Year of Publication

2012

College

Martin School of Public Policy and Administration

Degree Name

Master of Public Administration

Executive Summary

On the local news channels in major college football towns, there are anecdotal stories during that detail celebratory riots that took place during or after a college football game. Few empirical studies have focused on whether there is a relationship between college football games and crime. This paper attempts to determine that relationship by exploiting the fact that college football games are played in a home stadium and an away stadium.

More specifically, the study addressed the following two questions:

Do jurisdictions in which a home football game is played differ in crime rate from those where an away game is played?

Do crime rates in the same jurisdiction vary depending on whether a home or away game is played there?

Data for the analysis was collected for the 2007 NCAA Division I football season. Crime data for the analysis was obtained from the 2007 National Incident Based Reporting System; sixty jurisdictions having college football teams reported data. The data set included jurisdiction demographic information from the 2007 American Community Survey. Various other data sources were used to collect information specific to the college football game, such as rivalry identification, point spread, conference game, game time and outcome of the game. The analysis of data included t-test of difference in means for significance of home versus away games and other t-tests related to the game outcome of home games. The analysis included two regression models, a standard OLS and a fixed effects regression.

The regression results found that home games had a significant and positive impact on total crime, compared to away games (i.e. home games increase crime rates). Homes games had a significant and positive impact when both models were run with disorderly conduct, drunkenness and liquor law violation as the dependent variable crime category. In the fixed effects model, the relationship between home games and DUI crime was significant and positive although the magnitude was not meaningful. The game characteristics or outcome of the game had little impact on crime.

Recommendations are made for future research to increase the sample size by analyzing multiple years. Further research is needed to analyze the police force and policing practices during college football games. Recommending increasing the number of police on duty during college football games could have either a positive or negative impact on the crime rate. More police could either reduce the number of crimes as they are more visible to spectators or increase crime as police can cover a larger area.

Share

COinS