Decisions That Influence Police Officer Use of Deadly Force: A Systematic Review

Background and Purpose: A systematic review was conducted to examine use of deadly force among police officers and the race of victims. The purpose of a systemic review, and therefore this study, is to “sum up the best available research on a specific question” (The Campbell Collaboration, n.d.). A national database does not exist that can be used to analyze the race of victims, whether victims were armed or unarmed and other characteristics associated with victims of deadly force by police officers. Consequently little empirical evidence is available to guide social work practice and policy.

Methods: The Campbell Collaboration guidelines for conducting a systematic review were employed. These methods including developing a priori criteria for including and excluding retrieved studies, a search strategy, and codes for retrieved studies. The inclusion criteria required that retrieved studies use samples of police officers or police recruits; utilize a comparison group of victims that was dichotomous (i.e., Black/white, armed/unarmed); contain quantitative analysis; and be English language publications. These criteria were included in the Boolean search terms. Only electronic databases were searched, and the systematic search produced 632 sources that were identified through a search of 15 electronic databases. Based on titles and abstracts, 615 sources were excluded for not meeting the inclusion criteria and 17 full-text articles were obtained, screened and reviewed. Upon careful examination of the full-text articles it was found that 11 did not meet the study criteria. 

Results: The results of the systematic search yielded 6 studies that met the inclusion criteria.  The retrieved studies reveal that police officers’ decisions to shoot or not shoot, and racial bias, are gleaned from computer-based simulation experiments. Such experiments have been conducted with police officers and recruits, and college student samples (Based on the inclusion criteria samples of college students were excluded from the review). The procedures used among the experiments involved showing the study participants pictures that depicted African-American or white males who were either armed or unarmed, and the pictures were displayed on a laptop computer. Participants were instructed to press a computer key to shoot and a separate key to not shoot. Several findings emerged from these experiments: (1) police officers and recruits made fewer errors and more correct decisions to shoot based on armed images; (2) during initial phases of the experiments officers shot unarmed African-American males more frequently than white males who were not armed; and (3) as the experiments continued officers were no more likely to shoot African-American males than white males.

Implications: The findings suggest the problems examining such a difficult subject. These experiments are weak in external validity. Advocacy practice could be useful for developing and implementing a national database that describes the situational characteristics of victims of police shootings, and drafting policies that require local law enforcement agencies to participate in such data collection since at present participation is voluntary.