A Data-driven Investigation into the Northwestern Police Department
The murder of George Floyd by police officers in Minneapolis has once again turned the national spotlight to our anti-Black law enforcement system. After days of protests across the country, the Minneapolis City Council voted to disband their police department and promised sweeping reforms. For the first time, the ‘Defund The Police’ movement has mainstream appeal. Recently, a petition created by Northwestern University students demanding the abolition of the Northwestern Police Department (NUPD) has gained momentum, receiving over 7,500 signatures from community members. We wanted to take a closer look at the data surrounding the NUPD and ask tough questions about its impact on the Evanston community.
The Northwestern Police Department (NUPD) is a private police force paid for by Northwestern University, allowed under Section 110 of the Illinois Compiled Statutes. Northwestern police officers have the same powers as regular police such as the power to arrest. After confirmation from a NUPD representative, we found that their jurisdiction covers Evanston’s 1st and 5th wards.
We accessed data on the types of offenses handled by the NUPD in Evanston from 2016 to 2018. A breakdown of the 1,055 reported cases highlights the role they play for the community and provides important insight into how NUPD resources are used.
As shown in the above figure, 78.9% of the cases that the NUPD handled from 2016–2018 were Liquor or Drug Law offenses that resulted in disciplinary action. Clearly, a large majority of NUPD activity is focused on non-violent offenses. We argue that the over-policing of non-violent crimes shifts the focus of law enforcement away from community safety. Not only do narcotics laws fall short of keeping communities safe, but anti-drug laws have also been widely criticized as a tool to systematically target Black people. This begs the question: how can we quantify anti-Blackness in the NUPD?
The NUPD has faced significant criticism from the Northwestern student body regarding anti-Black policing. After Jeff Sessions’ talk in late 2019, students denounced NUPD’s anti-Black profiling in their response to the protests, and just a couple of days ago, in a Northwestern Daily article about the petition, SESP Junior and FMO coordinator Mari Gashaw was quoted as saying that “students at Northwestern have had NUPD called on them without reason, which demonstrates the lack of consequences for the police’s law mistreatment of [Black] people.” We wanted to see if these instances were outliers or part of a broader trend of the NUPD disproportionately policing Black people in their jurisdiction.
Both the Illinois Department of Transportation and NUPD report annual statistics on the department’s pedestrian stops. The criterion includes all non-traffic related incidents and searches conducted within NUPD’s jurisdiction. In order to tell the full story, we needed to create a statistic that accounted for Evanston’s demographics. To estimate the population in NUPD’s jurisdiction, we used 2010 Census tract demographic data and matched Census tracts to Evanston wards based on geographic location. We found that Census tracts 8092, 8093, 8087.02, and 8094 best estimate Evanston’s 1st and 5th wards.
Therefore, we computed a likelihood ratio by taking the percent of pedestrian stops involving Black and white people and dividing it by their respective population proportion. For example, Black people represent 18.79% of the population, and white people represent 57.74% of the population. In 2017, Black people comprised 40% of pedestrian stops, giving a likelihood ratio of 2.12. This indicates that in 2017, Black people were 2.12 times more likely to be the target of a pedestrian stop by the NUPD than a randomly selected person.
The figure above demonstrates that Black people have faced a significantly higher likelihood of being stopped compared to white people. In 2018, Black individuals were 3.73 times more likely to be subjected to a stop by the NUPD than a randomly-selected pedestrian whereas white individuals were 0.1 times as likely. On average, in these four years, Black people experienced a likelihood ratio of 2.68, while the white ratio was 0.67. This means that Black people were four times more likely to be stopped by the NUPD than white people from 2016–2019.
The data simply quantifies what Black communities have already known: the NUPD has an obvious pattern of discrimination against Black people. All the while, the majority of NUPD’s cases are nonviolent alcohol or drug-related offenses that are largely uncorrelated to community well-being yet tied with anti-Black policing. As two privileged people of color, we do not want to speak on behalf of the Black community in proposing specific policy changes. Therefore, we affirm that the administration should seriously reconsider NUPD’s presence on campus and consult with the Black community for alternative approaches to safety.
In conducting the analysis, we wanted to address two important limitations. First, we were unable to determine how much Northwestern actually spends on the NUPD. As with any article arguing for a decrease in funding for an organization, we recognize that it is important to indicate the current levels of funding. Unfortunately, the dataset on specific budget details is highly private. Second, when dealing with policing statistics, it is often difficult to identify a representative population. Our method is based on estimating the population of the 1st and 5th wards, but pedestrians are not confined to walking in the ward they live in. This is an inherent problem when trying to quantify discrimination in policing, and we will continue to improve our methods.
Ultimately, data matters in crafting demands and recommendations, especially when driving change within a large institution such as Northwestern. Hopefully, our results can supplement the current petition and bolster student leaders’ efforts to rethink policing on the Northwestern campus and its surrounding community. In the future, we want to look into the Evanston Police Department (EPD) and, more broadly, at university police departments nationally.
The analysis was accomplished using publicly available data from NUPD and IDOT. Thank you to our friends for providing feedback.
By Rahul Shukla and Shayak Bandyopadhyay
