Desktop version

Home arrow Business & Finance

  • Increase font
  • Decrease font


<<   CONTENTS   >>

Estimation results

The results from estimating equation (5.5) are shown in Table 5.3. The column headed “Stop Rate” shows the PSR associated with the various ethnicities, first computed over all Areas and then, separately, for the five largest police areas in E&W: Greater Manchester, Merseyside, Metropolitan, Thames Valley, West Midlands, and West Yorkshire. So, for the areas considered in aggregate, the PSR in terms of the major ethnicities were

12.9 stops/1,000 for British Whites, 22.6 for Black Africans, 62.7 for Black Caribbeans, 8.2 for Indians, 19.9 for Bangladeshis, and 22.3 for Pakistanis. The corresponding sample stop rates (dk) were, respectively, 11.6, 42.5, 86.3, 12.7, 37.1, and 23.6. The divergence between the sample stop rates, ak, and the PSR, dk, is analysed in detail in a later section, but suffice it here to note that, of the ethnic minorities which had a disproportionately large presence in the Metropolitan Area - Black Africans, Black Caribbeans, Indians, Bangladeshis, and Pakistanis - the sample stop rates were greater than the PSR (ak > dk) while for British Whites, who had a disproportionately small presence in the Metropolitan Area, the sample stop rate was smaller than the PSR (dk < ak).

The column in Table 5.3 headed “Marginal Rate” represents, for the relevant ethnic group, the difference between the PSR of that group and that of the reference group, denoted by [R], of British Whites. So, for example, the marginal rate of Black Caribbeans was 49.7 points, meaning that the PSR of Black Caribbeans (62.7) was 49.7 points greater than the PSR of British Whites (12.9). Dividing these marginal probabilities by their standard errors (shown in the next column) yielded the t-values. These showed whether the marginal rates were significantly different from zero in the sense that the likelihood of observing these values under the null hypothesis that the marginal rates were zero - that is, of no difference between the PSR of the target and reference group - was less than 5% (superscript ** in Table 5.3) or 10% (superscript *).

Although popular discussion of police stops in E&W pits the country’s Black, Asian, and Minority Ethnic (BAME) population against its White citizens in order to emphasise the greater frequency with which the former are stopped relative to the latter, the results in Table 5.3 suggest a subtler unfolding of police attitudes with respect to BAME persons. First, within the BAME group, Indians and Chinese were predicted to have significantly lower stop rates than British Whites (respectively, 8.2 and 5.5 vs. 12.9). Second, while on average, Black persons were more likely to be stopped

