Police unions have systematized and perfected the process of ensuring impunity for crooked cops, ensuring that even the most violent, lying, thieving, racist, authoritarian cops can stay on the force; at the same time, apologists for police violence and corruption tell us not to blame the whole force for the bad deeds of a few rotten apples.
Rotten apples spoil the barrel, though. In The network structure of police misconduct, a new Northwestern University study that analyzed 8 years worth of misconduct reports relating to more than 8,000 Chicago police officers, a trio of researchers from sociology, law and policy fields show that corruption is contagious, and that crooked cops create "misconduct networks" that initiate new officers into corrupt ways.
The defenders of the cops are right about one thing, though: most cops don't attract civilian complaints — "the modal number of civilian complaints is zero and average is something like 1.3"… but 3% of cops account for 27% of all complaints.
The dirty cop is likely to be male, between 25 and 45. 76% of cops who attract complaints are partnered with cops who are also named in the complaint. Cops who are paired with rookies are much less likely to engage in misconduct.
To reduce police corruption, the researchers recommend "mixed race" pairings of cops in high-crime areas, and more cops from "underrepresented groups."
This study set out to examine some of the possible factors associated with police misconduct in Chicago. Broadly, our results can be grouped into four key claims. First, police misconduct appears to be a networked phenomenon. Of the 6,780 Chicago Police officers who receive a civilian-facing complaint, 76% received at least one complaint alongside another officer, while 68% of the 6,015 officers who received a department-facing complaint did so alongside one or more of their colleagues. Furthermore, 55% and 42% of all civilian-facing and department-facing complaints, respectively, name more than one officer. These patterns of co-misconduct string together in various dyadic and extra-dyadic patterns to form larger networks that vary in size and connectivity across geographic police districts.
Second, levels of individual misconduct within networks are highly skewed, with a small number of officers receiving large numbers of co-complaints. Roughly 62% of all police officerswho were active for at least one year during 2010–2016 received at least one complaint from either a civilian or a fellow officer. The majority of officers receive relatively few complaints and tend to engage in misconduct with a small number of their fellow officers. However, as mall proportion of officers have a much larger number of complaints and a larger number of co-complainants: approximately 12% of officers have 10 or more unique co-complainants and the top 1% of officers have received a civilian co-complaint with 26 other officers and a departmental co-complaint with 14 other officers, on average. Coupled with the first key finding, this heavy-tailed degree distribution indicates that these high activity officers are likely responsible for much of the observed connectivity of the misconduct networks within police districts.
Third, there is significant age, gender, and racial variation in the receipt of civilian-facing and department-facing complaints. Age is strongly associated with misconduct, with younger officers more likely to have received both types of complaints than older officers, but especially complaints from civilians. Male officers have a 4 percentage points greater chance than a female officer of receiving a civilian-facing only complaint. Female officers, however, are 8 percentage points more likely to receive only departmental-facing complaints. Our finding on female officers and civilian complaints appears to confirm previous research showing that female officers receive fewer complaints and are less likely to be involved in use of force situations with civilians (14,79). Our finding that women are more likely to receive departmental facing complaints suggest that female officers may be receiving greater scrutiny or experiencing possible discrimination from fellow officers in the work environment—a finding not previously explored in the context of policing, but in line with findings around women's experience in the work place more generally(90).
In terms of race, white officers are marginally more likely than black or Hispanic officers to have received at least one complaint. Similar to female officers, however, black officers are more likely to receive departmental complaints than civilian complaints. As with gender, these findings also suggest that black officers might be experiencing greater scrutiny or discrimination on the job. Unfortunately, we lack sufficient data to explain the difference between civilian and departmental complaints for both black and female officers.
Fourth, experience and race affect the probability that officers are tied to other officers in misconduct networks. One of the largest effects in our models suggests that officers who are different in terms of their tenure are less likely to engage in misconduct together. While we lack the data needed to explore why this is the case, one hypothesis is that older officers might mitigate some of the inexperience or tendencies of younger officers, suggesting that pairing officers of different tenure might be given greater policy consideration.
The network structure of police misconduct [George Wood, Daria Roithmayr, and Andrew Papachristos/Northwestern University]