Does concealed handgun carry make campus safer? A panel data analysis of crime on college and university campuses
MetadataShow full item record
The purpose of this report is to recommend and test an empirical strategy for assessing the impact that concealed carry policies have on crime at college and university campuses. I use panel data obtained from the Department of Education for all crimes reported on four-year, undergraduate, federal financial aid-receiving institutions between 2001 and 2014 to model the impact of campus carry legislation. Differences in legislation across states, time, and school types allow for estimation of a triple difference regression model. Results of OLS estimations show that campus carry has no significant observable association with rates of aggravated assault, sexual assault, robbery, burglary, and motor theft committed on campus at the 95% confidence interval. These results are robust to a number of different assumptions, including time lag and negative binomial modeling approaches. However, true effects may be difficult to determine precisely as model estimations present large standard errors. Notably, my analysis does not attempt to control for variables that may also influence campus crime rates, such as local economic conditions, gun ownership rates, or rates of concealed carrying on campus. This analysis is therefore only a starting point for further research and the results contained here should be considered preliminary. At most, my analysis may throw partisan narratives surrounding campus carry into some measure of doubt. In particular, results fail to demonstrate a measurable deterrent effect theorized by campus carry advocates, or a criminal enabling effect theorized by opponents of the policy. Regardless of crime changes, I suggest that policymakers considering this controversial measure should also weigh how concealed carrying policies may influence a variety of other variables, including student suicides – a full understanding of which requires considerable caution and further research.