Table 5.3 Predicted stop rates per 1,000 in selected Police Areas of England and

Wales, by ethnicity

Stop Rate

Marginal

Rate

Standard

Error

t -Value

All Police Areas

White British [R|

12.9

White Irish

9.2

-3.8**

1.5

-2.5

Other White

15.1

2.1**

0.7

3.0

Black African

22.6

9.6**

1.8

5.4

Black Caribbean

62.7

49.7**

2.8

17.7

Other Black

87.7

74.8**

3.9

19.4

Indian

8.2

-4.7**

1.3

-3.6

Bangladeshi

19.9

7.0**

2.5

2.8

Pakistani

22.3

9.4**

2.3

4.2

Other Asian

18.0

5.1**

1.4

3.6

White/Asian

6.1

-6.8**

1.8

-3.8

White/African

12.7

-0.2

2.8

-0.1

White/Caribbean

24.0

11.0**

1.7

6.4

Other Mixed

19.8

6.9**

2.1

3.3

Chinese

5.5

-7.5**

1.8

-4.2

Other

10.8

-2.1

1.8

-1.1

Greater Manchester

White British [R|

11.1

White Irish

3.2

-7.9

5.0

-1.6

Other White

6.6

-4.5

3.5

-1.3

Black African

17.7

6.5

4.4

1.5

Black Caribbean

54.9

43.8**

6.9

6.3

Other Black

33.5

22.4**

8.5

2.6

Indian

5.1

-6.0

4.0

-1.5

Bangladeshi

8.4

-2.7

5.0

-0.6

Pakistani

15.7

4.6*

2.6

1.8

Other Asian

15.8

4.7

5.5

0.9

White/Asian

6.4

-4.7

7.4

-0.6

White/African

14.1

3.0

9.2

0.3

White/Caribbean

30.0

18.9**

6.1

3.1

Other Mixed

16.4

5.3

8.4

0.6

Chinese

1.7

-9.4

6.0

-1.6

Other

7.5

-3.6

5.6

-0.6

Merseyside

White British [R|

26.3

White Irish

9.1

-17.2**

8.0

-2.2

Stop Rate

Marginal

Rate

Standard

Error

t -Value

Other White

8.2

-18.1**

6.0

-3.0

Black African

14.8

-11.5

9.3

-1.2

Black Caribbean

79.7

53.4**

20.2

2.6

Other Black

136.4

110.1**

17.7

6.2

Indian

4.8

-21.4**

10.4

-2.1

Bangladeshi

9.6

-16.7

18.9

-0.9

Pakistani

15.6

-10.7

18.1

-0.6

Other Asian

22.6

-3.7

11.9

-0.3

White/Asian

8.8

-17.5

13.5

-1.3

White/African

15.0

-11.3

13.2

-0.9

White/Caribbean

23.3

-3.0

11.5

-0.3

Other Mixed

26.4

0.1

13.0

0.0

Chinese

12.4

-13.9

8.6

-1.6

Other

9.0

-17.3**

9.2

-1.9

Metropolitan

White British |R]

31.2

White Irish

27.4

-3.8*

2.2

-1.7

Other White

46.2

15.0**

1.0

14.7

Black African

62.8

31.6**

1.3

24.2

Black Caribbean

119.6

88.5**

1.6

54.1

Other Black

186.0

154.8**

2.3

68.0

Indian

22.2

-9.0**

1.3

-6.8

Bangladeshi

63.8

32.6**

2.0

16.2

Pakistani

42.9

11.7**

2.0

5.9

Other Asian

50.5

19.3**

1.5

12.6

White/Asian

11.0

-20.2**

2.9

-6.9

White/African

22.5

-8.7**

3.6

-2.4

White/Caribbean

51.6

20.4**

2.7

7.6

Other Mixed

48.4

17.2**

2.7

6.4

Chinese

12.9

-18.3**

2.7

-6.9

Other

29.0

-2.2

1.8

-1.2

Thames Valley

White British [R]

8.8

White Irish

7.5

-1.3

6.0

-0.2

Other White

7.6

-1.2

2.6

-0.5

Black African

13.2

4.4

4.5

1.0

Black Caribbean

39.8

31.0**

6.7

4.7

Other Black

46.7

37.9**

10.6

3.6

Indian

5.4

-3.4

3.5

-1.0

(Continued)

