Statistical Analysis of Police Arrest Decisions

Statistical Analysis of Police Arrest Decisions

What are the central variables that Smith and Visher studied?

Smith and Visher (1981) analyzed a series of independent variables that were found by previous studies to influence dependent variable 'likelihood of being arrested'. Although a Chi square test revealed several factors to be significantly associated with likelihood of being arrested (Table 1), the aim of their research was to reveal the relative contributions of the many independent variables examined. Towards this goal the authors used a multivariate probit model to gauge relative contributions from a set of independent variables on arrest likelihood. These variables are as follows: police contact initiated by a victim or the police, contact with suspect took place in a public or private location, presence of bystanders, victim is present or absent, victim's choice of how to handle the contact, suspect's race, age, and gender, suspect's demeanor, relationship between victim and suspect (known or stranger), and the seriousness of the crime.

How did Smith and Visher use and present their descriptive statistics?

The descriptive statistics were presented in a table format, and rather than provide probability values as is the common practice today the cumulative density functions (B) and raw Z scores were given (Tables 3 and 4). These values allow the reader to gauge the relative predicted contributions of the different variables on arrest likelihood.

The statistics were then used to evaluate the findings of prior studies, discuss the relative importance of the different variables on arrest likelihood, and allow the authors to comment on possible interrelationships between the independent variables.

How did the authors use statistical analysis to present their findings about police arrest decisions?

The authors took the position that a large number of variables likely contribute to whether an arrest takes place and that some of these variables probably interact in a dependent manner. The statistics were used to test this theory and the resultant data used to launch into an informed and authoritative discussion of the results. By taking this approach the authors were able to discuss and convey the significant degree of complexity involved in determining whether an arrest takes place, and this complexity of variables and interrelationships were depicted graphically in Figure 1 (Causal Model).

How did the authors use the idea of statistical significance in their study?