Black [O] vit Matter: Race, Crime et Fire to Kill aux États-Unis. Partie 2

Dans la première partie de l'article, j'ai décrit le contexte de l'étude, ses objectifs, ses hypothèses, ses contributions et ses outils. Maintenant, nous pouvons dire Gagarine sans plus tarder ...

Aller!

Nous importons les bibliothèques et définissons le chemin vers le répertoire avec tous les fichiers:

import pandas as pd, numpy as np

#      
ROOT_FOLDER = r'c:\_PROG_\Projects\us_crimes'

Mort aux mains de la loi

Commençons par analyser les données sur les victimes policières. Importons un fichier CSV vers DataFrame:

#    Fatal Encounters (FENC)
FENC_FILE = ROOT_FOLDER + '\\fatal_enc_db.csv'

#   DataFrame
df_fenc = pd.read_csv(FENC_FILE, sep=';', header=0, usecols=["Date (Year)", "Subject's race with imputations", "Cause of death", "Intentional Use of Force (Developing)", "Location of death (state)"])

, , : , ( ), ( , ), , .

, " " . , , FENC , , . , ( ). Fatal Encounters Excel ( ).

:

df_fenc.columns = ['Race', 'State', 'Cause', 'UOF', 'Year']
df_fenc.dropna(inplace=True)

, . . FENC, , (Hispanic/Latino), (Asian/Pacific Islander) (Middle Eastern). . :

df_fenc = df_fenc.replace({'Race': {'European-American/White': 'White', 'African-American/Black': 'Black', 
                          'Hispanic/Latino': 'White', 'Native American/Alaskan': 'American Indian',
                          'Asian/Pacific Islander': 'Asian', 'Middle Eastern': 'Asian',
                          'NA': 'Unknown', 'Race unspecified': 'Unknown'}}, value=None)

( ) :

df_fenc = df_fenc.loc[df_fenc['Race'].isin(['White', 'Black'])]

"UOF" ( )? , ( ) . , (, ) , . : 1) - (: , ; : ); 2) ; , , ( ) , . , :

df_fenc = df_fenc.loc[df_fenc['UOF'].isin(['Deadly force', 'Intentional use of force'])]

. CSV, :

df_state_names = pd.read_csv(ROOT_FOLDER + '\\us_states.csv', sep=';', header=0)
df_fenc = df_fenc.merge(df_state_names, how='inner', left_on='State', right_on='state_abbr')

df_fenc.head(), :

Race

State

Cause

UOF

Year

state_name

state_abbr

0

Black

GA

Gunshot

Deadly force

2000

Georgia

GA

1

Black

GA

Gunshot

Deadly force

2000

Georgia

GA

2

Black

GA

Gunshot

Deadly force

2000

Georgia

GA

3

Black

GA

Gunshot

Deadly force

2000

Georgia

GA

4

Black

GA

Gunshot

Deadly force

2000

Georgia

GA

, :

#     
ds_fenc_agg = df_fenc.groupby(['Year', 'Race']).count()['Cause']
df_fenc_agg = ds_fenc_agg.unstack(level=1)
#     UINT16  
df_fenc_agg = df_fenc_agg.astype('uint16')

2 : White ( ) Black ( ), ( 2000 2020). :

#        (- )
plt = df_fenc_agg.plot(xticks=df_fenc_agg.index, color=['olive', 'g'])
plt.set_xticklabels(df_fenc_agg.index, rotation='vertical')
plt.set_xlabel('')
plt.set_ylabel('-    ')
plt

:

() , .

2.4 . , . , .

( ):

#  CSV     (1991 - 2018)
POP_FILE = ROOT_FOLDER + '\\us_pop_1991-2018.csv'
df_pop = pd.read_csv(POP_FILE, index_col=0, dtype='int64')

:

#     -     2000 - 2018 .
df_pop = df_pop.loc[2000:2018, ['White_pop', 'Black_pop']]

#  ,    
df_fenc_agg = df_fenc_agg.join(df_pop)
df_fenc_agg.dropna(inplace=True)

#       
df_fenc_agg = df_fenc_agg.astype({'White_pop': 'uint32', 'Black_pop': 'uint32'})

. 2 , ( 1 . ):

df_fenc_agg['White_promln'] = df_fenc_agg['White'] * 1e6 / df_fenc_agg['White_pop']
df_fenc_agg['Black_promln'] = df_fenc_agg['Black'] * 1e6 / df_fenc_agg['Black_pop']

, :

