School Security Measures and the Prevention of School Crime and Violence

Background and Objectives: In an attempt to prevent crime and violence in schools, federal policies have provided funding for security measures such as security cameras, metal detectors, and law enforcement officers. Yet there is a noticeable dearth of rigorous empirical research examining the effects of those security measures on school safety. This study used rigorous quasi-experimental research methods to examine the patterns of security measure utilization in U.S. schools, and how those security measures are related to school crime and violence.

Theoretical Framework: Theories of deterrence and criminalization offer competing perspectives in terms of the effects of security measures on student outcomes. Deterrence theories suppose school security measures will have beneficial effects by deterring criminal and delinquent behavior, whereas theories of criminalization posit that they may produce detrimental effects in certain environments. Given these competing theoretical perspectives, we have taken an exploratory approach to examining the ability of school security measures to prevent school crime and violence.

Methods: We used cross-sectional data from multiple waves of two large nationally representative surveys – the School Crime Supplement (SCS) to the National Crime Victimization Survey and the School Survey on Crime & Safety (SSOCS). The SCS analytic sample included over 40,000 student responses collected at six time-points (2001, 2003, 2005, 2007, 2009, 2011). The SSOCS analytic sample included over 6,000 school administrator responses collected at four time points (2003-04, 2005-06, 2007-08, 2009-10). After identifying several common patterns of security measure utilization, we estimated generalized propensity scores as the predicted probability of respondents’ observed school security utilization pattern based on a multinomial logistic regression model that included a wide range of potentially confounding characteristics. We then used generalized linear models controlling for these propensity scores to estimate the effects of school security measures on measures of crime and violence (exposure to drugs, fighting, firearms, and property crime).

 Results: Results indicated that many utilization patterns of security measures were associated with increased crime and violence (particularly in terms of exposure to drugs, fighting, and firearms), although findings varied across student and school administrator reports. Utilization of multiple security measures, however, reduced the likelihood of exposure to property crime in high schools.

 Significance: Given the high levels of funding for school safety initiatives, understanding how school security measures are associated with crime and violence is a critical endeavor. Our findings suggest that some ongoing policies and funding mechanisms geared towards preventing crime and violence in schools may not be working as well as intended. We will discuss implications for the implementation and use of school security measures relative to school safety.