STUDIES:   

654/

“Police killings of unarmed Blacks and unarmed Hispanics have a significant, positive effect on Black homicide changes.” & “The actual killings of unarmed civilians … and not just the public’s protest against them, drive the effect [of increased homicides]."

A Ferguson Effect, the Drug Epidemic, Both, or Neither? Explaining the 2015 and 2016 U.S. Homicide Rises by Race and Ethnicity A Ferguson Effect, the Drug Epidemic, Both, or Neither? Explaining the 2015 and 2016 U.S. Homicide Rises by Race and Ethnicity Table 3. Ordinary Least Squares Regression for Police Killings and Race/Ethnic-Specific Homicide Rate Changes (n = 586 Counties). The results are compelling and give credence to the Ferguson effect hypothesis, especially for Black homicide changes. Over the 2014 to 2015 period, police killings of unarmed Blacks and unarmed Hispanics have a significant, positive effect on Black homicide changes. Specifically, a one-unit increase in police killings of unarmed Blacks and Hispanics increases Black homicide rates by a factor of 0.95 (rse = 0.18, p < .01) and 3.57 (rse = 1.61, p < .05), respectively. Consistent with our hypothesis regarding the conditional effect of police illegitimacy, the significant, positive interaction term for police killings of unarmed Blacks × structural disadvantage indicates that the criminogenic effect of police violence was stronger in counties with high structural disadvantage. Its coefficient of 0.33 (rse = 0.17, p < .1) is relatively smaller, indicating that while there is an additional component specific to disadvantaged communities, this is a more general phenomenon. Table 3. Ordinary Least Squares Regression for Police Killings and Race/Ethnic-Specific Homicide Rate Changes (n = 586 Counties). The results are compelling and give credence to the Ferguson effect hypothesis, especially for Black homicide changes. Over the 2014 to 2015 period, police killings of unarmed Blacks and unarmed Hispanics have a significant, positive effect on Black homicide changes. Specifically, a one-unit increase in police killings of unarmed Blacks and Hispanics increases Black homicide rates by a factor of 0.95 (rse = 0.18, p < .01) and 3.57 (rse = 1.61, p < .05), respectively. Consistent with our hypothesis regarding the conditional effect of police illegitimacy, the significant, positive interaction term for police killings of unarmed Blacks × structural disadvantage indicates that the criminogenic effect of police violence was stronger in counties with high structural disadvantage. Its coefficient of 0.33 (rse = 0.17, p < .1) is relatively smaller, indicating that while there is an additional component specific to disadvantaged communities, this is a more general phenomenon.
A Ferguson Effect, the Drug Epidemic, Both, or Neither? Explaining the 2015 and 2016 U.S. Homicide Rises by Race and Ethnicity
Homicide Studies
Gaston, Shytierra / Cunningham, Jamein P. / Gillezeau, Rob
Vol. 23 Issue 3, pp. 285–313, 2019
10.1177/1088767919849642
Thoughts?
Comment on this study on its Twitter thread.