Black

White

White_pop

Black_pop

White_promln

Black_promln

Year

2000

148

291

218756353

35410436

1.330247

4.179559

2001

158

353

219843871

35758783

1.605685

4.418495

2002

161

363

220931389

36107130

1.643044

4.458953

2003

179

388

222018906

36455476

1.747599

4.910099

2004

157

435

223106424

36803823

1.949742

4.265861

2005

181

452

224193942

37152170

2.016112

4.871855

2006

212

460

225281460

37500517

2.041890

5.653255

2007

219

449

226368978

37848864

1.983487

5.786171

2008

213

442

227456495

38197211

1.943229

5.576323

2009

249

478

228544013

38545558

2.091501

6.459888

2010

219

506

229397472

38874625

2.205778

5.633495

2011

290

577

230838975

39189528

2.499578

7.399936

2012

302

632

231992377

39623138

2.724227

7.621809

2013

310

693

232969901

39919371

2.974633

7.765653

2014

264

704

233963128

40379066

3.009021

6.538041

2015

272

729

234940100

40695277

3.102919

6.683822

2016

269

723

234644039

40893369

3.081263

6.578084

2017

265

743

235507457

41393491

3.154889

6.401973

2018

265

775

236173020

41617764

3.281493

6.367473

2 - . :

plt = df_fenc_agg.loc[:, ['White_promln', 'Black_promln']].plot(xticks=df_fenc_agg.index, color=['g', 'olive'])
plt.set_xticklabels(df_fenc_agg.index, rotation='vertical')
plt.set_xlabel('')
plt.set_ylabel('-    \n 1   ')
plt

:

df_fenc_agg.loc[:, ['White_promln', 'Black_promln']].describe()

White_promln

Black_promln

count ()

19.000000

19.000000

mean ( .)

2.336123

5.872145

std (. )

0.615133

1.133677

min (. )

1.330247

4.179559

25%

1.946485

4.890977

50%

2.091501

5.786171

75%

2.991827

6.558062

max (. )

3.281493

7.765653

:

1. 5.9 1 . 2.3 1 . ( 2.6 ).

2. () 1.8 , . ( , , .)

3. - 2013 . (7.7 ); - 2018 . (3.3 ).

4. ( 0.1 - 0.2 ), 2009 . 2011 - 2013 .

, :

- , , ?

- , . 2.6 , .

, - , , .

CSV :

CRIMES_FILE = ROOT_FOLDER + '\\culprits_victims.csv'
df_crimes = pd.read_csv(CRIMES_FILE, sep=';', header=0, index_col=0, usecols=['Year', 'Offense', 'Offender/Victim', 'White', 'White pro capita', 'Black', 'Black pro capita'])

- : , , , ( - "White", "Black" - "White pro capita", "Black pro capita").

(`df_crimes.head()`):

Offense

Offender/Victim

Black

White

Black pro capita

White pro capita

Year

1991

All Offenses

Offender

490

598

1.518188e-05

2.861673e-06

1991

All Offenses

Offender

4

4

1.239337e-07

1.914160e-08

1991

All Offenses

Offender

508

122

1.573958e-05

5.838195e-07

1991

All Offenses

Offender

155

176

4.802432e-06

8.422314e-07

1991

All Offenses

Offender

13

19

4.027846e-07

9.092270e-08

. :

#    ( )
df_crimes1 = df_crimes.loc[df_crimes['Offender/Victim'] == 'Offender']
#    (2000-2018)    
df_crimes1 = df_crimes1.loc[2000:2018, ['Offense', 'White', 'White pro capita', 'Black', 'Black pro capita']]

(1295 * 5 ):

Offense

White

White pro capita

Black

Black pro capita

Year

2000

All Offenses

679

0.000003

651

0.000018

2000

All Offenses

11458

0.000052

30199

0.000853

2000

All Offenses

4439

0.000020

3188

0.000090

2000

All Offenses

10481

0.000048

5153

0.000146

2000

All Offenses

746

0.000003

63

0.000002

...

...

...

...

...

...

