Kb 00:02:29 00:01:16 00:01:07 00:00:33 00:00:02