The role of race in Louisville traffic stop searches

   

   

Introduction:

• I examined a dataset from the Stanford Open Policing Project that contains several years' of police traffic stops in Louisville, Kentucky, where the police shooting of Breonna Taylor in March was a factor in sparking nationwide protests last year. My objective was to determine whether race played a role in the outcomes of traffic stops in Louisville. I was particularly interested in searches, which mark a more aggressive and invasive stance by an officer and inherently involve questions of civil rights.

   

The data:

https://openpolicing.stanford.edu/data/

The dataset contained details on more than 100,000 traffic stops over three years.

Features included:

• Age, race and gender of people stopped
• Age, race and gender of police officers who stopped them
• Location (latitude and longitude) of the stop
• Police division where the stop occurred
• The stated reason for the stop
• Whether the citizen was searched, frisked, given a citation

   

Among the questions asked and answered:

• Do Black citizens face a higher probability of being searched than white citizens?

• Are white officers or Black officers more likely to search a Black citizen?

• Are Black officers more likely to search a Black citizen or a white one?

• Are Latino officers more likely to search someone than a white officer?

       

An overview of stops, searches and citations by race of citizens

• Blacks are overrepresented as a percentage of the population in stops, searches and citations.

• Whites, Latinos and Asians are underrepresented.

       

   

• The percentage of Black motorists who were searched was nearly double that of whites.

• But white and Latino citizens were given citations at a higher rate.

   

       

The probabilities of Black and white citizens being searched

• In 10,000 Monte Carlo simulations, the probability that Blacks face a higher probability of being searched than whites: 1.0.  
• Worlds apart: plotting the beta distributions of the rates of searches of Blacks and whites.

       

       

Considering an officer's race in outcomes

• White, Black and Asian officers were more likely to search a Black motorist than one of another race.    
• Latino officers searched white motorists the most often, followed by Latinos.

       

       

• White, Latino and Black officers gave citations to white and Latino citizens at higher rates than they did Black citizens.    
• Latino officers appear far less aggressive in giving citations than in conducting searches.

             

 

Do Black citizens face a higher probability of being searched by a white officer than a Black one?

• Ten thousand simulations found that the probability that Black citizens face a higher probability of being searched by a white officer is 1.0.    
• Plotting the distributions of Black and white officers' searches of Black citizens illustrates the point.

           

 

Are Black officers more likely to search Black citizens than white citizens?

• Yes, the probability that a Black citizen faces a higher probability of being searched by a Black officer is 1.0.    

       

Do people stopped by Latino officers face a higher probability of being searched versus a white officer?

• In 10,000 simulations, the probability that the chance of being searched by a Latino officer is higher: 1.0.    
• Plotting the distribution of searches by Latino and white officers.        

       

The role of geography in outcomes

   

Louisville ranks as 30th most segregated city in U.S., according to Census Bureau

       

       

The one division where white motorists more likely to be searched than Black ones

• In second division: 73% of all stops were of Black motorists, but white motorists faced higher probability of being searched.

• In 10,000 simulations, the probability of white motorists facing a greater chance of search was 1.0.

• Rates of search were high for both: 22.8% for whites, 18% for Blacks.

• More than twice as many stops of Black motorists than white ones.

   

   

   

Comparing search probabilities for Black citizens in affluent, mostly white neighborhoods

• Black motorists face a greater chance of being searched in the 1st Division than in the affluent, mostly white 5th Division.