Impact of Reform on Chicago Police Accountability

(Github Project Repo)

In the past 20 years in Chicago, there have been a few attempts at reforming how complaints and allegations against Chicago police are handled, usually in response to major scandals. 

Up until 2007, police misconduct complaints were handled by the Office of Professional Standards (OPS), a branch of the Chicago Police Department. In response to several scandals with regard to the 'code of silence' that suppressed police officers from holding each other accountable, the Chicago city council voted to dissolve OPS, and establish two separate entities to review police complaints. The Independent Police Review Authority (IPRA) would be outside CPD and have leadership appointed by the mayor, and it would review cases related to “excessive force, domestic violence, coercion through violence, or verbal bias-based abuse.” The CPD-internal Bureau of Internal Affairs (BIA) would review all other complaints.

In the aftermath of several scandals regarding CPD, including the murder of unarmed teenager Laquan McDonald by a Chicago police officer and the uncovering of secret police torture sites, a coalition formed demanding citizen accountability of police. A broad alliance of activists and groups have proposed forming a Civilian Police Accountability Council, an independent and fully-elected board that would investigate allegations of police misconduct and police shootings, and be the final authority on police discipline.

The purpose of my project was to build an explanatory model that would uncover what features of a police complaint make it more likely to result in discipline, and to determine if the formation of IPRA led to any meaningful increase in accountability. 

The data for this project came from the Citizen's Police Data Project (CPDP), an effort of Chicago's Invisible Institute to bring internal practices of the criminal justice system under public scrutiny.

I specifically focused on police complaints that were made between 2001 and 2015, in order to be symmetric around the formation of IPRA, and to avoid recent complaints that may still be in process.

A broad overview of the data indicated that both before and after the formation of IPRA, the vast majority of complaints do not result in discipline. 

In my dataset, each unique complaint was associated with information related to the date/time of the complaint, the nature of the complaint, whether the complaint had witnesses, and the disciplinary recommendation and final outcome, as well as the implicated officer. I also pulled in data about the officers, including their race, sex, age, rank and time on the force at the time of the complaint.

I wanted to make my model as simple and user-friendly as possible, so I wanted to trim down my features from the over 50 or so that I was starting with. By building a simple logistic regression model and examining the coefficients associated with the features, I was able to identify the more critical features to determining the outcome of a complaint:

- Officer-Filed or Civilian-Filed: A complaint filed by another police officer was significantly more likely to result in discipline than a complaint filed by a civilian. This result corroborates a similar finding by the CPDP released on their own website

- Type of Complaint: Certain types of complaints, like illegal search complaints, essentially never result in discipline for the police officer. This could be because illegal searches are addressed in courtrooms by ruling evidence as inadmissible - a consequence which has no bearing on the behavior of the officer. This provides little disincentive to police officers to respect the rights of people they have stopped in the moment.

- Number of Complaints Against Officer: The total number of complaints against an officer was positively correlated with a particular complaint ending in discipline - however, only for certain violations. When I disaggregated the complaint count against officers by type of complaints, I found that there were several types of complaints (use of force, violation of civil rights, illegal searches, etc) where being a repeat offender was actually negatively correlated with discipline. 

- Number of Witnesses: The primary factor under the control of a complainant that was positively correlated with officer discipline was the number of witnesses to the complaint willing to give affidavits. Approximately 6% of civilian complaints with greater than 5 witnesses resulted in discipline, as opposed to <1% of civilian complaints with 1 or fewer witnesses.

---

The Effect of Reform

In general, the implementation of IPRA had minimal effect on the overall rate of police discipline, as indicated in the pie chart above. However, when I disaggregated complaints by type, I did see one minimal effect of IPRA, which was specifically confined to the kind of complaints IPRA (and today, COPA) is intended to investigate. After the formation of IPRA, there was a slight increase in the frequency of discipline for complaints related to use of force and verbal abuse.

Allegations of verbal abuse such as racial slurs were disciplined 6% of the time after IPRA, despite being disciplined less than 2% of the time before.

In general, it seems that independent bodies have a small impact on increasing police accountability, but only for the complaints which are sent to them (a process which in itself needs greater transparency).  The vast majority of complaints are never sent to the review boards, a statistic that has remained true under the new review board COPA.

Even when the complaint is diverted to an independent review board, the rates of discipline for police complaints are still pitifully small, especially when the complainant is not a fellow officer. These findings emphasize the need for a more transparent review process that can be subjected to democratic scrutiny.