Connecticut's novel approach to collecting and analyzing traffic stop data for evidence of disparate treatment is widely considered to be a model of best practice. Here, we provide an overview of Connecticut's framework, detail solutions to the canonical empirical challenges of analyzing traffic stop, and describe a data-driven approach to early intervention. Unlike most jurisdictions that simply produce an annual traffic stop report, Connecticut has developed an ongoing system for identifying and mitigating disparity. Connecticut's framework for identifying significant disparities on an annual basis relies on the so-called "preponderance of evidence" approach. Drawing from the cutting-edge of the empirical social science literature, this approach applies several, as opposed to a single, rigorous empirical test of disparity. For departments identified as having a disparity, Connecticut has developed a process for intervening on an annual basis. In that process, policing administrators engage with researchers to conduct an empirical exploration into possible contributing factors and enforcement policies. In Connecticut, this approach has transformed what had once been a war of anecdotes into a constructive data-driven conversation about policy. Variants of the Connecticut Model have recently been adopted by the State of Rhode Island, Oregon, and This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.

This paper uses state police stop data in Texas to assess patrol activity. We find that both the types of stops and allocation of resources over space change in darkness relative to daylight, and that the changes in stop type and manpower allocation are correlated within police officers. We also find that the counties receiving more police resources in darkness have a higher share of minority residents. Veil of Darkness (VOD) tests of racial discrimination in traffic stops require that the distribution of motorists be independent of darkness, which is unlikely to be the case without detailed geographic controls.

No abstract

No abstract

African-American motorists may adjust their driving in response to increased scrutiny by police. In daylight, when their race is more easily observable, minority motorists are the only group less likely to have fatal motor vehicle accidents. In Massachusetts and Tennessee, we find that African-Americans are the only group of stopped motorists with slower speeds in daylight.Consistent with an illustrative model, these speed shifts are concentrated at higher percentiles of the distribution. Calibration of this model indicates this behavior creates substantial bias in conventional tests of discrimination that rely on changes in the odds a stopped motorist is a minority.