This notebook looks at data from the State of California Department of Justice’s Open Justice site in an attempt to analyze the role played by race in determining the outcomes – “dispositions” – of arrests and the use of force during arrests. Below is the code used to experiment and investigate.

Code

#import CA DOJ data on arrest dispositions

Code

import pandas as pd
df = pd.read_csv('./OnlineArrestDispoData1980-2020.csv')

Code

#select for arrests after the year 2010
#separate the data by the race of people arrested and the disposition code

Code

df = df[df['YEAR'] > 2010]
df['TOTAL'] = df['F_TOTAL'] + df['M_TOTAL'] + df['S_TOTAL']

Code

import matplotlib.pyplot as plt

df_race = df[df['RACE'] == "Black"]

df_release = df_race[df_race['ARREST_DISP_CODE'] == "Released"]
df_complaint_sought = df_race[df_race['ARREST_DISP_CODE'] == "Complaint Sought"]
df_to_other = df_race[df_race['ARREST_DISP_CODE'] == "To Other Agency"]
df_within_dept = df_race[df_race['ARREST_DISP_CODE'] == "Within Department"]
df_probation = df_race[df_race['ARREST_DISP_CODE'] == "Juvenile Probation"]

total_arrest = df_race['TOTAL'].sum()

total_release = df_release['TOTAL'].sum()
total_complaint_sought = df_complaint_sought['TOTAL'].sum()
total_to_other = df_to_other['TOTAL'].sum()
total_within_dept = df_within_dept['TOTAL'].sum()
total_probation = df_probation['TOTAL'].sum()

# initialize list of lists
data = [['Black',(total_release/total_arrest)*100,
         (total_to_other/total_arrest)*100, (total_within_dept/total_arrest)*100, 
          (total_probation/total_arrest)*100, (total_complaint_sought/total_arrest)*100]]
  
df_data_black = pd.DataFrame(data, columns = ['Race', 'Released', 'To Other Agency', 
                                         'Within Department', 'Juvenile Probation', 'Complaint Sought'])
                        

Code

df_race = df[df['RACE'] == "White"]

df_release = df_race[df_race['ARREST_DISP_CODE'] == "Released"]
df_complaint_sought = df_race[df_race['ARREST_DISP_CODE'] == "Complaint Sought"]
df_to_other = df_race[df_race['ARREST_DISP_CODE'] == "To Other Agency"]
df_within_dept = df_race[df_race['ARREST_DISP_CODE'] == "Within Department"]
df_probation = df_race[df_race['ARREST_DISP_CODE'] == "Juvenile Probation"]

total_arrest = df_race['TOTAL'].sum()

total_release = df_release['TOTAL'].sum()
total_complaint_sought = df_complaint_sought['TOTAL'].sum()
total_to_other = df_to_other['TOTAL'].sum()
total_within_dept = df_within_dept['TOTAL'].sum()
total_probation = df_probation['TOTAL'].sum()

# initialize list of lists
data = [['White',(total_release/total_arrest)*100,
         (total_to_other/total_arrest)*100, (total_within_dept/total_arrest)*100, 
          (total_probation/total_arrest)*100, (total_complaint_sought/total_arrest)*100]]
  
df_data_white = pd.DataFrame(data, columns = ['Race', 'Released', 'To Other Agency', 
                                         'Within Department', 'Juvenile Probation', 'Complaint Sought'])
                    

Code

df_race = df[df['RACE'] == "Hispanic"]

df_release = df_race[df_race['ARREST_DISP_CODE'] == "Released"]
df_complaint_sought = df_race[df_race['ARREST_DISP_CODE'] == "Complaint Sought"]
df_to_other = df_race[df_race['ARREST_DISP_CODE'] == "To Other Agency"]
df_within_dept = df_race[df_race['ARREST_DISP_CODE'] == "Within Department"]
df_probation = df_race[df_race['ARREST_DISP_CODE'] == "Juvenile Probation"]

total_arrest = df_race['TOTAL'].sum()

total_release = df_release['TOTAL'].sum()
total_complaint_sought = df_complaint_sought['TOTAL'].sum()
total_to_other = df_to_other['TOTAL'].sum()
total_within_dept = df_within_dept['TOTAL'].sum()
total_probation = df_probation['TOTAL'].sum()

