In recent years, many police departments have mandated or encouraged anti-bias training. This has occurred in response to government-imposed measures such as consent decrees or as a proactive attempt to enhance public perceptions of police following actions that have raised concerns about racially motivated and other discriminatory practices.
In a new study, researchers evaluated the effectiveness of an anti-bias training intervention in one Californian jurisdiction. The study found that officers who received the training had improved performance scores (measured by body-worn camera footage), decreased disparity in how they treated different groups of people, and fewer discrimination-based complaints. Results were small but significant.
The study, by researchers at Washington State University (WSU) and RTI International, is published in Policing: An International Journal.
Through anti-bias training programs, police departments seek to promote fairness in police decision-making, enhance public perceptions of police, and ultimately improve interactions between police and community members. Many programs target bias, with two methods used most frequently: implicit bias training (classroom-based training with lectures on the science of bias, strategies to recognise bias, and scenarios for practising strategies) and counter-bias training (simulation-based training in which officers respond to scenarios to facilitate their recognition of biases in behaviour and practice unbiased decision-making).
“Despite concerns around bias and departmental commitments to anti-bias training, there is limited evidence on the impact of the trainings on officers’ behaviour and decision making in interactions with the community,” explained Lois James, Assistant Dean of Research at WSU’s College of Nursing, who led the study. James is an expert whose work is promoted by the NCJA Crime and Justice Research Alliance, which is funded by the National Criminal Justice Association.
Between 2019 and 2021, James and her colleagues evaluated the effects of an anti-bias intervention that combined implicit and counter-bias training on two distinct outcomes: officers’ performance as measured by coded body-worn camera (BWC) footage of encounters with community members and discrimination-based citizen complaints issued against the police.
50 patrol officers from the Sacramento Police Department were randomly selected to participate in the anti-bias intervention. Before and after the intervention, researchers coded a random selection of BWC videos from the intervention group and from a control group of officers who did not take part in the training to assess police performance in interactions with the public. They also collected discrimination-based complaints by community members before and after the intervention for both groups.
Officers who participated in the anti-bias training had a small but significant increase (approximately 5%) in performance scores compared to officers who did not take part in the training. In addition, officers who were trained had a reduction (approximately 50%) in discrimination-based complaints compared to officers who were not trained.
The improvements in officers’ performance scores appeared to be driven by reductions in disparities in how officers treated community members who were suffering homelessness. The study did not find disparities in how officers interacted with community members based on race, but disparities were observed for gender, with officers receiving lower performance scores in interactions with men than with women.
“Although our results are from a single municipal police department, ours is the first study to suggest that anti-bias training could have a positive behavioural impact on officers’ behaviours during interactions with the public and on public perceptions of biased treatment by officers,” noted Stephen James, Assistant Professor in WSU’s Elson S. Floyd College of Medicine, who co-authored the study. “As such, this is a tentatively encouraging step forward in demonstrating the effectiveness of anti-bias training.”
“However, our findings should be interpreted with caution and do not necessarily mean that anti-bias training will be successful in improving police interactions with the public,” added Renée Jean Mitchell, who was a senior research scientist at RTI when she co-authored the study. “Other factors could contribute to these findings and so we encourage more research on this important issue.”
Moreover, the authors caution that the combined effects of the Covid-19 pandemic and the civil unrest of 2020 might have influenced results in ways that are difficult to account for, even with a no-training control group. The way police in the jurisdiction studied dealt with these factors might have differed from how other departments dealt with them, so the findings might not generalise to other jurisdictions.