Disproportionality in Officer-Initiated Contacts with Civilians and Contact Outcomes: A Bivariate and Multivariate Analysis

This study used prospective, officer-collected data from 39,730 officer-initiated contacts, occurring over a two-year period, to answer these questions: (1) What is the relationship between civilians’ race and the occurrence of officer-initiated contacts, and the outcomes of search, arrest, and use-of-force, resulting from officer-initiated contacts? (2) What is the relationship between neighborhood characteristics, contact characteristics, officers’ race, and civilians’ gender and race, and the search, arrest, and use-of-force of officer-initiated contacts?  There were 237 (88.4%) out of 268 commissioned officers who collected data using an electronic instrument that was developed by the researcher, for an officer-level 2.2% margin of error. Additional data were collected from the U.S. Census Bureau American Community Survey, and police department episode and personnel records.  The bi-variate results, which answered research question one, included: (1) African Americans and Native Americans were significantly more likely to experience officer-initiated contacts than members of other racial groups; (2) African Americans and Native Americans were significantly more likely to be searched or arrested, in comparison to other racial groups; (3) The reasons for searches and arrests did not differ significantly between racial groups; (4) Native Americans were also more likely to experience a use-of-force incident than members of other racial groups.  Binomial logistic regression was used to answer research question two, and analyzed predictors of outcomes from officer-initiate contacts. The significant multivariate results included: (1) Contacts occurring in higher index crime neighborhoods are between 32% and 72% more likely to result in searches or arrests; (2) The race of the civilian involved in a contact significantly increases the likelihood of searches and arrests, increasing likelihoods by between 45% and 103% for searches, 49% and 348% for arrests, among African American and Native American civilians, respectively; (3) When the race of the contacting officer is White, the likelihood of searches significantly decreases by between 35% and 41%, the likelihood of arrests decreases by between 28% and 35%, and the likelihood of force being used does not significantly change; (4) When force was used, being male, or Native American, increased the likelihood of force being used by 119% and 169%, respectively, and being arrested increased the likelihood by 379%.  Conducting these analyses revealed that disproportionate officer-initiated contacts with civilians who are minority groups, and the outcomes of these contacts, is a significant occurrence in the community, and is also a complex phenomenon that eludes easy explanation.  In keeping with the conference theme of “reducing racial and economic inequality,” the author will also discuss how these data are being used in conversations between the police department and members of minority communities to move police training practices to more thoroughly address disproportionality and implicit bias in policing, and to inform future research efforts.