122 Racial bias in policing Table 5.3 (Continued)

Stop Rate

Marginal

Rate

Standard

Error

t -Value

Bangladeshi

25.7

16.9

10.4

1.6

Pakistani

22.7

13.9**

3.5

4.0

Other Asian

6.9

-1.9

4.6

-0.4

White/Asian

6.9

-1.9

6.9

-0.3

White/African

15.8

7.0

11.1

0.6

White/Caribbean

36.4

27.7**

6.9

4.0

Other Mixed

13.3

4.5

8.1

0.6

Chinese

2.2

-6.6

7.1

-0.9

Other

6.6

-2.2

7.3

-0.3

West Midlands

White British [R|

6.0

White Irish

2.7

-3.2

4.7

-0.7

Other White

8.6

2.6

3.5

0.8

Black African

10.8

4.8

4.0

1.2

Black Caribbean

26.6

20.6**

3.3

6.2

Other Black

21.1

15.1**

5.4

2.8

Indian

6.8

0.8

2.2

0.4

Bangladeshi

6.5

0.5

4.2

0.1

Pakistani

18.1

12.1**

2.2

5.6

Other Asian

9.0

3.0

3.8

0.8

White/Asian

5.3

-0.7

6.2

-0.1

White/African

13.1

7.1

11.8

0.6

White/Caribbean

14.1

3.6

7.5

0.5

Other Mixed

9.6

8.1**

4.0

2.0

Chinese

2.6

-3.4

6.3

-0.5

Other

3.8

-2.2

4.5

-0.5

West Yorkshire

White British [R|

14.3

White Irish

7.2

-7.1

7.6

-0.9

Other White

9.2

-5.0

3.9

-1.3

Black African

8.1

-6.2

5.9

-1.1

Black Caribbean

49.1

34.9**

7.4

4.7

Other Black

31.2

17.0

11.7

1.5

Indian

8.7

-5.6

4.0

-1.4

Bangladeshi

10.1

-4.2

7.7

-0.5

Pakistani

21.4

7.1**

2.2

3.2

Other Asian

13.2

-1.0

6.2

-0.2

White/Asian

8.7

-5.6

7.5

-0.8

Stop Rate

Marginal

Rate

Standard

Error

t -Value

White/African

13.9

-0.4

13.5

0.0

White/Caribbean

27.6

13.3**

6.4

2.1

Other Mixed

16.5

2.2

10.6

0.2

Chinese

4.0

-10.3

8.9

-1.2

Other

4.8

-9.5

6.5

-1.5

Source: Own calculations from www.ethnicity-facts-figures.service.gov.uk/crime-justice-and- the-law/policing/stop-and-search/latest.

** Marginal probability significant at 5% level; * Marginal probability significant at 10% level.

than British Whites, within the “Black” category Black Africans had a significantly lower PSR than Black Caribbeans (22.6 vs. 62.7). In fact, the gap between the two groups has grown: in 2017-18, the PSR for Black Africans and Black Caribbeans were, respectively, 11.7 and 51.8. Third, the model predicted that, among South Asians, although both Bangladeshis and Pakistanis were significantly more likely to be stopped than British Whites, the predicted stop rate for Pakistanis (22.3) was significantly higher than that for Bangladeshis (19.9). It is very plausible that the higher stop rate for Pakistanis relates to the regularity of Islamic terrorist attacks in E&W since July 2005. Fourth, in terms of mixed ethnicities, the model’s prediction was that White/Asians were significantly less likely to be stopped (6.1/1,000), White/Africans were as likely to be stopped (12.7), and White/Caribbeans were significantly more likely to be stopped (24) than British Whites (12.9).

Following the results for Areas in their entirety, the model presents results for the six largest police forces in E&W in terms of the PSR of the different ethnic groups. Of the five major non-White ethnic groups - Black African, Black Caribbean, Indian, Bangladeshi, and Pakistani - the model predicted that in every area: (1) Black Caribbeans were most likely, and Indians least likely, to be stopped, and (2) Black Africans were significantly less likely to be stopped than Black Caribbeans. The predicted experience of Bangladeshis compared to that of Pakistanis was mixed. In Greater Manchester, Merseyside, West Midlands, and West Yorkshire, Bangladeshis were predicted to be significantly less likely to be stopped than Pakistanis, but they were predicted to be significantly more likely to be stopped than Pakistanis in the Metropolitan and Thames Valley areas.

