A Critical Reanalysis of Maryland State Police Searches

Abstract

This article argues that previous analyses of the Maryland State Police search data may be unreliable, since nonstationarity of these data precludes the use of standard statistical inference techniques. In contrast, proper statistical graphics seem better suited to capture the complexities of the racial bias issue.

KEY WORDS::

• Graphical analysis
• Nonstationarity
• Policing
• Racial profiling
• Stop and search data

Notes

Gross and Barnes (2002) gave evidence that the behavior of the MSP changed over time, but failed to point out that it invalidates standard statistical tests. Hernandez-Murillo and Knowles (2004) pointed out that when “statistics conflate searches involving different levels of police discretion, standard tests for racial bias are not applicable.”

We do, however, consider “Hispanics” and “Others” in a graphical analysis of the largest drug seizures.

KPT did disaggregate the data into three subperiods, and did conclude that guilt rates were statistically equal within each subperiod. However, our graphic disaggregates the data to an even finer level (monthly).

An “obvious” grounds for search may not always lead to guilt and arrest.

The absolute amount cannot be inferred from the vertical lines in the graph, but what is important for our discussion is the fluctuation of the relative amounts over time. Note though that the absolute amount is shown in the “median cds amounts recovered” panel of Figure 1.

Gross and Barnes (2002) used regression techniques to analyze these data.