Public support for law enforcement agencies’ recording of arrest statistics by race: a research note

Abstract

This paper discusses research pertaining to law enforcement agencies' recording of crime statistics by race. Using data from a 2010 statewide poll in Pennsylvania, we examined whether there was support for the current policy that requires police to record the race of arrestees. In addition, the poll asked Pennsylvanians (N = 758) whether they believe that recording race and crime statistics promotes racial stereotypes. The study also examined the influence of several covariates (gender, age, race, and education) on the dependent measures. In particular, two hypotheses were tested; first, whether Whites would be more likely to support the recording of race statistics than non-Whites, and second, whether non-Whites would be more likely than Whites to believe that the recording of race statistics promotes racial stereotypes. The study found support for both hypotheses. The implications of these results are discussed.

Keywords:

• race and crime statistics
• public opinion
• law enforcement
• arrest statistics
• Uniform Crime Reports

Acknowledgements

The authors thank Dr. Auden Thomas, Director, Center for Survey Research at Penn State Harrisburg, for including the race and crime statistics questions on the 2010 Penn State poll.

Notes

1. It is important to note here that in 2003 Californians considered Proposition 54 (also referred to as the Racial Privacy Initiative). The proposition sought to amend the California Constitution by banning state and local governments from using racial classification. It was considered a first step toward a colorblind state. The proposition failed 36% (in favor) and 64% (opposed). In raw numbers, those in favor represented more than three million Californians. This represents one of the few times in recent years that Americans considered the removal of race from government statistics. It is ironic, though, that had it passed, among the exceptions were: medical data; law enforcement descriptions; prisoner and undercover assignments; and actions maintaining federal funding (http://vote2003.sos.ca.gov/propositions/2-3-prop-54.html).

2. The COOP1 and COOP2 formulas were also used and they arrived at similar results.

3. We originally included a measure of income in this analysis. Our results showed that income and education were highly correlated. The correlation suggested that there may be a coding problem. After further inspection, there was not a coding problem. In these data, those respondents with more education were more likely to have higher incomes. We performed the same analyses presented in the paper using income and not education and arrived at substantively the same results.

4. The specific information for the test of parallel lines may be obtained from the second author on request.

5. We thank an anonymous reviewer for sharing some ideas concerning this finding.