Executive Summary
As part of the preparations for the 2011 Census, the Scottish Government commissioned Ipsos MORI to undertake cognitive question testing on a provisional set of questions on Country of Birth, Religion, National Identity and Ethnicity.
The overall aim of the research was to provide evidence on how the different questions were interpreted by respondents and whether the questions elicited the required information.
Eighty-eight cognitive interviews were conducted with a broad range of respondents in terms of ethnic background, religion, age, sex and living in different parts of Scotland.
1) Country of Birth
This question was relatively unproblematic. There were occasional problems with the write-in box not being long enough for countries with particularly long names. Most people understood the impracticality of having a longer list of response options and were happy with the options provided.
The question should be retained in its current format.
2) Religion
The main problem with this question was the layout. Several respondents interpreted the "Other Christian" write-in box as a visual cue that signalled the end of the question and wrote their religion (e.g. Hindu) in that box. This resulted in several respondents answering this question more than once. Consideration should be given to amending the layout of this question to avoid the visual break caused by the "Other Christian" write-in boxes in the middle of the response options.
Consideration should also be given to whether there is a data requirement to specify other Christian churches or whether simply having an "Other Christian" option (with no write-in boxes) would be sufficient.
Most people understood why it might be useful to collect data on religion and were happy with the inclusion of the question. However, a few people thought the question was being asked in connection with terrorism and, specifically, to find out the number of Muslims in Scotland.
Most people noticed that the question was asking about "current" religion rather than religion of upbringing but a few did not. Consideration should be given to underlining, or otherwise emphasising, the word "current".
Respondents were generally happy with the response options, although there was some confusion from people who had not lived in Scotland for long about whether "Church of Scotland" was the same as "Church of England". The inclusion of a "Church of England" response option should be considered.
There was some concern that the write-in boxes were too short. Consideration should be given to increasing the length of the write-in boxes (e.g. by having two rows of boxes).
3) National Identity
Consideration should be given to not including this question.
Most people understood the meaning of the question, however there were some misinterpretations. The word "identify" caused some ambiguity. There was also some confusion about the word "nation."
Many respondents did not notice the instruction to "Tick all boxes that apply" and felt uncomfortable having to choose one nation. If the question is retained, the instruction should be made clearer.
Several respondents, and British respondents in particular, thought that the purpose of the question was to explore their views on Scottish devolution and independence.
There was also evidence that a social desirability effect might bias the results of this question: there was a view that immigrants should try to integrate and therefore the "correct" answer was Scotland and/or Britain.
As the only specified response options were all in the UK and the Republic of Ireland, some respondents inferred that there was less interest in "other" nations.
The write-in box was frequently described as too short, so respondents could not write-in more than one "other" nation. For some, this seemed to imply that the expected answer was from one or more UK countries, or UK countries and one other . If the question is retained, the length of the write-in box should be increased.
Some respondents felt the question was important as it meant that they could express their "Scottishness" or "Britishness". If the question is deleted, the recommendation to include "Scottish" and "British" in the Asian ethnic group response options becomes even more important.
4) Ethnicity question
The main problem with this question was the layout. Many respondents initially thought that the White section constituted the entire ethnicity question because they interpreted the "Other white ethnic group" write-in space as a visual cue that signalled the end of the question and wrote their ethnicity (e.g. Black African) in that space.
Several versions of the questionnaire tried to correct this problem, mainly by increasing the indentation of the write-in spaces. The final version tested also put a letter in front of each section heading. However, not enough interviews with non-white respondents were conducted to fully assess the impact of the change. There appeared to be very little objection to lettering being used in relation to these headings. The layout and lettering used in Version 6 (see Annex 1) should be tested further with more non-white respondents.
A couple of respondents objected to the ordering of the sections in the question, feeling this should be done in alphabetical order as they were offended the "White" section was listed first. However, most non-white respondents were happy with the ordering and assumed it reflected the numbers of people from different ethnic groups in the Scottish population.
- Interpretation of the question
Most respondents seemed to understand the term "ethnic group". However, there were several respondents who struggled with the term. Common misinterpretations included the ethnic groups the respondent came into contact with or socialised with.
Respondents felt that the explanation included in the 2001 Census would have helped them to understand the question better. Consideration should be given to amending "ethnic group" to "ethnic background."
- Feelings towards the question
Several non-white respondents did not like answering this question. It made some of them feel uncomfortable and some did not understand why this information was important.
The respondents who understood why this data was being collected were more likely to accept the question. However, simply adding an explanation of why the question is included will not completely solve the problem, as it was clear that some non-white respondents, even those who were in favour of the question and understood its potential usefulness in planning services and tackling discrimination, at some level "hated" this question being asked. Consideration should be given to including an explanation about why it is necessary to collect information on ethnicity in the Census.