The results from estimating equation (5.5) are presented in a different form in Table 5.4 in terms of the stop experience of each of six ethnic groups - British White, Black African, Black Caribbean, Indian, Bangladeshi, and Pakistani - for the six Police Areas distinguished. Considered over the 16 ethnicities in their entirety, the most striking feature of these

Stop Rate

Marginal

Rate

Standard

Error

t -Value

All ethnicities

Greater Manchester

11.6

-23.1s'*

0.7

-33.1

Merseyside

24.9

-9.8 s*

1.0

-9.6

Metropolitan Police [R|

34.6

Thames Valley

9.8

-24.8 s*

0.7

-33.9

West Midlands

6.8

-27.8 s'*

0.7

-39.0

West Yorkshire

14.2

-20.4 s*

0.8

-27.2

British White

Greater Manchester

11.1

-20.1s*

0.8

-25.4

Merseyside

26.3

-4.9 s*

0.9

-5.2

Metropolitan Police [R|

31.2

Thames Valley

8.8

-22.4s*

0.8

-26.6

West Midlands

6.0

-25.2 s'*

0.8

-30.2

West Yorkshire

14.3

-16.9s*

0.8

-20.0

Black African

Greater Manchester

17.7

-45.1s*

4.5

-10.0

Merseyside

14.8

-48.0 s*

9.4

-5.1

Metropolitan Police [R|

62.8

Thames Valley

13.2

-49.5 s*

4.6

-10.7

West Midlands

10.8

-52.0s*

4.1

-12.7

West Yorkshire

8.1

-54.6 s*

6.0

-9.2

Black Caribbean

Greater Manchester

54.9

-64.8s*

7.1

-9.2

Merseyside

79.7

-40.0s*

20.3

-2.0

Metropolitan Police [R|

119.6

Thames Valley

39.8

-79.9 s*

6.8

-11.7

West Midlands

26.6

-93.1s*

3.6

-25.8

West Yorkshire

49.1

-70.5 s'*

7.5

-9.4

Indian

Greater Manchester

5.1

-17.1s*

4.2

-4.1

Merseyside

4.8

-17.3**

10.4

-1.7

Metropolitan Police [R|

22.2

Thames Valley

5.4

-16.8s*

3.7

-4.6

West Midlands

6.8

-15.4s*

2.5

-6.2

West Yorkshire

8.7

-13.5s*

4.2

-3.2

Bangladeshi

Greater Manchester

8.4

-55.4 s'*

5.3

-10.4

Stop Rate

Marginal

Rate

Standard

Error

t -Value

Merseyside

9.6

-54.2**

19.0

-2.9

Metropolitan Police [R|

63.8

Thames Valley

25.7

-38.1**

10.6

-3.6

West Midlands

6.5

-57.3**

4.6

-12.5

West Yorkshire

10.1

-53.7**

7.9

-6.8

Pakistani

Greater Manchester

15.7

-27.2**

3.2

-8.5

Merseyside

15.6

-27.3

18.2

-1.5

Metropolitan Police [R|

42.9

Thames Valley

22.7

-20.2**

3.9

-5.2

West Midlands

18.1

-24.8**

2.8

-8.8

West Yorkshire

21.4

-21.6**

2.9

-7.5

Source: Own calculations from www.ethnicity-facts-figures.service.gov.uk/crime-justice-and the-law/policing/stop-and-search/latest.

** Marginal probability significant at 5% level; * Marginal probability significant at 10% level.

results in Table 5.4 is the high PSR in the Metropolitan Area compared to the other areas. The overall PSR for the different areas, arranged in descending order, were Metropolitan (34.6), Merseyside (24.9), West Yorkshire (14.2), Greater Manchester (11.6), Thames Valley (9.8), and West Midlands (6.8), with all the PSR being significantly different from each other. While each ethnicity, in terms of the predicted stop rate, came off worst in the Metropolitan area, the experience of the ethnicities differed across the areas. The highest to lowest ratio of predicted stop rates was 5.2 for British Whites (31.2 in Metropolitan and 6 in West Midlands), 5.8 for Black Africans,

4.5 for Black Caribbeans, 4 for Indians, 9.8 for Bangladeshis, and 1.9 for Pakistanis.

 
<<   CONTENTS   >>

Related topics