2018

Larceny Theft Offenses

1961

0.000008

1669

0.000040

2018

Larceny Theft Offenses

48616

0.000206

30048

0.000722

2018

Drugs Narcotic Offenses

555974

0.002354

223398

0.005368

2018

Drugs Narcotic Offenses

305052

0.001292

63785

0.001533

2018

Weapon Law Violation

70034

0.000297

58353

0.001402

1 1 ( ). :

df_crimes1['White_promln'] = df_crimes1['White pro capita'] * 1e6
df_crimes1['Black_promln'] = df_crimes1['Black pro capita'] * 1e6

, ( ), :

df_crimes_agg = df_crimes1.groupby(['Offense']).sum().loc[:, ['White', 'Black']]

White

Black

Offense

All Offenses

44594795

22323144

Assault Offenses

12475830

7462272

Drugs Narcotic Offenses

9624596

3453140

Larceny Theft Offenses

9563917

4202235

Murder And Nonnegligent Manslaughter

28913

39617

Sex Offenses

833088

319366

Weapon Law Violation

829485

678861

:

plt = df_crimes_agg.plot.barh(color=['g', 'olive'])
plt.set_ylabel(' ')
plt.set_xlabel('-   ( 2000 - 2018 )')

, , :

  • , , " " , ,

  • , ( 2 " ")

, "" . , :

df_crimes_agg1 = df_crimes1.groupby(['Offense']).sum().loc[:, ['White_promln', 'Black_promln']]

White_promln

Black_promln

Offense

All Offenses

194522.307758

574905.952459

Assault Offenses

54513.398833

192454.602875

Drugs Narcotic Offenses

41845.758869

88575.523095

Larceny Theft Offenses

41697.303725

108189.184125

Murder And Nonnegligent Manslaughter

125.943007

1016.403706

Sex Offenses

3633.777035

8225.144985

Weapon Law Violation

3612.671402

17389.163849

:

plt = df_crimes_agg1.plot.barh(color=['g', 'olive'])
plt.set_ylabel(' ')
plt.set_xlabel('-    1    ( 2000 - 2018 )')

. ( ) , . " " 3 .

" " (All Offenses) , ( ) ( - , ).

#   'All Offenses' =  
df_crimes1 = df_crimes1.loc[df_crimes1['Offense'] == 'All Offenses']
#    , , ,    :
#df_crimes1 = df_crimes1.loc[df_crimes1['Offense'].str.contains('Assault|Murder')]

#       
df_crimes1 = df_crimes1.groupby(level=0).sum().loc[:, ['White_promln', 'Black_promln']]

:

White_promln

Black_promln

Year

2000

6115.058976

17697.409882

2001

6829.701429

20431.707645

2002

7282.333249

20972.838329

2003

7857.691182

22218.966500

2004

8826.576863

26308.815799

2005

9713.826255

30616.569637

2006

10252.894313

33189.382429

2007

10566.527362

34100.495064

2008

10580.520024

34052.276749

2009

10889.263592

33954.651792

2010

10977.017218

33884.236826

2011

11035.346176

32946.454471

2012

11562.836825

33150.706035

2013

11211.113491

32207.571607

2014

11227.354594

31517.346141

2015

11564.786088

31764.865490

2016

12193.026562

33186.064958

2017

12656.261666

34900.390499

2018

13180.171893

37805.202605

:

plt = df_crimes1.plot(xticks=df_crimes1.index, color=['g', 'olive'])
plt.set_xticklabels(df_fenc_agg.index, rotation='vertical')
plt.set_xlabel('')
plt.set_ylabel('-  \n 1   ')
plt

:

1. 2 , , , 3 ( ).

2. ( 2 18 ). , : 2001 2006 . , 2007 2016 , 2017 . 2 ( ).

3. 2007-2016 ., , .

, :

- ?

- 3 .

: , " , ?"

- - .

, :

#  
df_uof_crimes = df_fenc_agg.join(df_crimes1, lsuffix='_uof', rsuffix='_cr')
#    (.   )
df_uof_crimes = df_uof_crimes.loc[:, 'White_pop':'Black_promln_cr']

?

White_pop

Black_pop

White_promln_uof

Black_promln_uof

White_promln_cr

Black_promln_cr

Year

2000

218756353

35410436

1.330247

4.179559

6115.058976

17697.409882

2001

219843871

35758783

1.605685

4.418495

6829.701429

20431.707645

2002

220931389

36107130

1.643044

4.458953

7282.333249

20972.838329

2003

222018906

36455476

1.747599

4.910099

7857.691182

22218.966500

2004

223106424

36803823

1.949742

4.265861

8826.576863

26308.815799

2005

224193942

37152170

2.016112

4.871855

9713.826255

30616.569637

2006

225281460

37500517

2.041890

5.653255

10252.894313

33189.382429

2007

226368978

37848864

1.983487

5.786171

10566.527362

34100.495064

2008

227456495

38197211

1.943229

5.576323

10580.520024

34052.276749

2009

228544013

38545558

2.091501

6.459888

10889.263592

33954.651792

2010

229397472

38874625

2.205778

5.633495

10977.017218

33884.236826

2011

230838975

39189528

2.499578

7.399936

11035.346176

32946.454471

2012

231992377

39623138

2.724227

7.621809

11562.836825

33150.706035

2013

232969901

39919371

2.974633

7.765653

11211.113491

32207.571607

2014

233963128

40379066

3.009021

6.538041

11227.354594

31517.346141

2015

234940100

40695277

3.102919

6.683822

11564.786088

31764.865490

2016

234644039

40893369

3.081263

6.578084

12193.026562

33186.064958

2017

235507457

41393491

3.154889

6.401973

12656.261666

34900.390499

2018

236173020

41617764

3.281493

6.367473

13180.171893

37805.202605

, :

