Jump to ContentJump to Main Navigation
The Political Integration of Ethnic Minorities in Britain$

Anthony F. Heath, Stephen D. Fisher, Gemma Rosenblatt, David Sanders, and Maria Sobolewska

Print publication date: 2013

Print ISBN-13: 9780199656639

Published to Oxford Scholarship Online: September 2013

DOI: 10.1093/acprof:oso/9780199656639.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2017. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in OSO for personal use (for details see http://www.oxfordscholarship.com/page/privacy-policy). Subscriber: null; date: 24 February 2017

(p.207) Appendix 1 The Ethnic Minority British Election Survey (EMBES)

(p.207) Appendix 1 The Ethnic Minority British Election Survey (EMBES)

Source:
The Political Integration of Ethnic Minorities in Britain
Publisher:
Oxford University Press

(p.207) Appendix 1

The Ethnic Minority British Election Survey (EMBES)

EMBES is the first nationally representative probability sample of the ethnic minority electorate in Britain since 1997, when a relatively small ‘booster’ sample was added to the main British Election Study (BES) post-election survey. The 2010 survey was much larger and was a ‘stand-alone’ one with a separate sampling frame and fieldwork. The two surveys have been closely coordinated, however, and around half the questions in the EMBES are exact replications of questions in the main post-election wave of the BES. Questions shared with the main BES included all the standard outcome variables, such as registration, turnout, party identification, vote choice, and so on, as well as explanatory variables covering valence issues, perceptions of party positions, and attitudes towards the main contemporary issues. In addition, there were questions in EMBES on ethnic and religious identities, social organization and mobilization, perceptions and experiences of discrimination, and attitudes towards a range of ethnic issues.

The original aim of the survey was to have a probability sample with more or less equal sample sizes (around 500) of the five established groups (as defined in the 2001 Census). The design was to be a clustered, stratified one using the Postcode Address File as the sampling frame, just as in the main BES, with areas of high ethnic minority density being over-sampled and areas with the lowest density (〈 2 per cent ethnic minorities) being excluded. We should note however that ethnic minority individuals from these lowest density areas were eligible to be sampled in the main BES. The primary sampling unit (PSU) for the EMBES was the Lower layer Super Output Area (LSOA) in England and Wales (Data Zones in Scotland) whereas for the main BES it was the ward. (Each LSOA contains around 1,500 people. LSOAs are designed to be more homogeneous with respect to population size than are wards.) In EMBES addresses in the selected PSUs were screened for the presence of ethnic minority individuals at the address, the number of addresses issued for screening varying inversely with the ethnic density of the PSU. Over 30,000 addresses were issued for screening in the 620 selected PSUs. To adjust for the over-sampling in higher density PSUs, descriptive statistics are routinely weighted throughout the book.

Respondents to EMBES were paid a conditional incentive of £20. Response rates were between 58 per cent and 66 per cent, depending on the method of treating ‘unknowns’ in the screening process. This compares very favourably with 1997, when the response (p.208) rate was 44 per cent. We should also note that in 1997, for cost reasons, areas where the ethnic density was 〈 10 per cent were excluded from the sampling frame. (For further details of the 1997 survey see Taylor and Thomson 1999, Heath and Saggar 1999.) The achieved sample size in 2010 was 2,787.

The questionnaire was administered face to face by CAPI, including a module which three-quarters of respondents completed themselves on the laptop. There was also a mailback questionnaire. Interviews were conducted in English but translators (aged over twelve) were allowed from within the household, and paper versions of the questionnaire in the main minority languages were provided for use by the household translator in such situations.

Following discussions with TNS-BMRB, our selected fieldwork agency, and agreement with the ESRC, some modifications were made to this initial design. It was agreed, in the interests of maximizing the effective sample size, to aim for larger samples (around 600) of the four larger groups with a smaller target sample of Bangladeshis. We also agreed that if, at the screening stage, a potential respondent from one of the five targeted groups was identified, but on subsequent interview the respondent self-identified as coming from a mixed or some other ethnic minority group, the interview would nonetheless be continued. This decision was taken partly for practical reasons, and partly because self-reported identities, especially for people from mixed backgrounds, are somewhat fluid. We also thought that people of mixed background are an important and growing group in their own right. Table 1 shows the breakdown by ethnic group, using the official census classification, together with the Office for National Statistics’ (ONS) mid-year estimates for 2007 (the latest available at the time of writing) of the population aged sixteen and over in England and Wales.

The ONS mid-year estimates shown in Table A1.1 suggest that ethnic minorities, including the Other White category, are now approaching 15 per cent of the population in England and Wales. They will, to be sure, be a somewhat lower proportion of the electorate, since many of the most recent arrivals will not be eligible to vote in a general election (and the proportion of ethnic minorities resident in Scotland is considerably lower than that in England and Wales). However, the great majority of the Indian, Pakistani, Bangladeshi, and Black Caribbean groups, and just under 90 per cent of the Black African group, are eligible to register as either British or Commonwealth citizens. The groups covered in EMBES, therefore, probably made up around 8 per cent of the eligible electorate in 2010. Recently released figures from the 2011 Census confirm that around 8 per cent of the resident population of England and Wales aged 18 and over belonged to the ethnic groups covered in EMBES. (p.209)

Table A1.1. The achieved sample sizes in BES, EMBES, and ONS mid-year estimates for 2007

frequencies

Cultural background

EMBES

BES

ONS (%)

White British

0

2761

84.7

Other White

0

44

4.9

Mixed—White/BC

70

7

0.3

Mixed—White/BA

23

8

0.2

Mixed—White/Asian

5

7

0.3

Mixed—White/Other

9

2

0.3

Indian

587

40

2.5

Pakistani

668

12

1.5

Bangladeshi

270

6

0.6

Other Asian

16

10

0.6

Black Caribbean

597

23

1.1

Black African

524

29

1.3

Other Black

6

7

0.2

Other ethnic group

11

27

1.6

Other answers

83

Refused

1

9

All

2787

3075

100.1

Notes: Unweighted frequencies for BES and EMBES. The figures in the third column are for residents aged sixteen and above. They do not include residents in Scotland.

Sources: 2010 BES post-election wave, EMBES 2010, and ONS mid-year estimates for 2007, table EE2