Our new study on the police shooting of Jacob Blake

We found that the shooting reduced Black Chicago residents' trust in police (and no effect for whites)

Hey folks! A couple weeks ago, we published a new study in PLOS ONE which takes a look at the aftermath of the police shooting of Jacob Blake in August 2020. The study is open access, so feel free to give it a read.

Our main finding is that the police shooting reduced Black Chicago residents’ trust in police. Although previous studies also found that high-profile police violence damages perceptions of police, we found these changes were most pronounced among three groups: Black residents overall, young Black residents, and Black residents who had previously been arrested or charged with a crime. This is particularly notable given that these groups already reported very low levels of trust in police prior to the shooting.

This might sound like a “water is wet” finding, but we were surprised to only observe changes in trust among Black respondents – for example, prior work found big changes in white perceptions of police following the murder of George Floyd in May 2020. For me, it was particularly surprising to not see any movement among e.g. white respondents who had previously been arrested or charged with a crime. Past research found that white perceptions of police changed after George Floyd was killed in May; Jacob Blake was shot in August, so it could be the case that white ‘learning’ about police violence hit a ‘ceiling.’

Also, in case you’re wondering why the study looks at Chicago rather than Kenosha (where Jacob Blake was shot) – the study uses survey data from Chicago, so that’s what we’ve got to look at. (There were lots of protests regarding the shooting in Chicago as well, it’s kind of like one big extended metro area.) We identify a causal effect through an “unexpected event during survey design”: Blake’s shooting occurred at the midpoint of a health survey which included questions about policing. We compared responses from before and after the shooting (these are treatment and control groups).

Our study may be the first to examine whether perception changes differ between residents who have prior exposure to criminalization and those who don’t. This is significant given my prior research finding that police contact reduces voter turnout.

Past research has also found an association between low levels of police trust and willingness to carry a firearm for self-protection. So, our findings contribute to a scientific understanding of how policing could contribute to America’s high rate of interpersonal violence.

(P.S. I haven’t been posting much because I’ve been trying to chip away at my dissertation. Hope to have more research to share with you all soon, and maybe some blog posts if something comes to mind!)