Skip to Content, Navigation, or Footer.
Tuesday, Dec. 5
The Indiana Daily Student

academics & research crime & courts

IU study finds average higher police presence in non-white neighborhoods


In a recent joint study, researchers from Indiana University, the University of California Los Angeles and Irvine, and American University used smartphone data to find there is a higher police presence in non-white neighborhoods.  

The researchers, according to an article from News at IU, used anonymized smartphone data from almost 10,000 police officers in 21 large U.S. cities, including Indianapolis. According to the article, this marks the first-time anonymized smartphone location data has been used to identify and study police officer’s movements while on patrol. 

Kate Christensen, IU assistant professor of marketing and one of the article’s authors, said different police departments release varying amounts of data, making it difficult to compare different departments. Additionally, when most of that data is released, she said, it’s about arrests, stops and crime rates, rather than knowing where police were 

“I was interested to see if where the police spend time was equal after controlling for neighborhood characteristics between racial groups,” Christensen said, “If there were differences, was that connected to different levels of arrest rates for different groups?” 

The researchers also found that, while the higher presence of police in non-white neighborhoods could be explained by neighborhood characteristics — like the location of murders, local income and education — for Black and Hispanic neighborhoods the difference in time spent remains after controlling for these characteristics.  

Christensen said researchers also found differences in time spent by police between cities. In some cities, she said, most of where police spent their time can be explained by racial demographics while in some cities, it can be explained by socioeconomics. 

On average, according to the article, police spent 0.36% more time in neighborhoods for each percentage point increase in Black residents, 0.52% more time in neighborhoods for each percentage point increase in Hispanic residents and 0.37% more time in neighborhoods for each percentage point increase in Asian residents.  

Christensen said she and other researchers tried to separate things related to socioeconomics, social cohesion and crime, from things like the racial demographics of where police spend their time, which departments might need to be more conscious of. 

In Black neighborhoods, Christensen said, the difference in time spent in neighborhoods can statistically explain over half of the higher arrest rate of Black people. 

In 2020, according to the Office of Juvenile Justice and Delinquency Prevention, Black people of all ages made up 26% of arrests, despite being 13% of the U.S. population, according to the U.S. Census. 

The analysis, she said, is preliminary, so researchers cannot say if there is over or under policing in specific areas based on this research and the researchers don’t know exactly why these things are happening. 

Black American’s experience of being over-policed, according to a research article, stems from law-enforcement's disproportionate attention to real, perceived or potential criminal activity suspect of racial minorities. The article also found police may ignore communities of color because they believe these citizens are responsible for the crime and disorder in their neighborhoods. 

Christensen said the researchers tried, to the best of their abilities, to identify who might be a police officer with phone location data. When they identified a police officer, she said the researchers wanted to try and figure out where they went during their shift.  

“It's not perfect, there's a lot of noise in the data,” Christensen said. 

The researchers used data from Safegraph, according to the article from IU, which recorded “pings” indicating where smartphones are at a certain time. This information was connected to police station location data published by the Department of Homeland Security and geofence data provided by Microsoft. The sample included phones used at 316 police stations in 21 cities between February and November 2017 in order to identify patrol officers.  

When people think about crime, Christensen said people just think about statistics released by departments instead of the impact of where police go on which crimes are observed and reported.  

Christensen said she doesn’t know what impact the research will have on policing practice, but she presented the work to the Toronto Police Department, and it was very interested. 

One possibility, Christensen said, is police departments could have GPS location data or the ability to collect it themselves and measure it. 

Christensen said the study provides a different way of looking at policing.  

“I'd like us to think about policing as not just about the end result, which is the arrest, but also about time’s spent within cities, and how citizens feel about that time,” Christensen said. 

Get stories like this in your inbox