Table 1.
Demographic characteristics of Black and white survey respondents in the included and excluded samples.
Fig 1.
Trust in law enforcement among the Black and white respondents included in the analyses.
Data are from the 2020 Healthy Chicago Survey. The sample includes 584 Black respondents and 939 white respondents interviewed in the four weeks prior and four weeks after the shooting of Jacob Blake. The survey question underlying the graphs is “To what extent do you trust your law enforcement agency?”.
Fig 2.
City-wide share of survey participants that reported trust in law enforcement in the days before and after Blake’s shooting.
Table 2.
Demographic characteristics of survey respondents 4 weeks before and 4 weeks after Jacob Blake’s shooting.
Fig 3.
Changes in trust in law enforcement after Jacob Blake’s shooting among Black respondents.
Data are from the 2020 HCS survey. The survey question used to create the measure of trust in law enforcement is “To what extent do you trust your law enforcement agency?” We code answers to this question as binary, taking on value 1 for those that responded “a great extent” or “somewhat” and 0 for the rest. Each pair of point estimates shows the change in the probability of expressing trust for those surveyed one, two, three, and four weeks after Jacob Blake’s shooting, relative to the pool of survey respondents interviewed in the four weeks prior. Within each pair of estimates, we show results with and without the set of controls listed in Table 2. Confidence intervals are at the 95% level with standard errors robust to heteroskedasticity.
Fig 4.
Changes in trust in law enforcement after Jacob Blake’s shooting among White respondents.
Data are from the 2020 HCS survey. The survey question used to create the measure of trust in law enforcement is “To what extent do you trust your law enforcement agency?” We code answers to this question as binary, taking on value 1 for those that responded “a great extent” or “somewhat” and 0 for the rest. Each pair of point estimates shows the change in the probability of expressing trust for those surveyed one, two, three, and four weeks after Jacob Blake’s shooting, relative to the pool of survey respondents interviewed in the four weeks prior. Within each pair of estimates, we show results with and without the set of controls listed in Table 2. Confidence intervals are at the 95% level with standard errors robust to heteroskedasticity.