  1. White_pop -

  2. Black_pop -

  3. White promln_uof - ( 1 )

  4. Black promln_uof - ( 1 )

  5. White promln_cr - , ( 1 )

  6. Black promln_cr - , ( 1 )

, ... , :)

, . - :)

plt = df_uof_crimes['White_promln_cr'].plot(xticks=df_uof_crimes.index, legend=True)
plt.set_ylabel('-     1  .')
plt2 = df_uof_crimes['White_promln_uof'].plot(xticks=df_uof_crimes.index, legend=True, secondary_y=True, style='g')
plt2.set_ylabel('-     1  .', rotation=90)
plt2.set_xlabel('')
plt.set_xlabel('')
plt.set_xticklabels(df_uof_crimes.index, rotation='vertical')
plt

:

, . , :

plt = df_uof_crimes['Black_promln_cr'].plot(xticks=df_uof_crimes.index, legend=True)
plt.set_ylabel('-     1  .')
plt2 = df_uof_crimes['Black_promln_uof'].plot(xticks=df_uof_crimes.index, legend=True, secondary_y=True, style='g')
plt2.set_ylabel('-     1  .', rotation=90)
plt2.set_xlabel('')
plt.set_xlabel('')
plt.set_xticklabels(df_uof_crimes.index, rotation='vertical')
plt

:

: "", : , .

, :

df_corr = df_uof_crimes.loc[:, ['White_promln_cr', 'White_promln_uof', 'Black_promln_cr', 'Black_promln_uof']].corr(method='pearson')
df_corr.style.background_gradient(cmap='PuBu')

:

White_promln_cr

White_promln_uof

Black_promln_cr

Black_promln_uof

White_promln_cr

1.000000

0.885470

0.949909

0.802529

White_promln_uof

0.885470

1.000000

0.710052

0.795486

Black_promln_cr

0.949909

0.710052

1.000000

0.722170

Black_promln_uof

0.802529

0.795486

0.722170

1.000000

: = 0.885, = 0.722. , , , ( ), . , , , .

, . ( , , ). : ( 100, %):

#   ( )
df_uof_crimes_agg = df_uof_crimes.loc[:, ['White_promln_cr', 'White_promln_uof', 'Black_promln_cr', 'Black_promln_uof']].agg(['mean', 'sum', 'min', 'max'])
# ""   
df_uof_crimes_agg['White_uof_cr'] = df_uof_crimes_agg['White_promln_uof'] * 100. / df_uof_crimes_agg['White_promln_cr']
df_uof_crimes_agg['Black_uof_cr'] = df_uof_crimes_agg['Black_promln_uof'] * 100. / df_uof_crimes_agg['Black_promln_cr']

:

White_promln_cr

White_promln_uof

Black_promln_cr

Black_promln_uof

White_uof_cr

Black_uof_cr

mean

10238.016198

2.336123

30258.208024

5.872145

0.022818

0.019407

sum

194522.307758

44.386338

574905.952459

111.570747

0.022818

0.019407

min

6115.058976

1.330247

17697.409882

4.179559

0.021754

0.023617

max

13180.171893

3.281493

37805.202605

7.765653

0.024897

0.020541

:

plt = df_uof_crimes_agg.loc['mean', ['White_uof_cr', 'Black_uof_cr']].plot.bar(color=['g', 'olive'])
plt.set_ylabel(' -   - ')
plt.set_xticklabels(['', ''], rotation=0)

, , . , , - .

:

1. ( ). : , .

2. , " " , ( ). , "" ( -> -> -> ).

3. , . .

, :

- , ?

- Oui, une telle corrélation est observée, bien qu'elle soit hétérogène par race: pour les blancs c'est presque idéal, pour les noirs c'est presque imparfaite.

Dans la prochaine partie de l'article, nous examinerons la répartition géographique des données analysées dans les États américains.

Lien vers la version anglaise de l'article (à la demande des travailleurs).




All Articles