Virtually all respondents were comfortable with this section heading. The few respondents who objected were those that objected to colour labels throughout the form.
Scottish/British respondents sometimes found it difficult to choose between response options.
All of the Polish respondents interviewed were happy that a Polish response option was included - although a couple were unsure it should be included without recognising any other Eastern European countries. However, respondents from other Eastern European countries had mixed views towards the Polish option, with some feeling it should be removed as other countries were not represented, and others feeling it was good that the Polish had been recognised. We therefore suggest that the Polish response option should be retained.
- "Mixed or Multiple Ethnic Groups" section
Although we did not specifically recruit respondents from "Mixed or Multiple Ethnic Groups", several respondents were from this group or had children who were.
The term "Mixed" was viewed unfavourably by several respondents who believed the term was, or could be, offensive. A few respondents favoured simply "Multiple Ethnic Groups". However, some other respondents were not familiar with this term and indicated that the term "mixed" was needed to explain the section. We suggest the phrase "Mixed or Multiple Ethnic groups" should be retained for the time being. As people become more familiar with the term "multiple ethnic groups", it may be possible to phase out the word "Mixed."
There was confusion about who should be included in this section. One interpretation was that it was only for a mix of different White ethnicities. Another interpretation was that it was only for a mix of White plus another ethnicity (and not, for example, Asian and African). Both of these interpretations were based on the fact that the "Mixed or Multiple Ethnic Groups" section followed on from the "White" section. Consideration should be given to adding a short explanation of who should be included in this group.
It was also suggested that consideration should be given to moving this section so it does not appear immediately after the "White" section. However, previous research has shown that this section was likely to get missed if placed towards the end of the question.
Further confusion was based on the use of the terms "Asian Scottish, African Scottish and Caribbean Scottish" and "Asian British, African British and Caribbean British" in other sections. Several respondents believed that these terms described people from mixed or multiple ethnic backgrounds (e.g. one Asian parent and one Scottish parent) which made it unclear who the "Mixed or Multiple Ethnic Groups" section was intended for.
Several of the respondents also commented that the write-in box in this section was not long enough to describe multiple ethnic groups. This write-in box should be expanded so people feel able to write in an adequate description of their mixed or multiple ethnic groups.
Consideration should be given to conducting testing with more people from mixed or multiple ethnic groups.
- "Asian, Asian Scottish or Asian British" section
Respondents were generally happy about the phrasing of the heading, feeling it recognised the identity of Asians who are born and raised in the UK. We suggest that the phrasing of the section heading should therefore be retained. However, it is important to recognise that this may confuse some people from mixed or multiple ethnic groups.
Most respondents were happy with the response categories that were included in this section, feeling most groups of Asians in the UK had been included. However, several UK-born Asian respondents (and parents of UK-born Asians) found it difficult to accept that the response categories (unlike the heading) did not include any reference to Britain or Scotland. Consideration should be given to adding Scottish and British to the individual response options (e.g. Pakistani, Pakistani Scottish or Pakistani British.)
Most Sikh respondents ticked the Indian response option and believed that the Sikh response option should be removed. They felt that Sikhism was a religion rather than an ethnicity and they were confused about its inclusion as an ethnic group. Most had either not noticed the Sikh option and ticked Indian "as usual", or had noticed and been uncomfortable about having to chose between the two options. A couple of respondents had selected the Sikh option rather than the Indian option, but this was simply because they felt they were being more specific: they did not feel strongly about the inclusion of a Sikh option. We suggest that the Sikh response option should therefore be removed.
- "African, Caribbean or Black" section
The inclusion of the term "or Black" led to considerable confusion. Many respondents felt that the term "or Black" implied that Africans or Caribbeans were not Black, or were not necessarily Black. They wondered who the "or Black" people could be, or assumed that non-Black Africans or Caribbeans should be included. Most Black Africans and some White Africans therefore assumed that White Africans should be included in this section. Consideration should be given to changing the heading back to the 2001 Census wording of "Black, Black Scottish or Black British."
Black Africans and Black Caribbeans were confused about whether to tick the "African…." or "Caribbean…" response option or the "Black…" option. Most selected "African…" or "Caribbean…." rather than "Black….". This was because they felt it was more specific and they preferred to emphasise their area of origin rather than their skin colour. Consideration should be given to deleting the "Black, Black Scottish or Black British" response option.
In most cases, choice of response option did not stem from any objection to the word "Black…" and, indeed, many respondents had spontaneously described themselves as "Black African". However, there were a few respondents who objected to the term "Black".
The inclusion of the terms "Scottish" and "British" were generally seen as positive, but did lead to some confusion about whether this meant people from mixed or multiple ethnic groups should be included.