# initialize list of lists
data = [['Hispanic',(total_release/total_arrest)*100,
         (total_to_other/total_arrest)*100, (total_within_dept/total_arrest)*100, 
          (total_probation/total_arrest)*100, (total_complaint_sought/total_arrest)*100]]
  
df_data_hispanic = pd.DataFrame(data, columns = ['Race', 'Released', 'To Other Agency', 
                                         'Within Department', 'Juvenile Probation', 'Complaint Sought'])
                    

Code

df_race = df[df['RACE'] == "Other"]

df_release = df_race[df_race['ARREST_DISP_CODE'] == "Released"]
df_complaint_sought = df_race[df_race['ARREST_DISP_CODE'] == "Complaint Sought"]
df_to_other = df_race[df_race['ARREST_DISP_CODE'] == "To Other Agency"]
df_within_dept = df_race[df_race['ARREST_DISP_CODE'] == "Within Department"]
df_probation = df_race[df_race['ARREST_DISP_CODE'] == "Juvenile Probation"]

total_arrest = df_race['TOTAL'].sum()

total_release = df_release['TOTAL'].sum()
total_complaint_sought = df_complaint_sought['TOTAL'].sum()
total_to_other = df_to_other['TOTAL'].sum()
total_within_dept = df_within_dept['TOTAL'].sum()
total_probation = df_probation['TOTAL'].sum()

# initialize list of lists
data = [['Other',(total_release/total_arrest)*100,
         (total_to_other/total_arrest)*100, (total_within_dept/total_arrest)*100, 
          (total_probation/total_arrest)*100, (total_complaint_sought/total_arrest)*100]]
  
df_data_others = pd.DataFrame(data, columns = ['Race', 'Released', 'To Other Agency', 
                                         'Within Department', 'Juvenile Probation', 'Complaint Sought'])
                    

Code

#display the disposition data, now sorted by race & disposition

Code

df_final = pd.concat([df_data_black, df_data_white, df_data_hispanic, df_data_others])
df_final = df_final.set_index('Race')
df_final.plot.barh(stacked=True)

<AxesSubplot:ylabel='Race'>

The majority of arrests by far result in a complaint sought, regardless of race. For the disposition of a referral to juvenile probation, Black and Hispanic individuals dominate numerically.

Code

#import CA DOJ data on use of force during arrests 

Code

force_df = pd.read_csv('./URSUS_Civilian-Officer_2020.csv')

Code

#break down the population of victims of force during arrests by race

Code

officer_force_df = force_df[force_df['Civilian_Officer'] ==  "Civilian"]
force_racial_breakdown_df = officer_force_df.groupby(['Race_Ethnic_Group'], as_index=True).size()
force_racial_breakdown_df.plot.barh()

<AxesSubplot:ylabel='Race_Ethnic_Group'>

Hispanic individuals are clearly the victims of force during arrests at a far greater rate than individuals of any other race or ethnic group, with white people coming at second.

Code

#break down the (assaultive) resistance of defendants during arrests by race

Code

civilian_resistance_df = force_df[force_df['Civilian_Officer'] ==  "Civilian"]
civilian_resistance_df = civilian_resistance_df[civilian_resistance_df['CIVILIAN_Resistance_Type'] =="assaultive"]
racial_breakdown_df = civilian_resistance_df.groupby(['Race_Ethnic_Group'], as_index=True).size()
racial_breakdown_df.plot.barh(stacked= True)

<AxesSubplot:ylabel='Race_Ethnic_Group'>

Here, too, Hispanics outnumber every other racial and ethnic group.