National Records of Scotland

Preserving the past, Recording the present, Informing the future

2001 Census Ethnicity Reports - Report on Coding of Any Mixed Background

2001 Census Ethnicity Reports - Report on Coding of Any Mixed Background

5. Findings

The result was to generate 18,396 descriptors for the 9,776 persons with text; an average of 1.88 descriptors per person. Appendix D provides a list of all descriptors with more than 20 occurrences.  Up to 6 descriptors were distilled for each person. Table 3 shows the number cross-tabulated by Section prefix.

Table 3: Person in Tick and Text database (T&T) by Section prefix and number of ethnicity descriptors

Section Prefix

Total

Number of ethnicity descriptors per person

0

1

2

3

4

5

6

Total

12,066

2,290

1,663

7,662

406

37

5

3

Section A White

420

21

190

188

19

2

-

-

Section B Mixed

8,805

1,309

866

6,359

257

14

-

-

Section C Asian

302

10

59

224

9

-

-

-

Section D Black

66

2

21

37

6

-

-

-

Section E Other

288

19

105

154

9

1

-

-

Ticks or text in more that one section

1,208

69

388

639

87

18

4

3

Ticks in more than one section but no text

859

858

1

-

-

-

-

-

Text in more than one section but no ticks

118

2

33

61

19

2

1

-

Turning to cases with only one descriptor, 866 (out of 1,663) were assigned codes indicating that the form-filler supplied information in Section B. There were also 422 persons with tick or text in more than one Section (last three rows of the table). This leaves 375 with information supplied in a single Section that was not Section B. An examination of these shows that 137 mentioned the ethnicity of one ancestor which was, presumably, different from the unmentioned ethnicity of other ancestors. Of the remainder of 238, 64 had entered text such as ‘Eurasian’ or ‘Coloured’ that may indicate mixed ethnic origin. For the remaining 174 cases, there was no indication of mixed ethnic origin. (These findings are not shown in a table in the main part of the report; they – and those of the following two paragraphs - are included in the summary given in Appendix D.)

The 8,113 cases with two or more descriptors were analysed to ascertain the extent to which a person’s descriptors straddled categories defined by the tick boxes on the form. Each descriptor was assigned to one of the 14 tick boxes, as if a person having ethnicity given by that descriptor alone was answering the ethnicity question. In general, a descriptor would be assigned in this way to a tick box other than box 5. Each person with two or more descriptors is thus assigned to one or more of the 14 tick box categories. In very few (115) cases did all the descriptors for a person correspond to a single tick box. Further, in most of these cases (65) the single tick box was 10 ‘Other Asian’ which still leaves scope for being of ‘Any Mixed background’.

When descriptors were assigned Sections rather than tick boxes, 367 persons might be considered as not being of ‘Any Mixed background’. Of these, 110 persons had all descriptors in Section A ‘White’, and 203 had all in Section C: ‘Asian’.

The tables in Appendix E summarise the information in this and the previous section about the validity of the coding of the 12,764 persons in the category of ‘Any Mixed background’. Table E1 shows that 11,021 of these cases (86%), were confirmed as having been correctly coded. The number of indeterminate cases corresponds to information from elsewhere about the number of records added by the One Number Census (ONC) process and, for persons enumerated, with the number with ethnicity imputed because missing. (Compared with ethnic groups generally, persons of ‘Any Mixed background’ were slightly more likely to have had their ethnicity imputed but slightly less likely to have been added by the ONC process.)

Table F1 in Appendix F shows the most common combination of first and second descriptor (with descriptors for each combination shown in alphabetical order). The most common combinations with over 200 cases each were ‘Black’/’White’, ‘African’/’Scottish’, ‘Iranian’/’Scottish’ and ‘Arab’/’Scottish’. Table F2 shows that the most common combination of ethnicity defined in terms of tick box categories with over 1,000 cases each were ‘Other-Asian’/’White-Scottish’ and ‘Other Ethnic-Group’/’White-Scottish’. Finally at the Section level, with over 1,000 cases each,  ‘Asian’/’White’, ‘Black’/’White’ and ‘Other’/’White’ were most common (Table F3).