Discrimination
I warmly welcome Professor John Curtice from the National Centre for Social Research, who is here to talk about the centre's report on attitudes towards discrimination in Scotland. Professor Curtice will take us through a presentation; members have a copy of the slides in front of them.
Professor John Curtice (National Centre for Social Research):
I offer apologies from my co-author, Catherine Bromley. We discovered that she had another engagement this morning, from which she could not free herself, in connection with a survey that she is conducting.
The project was very much a collaborative effort. It started with discussions between the National Centre for Social Research, the Equal Opportunities Commission, the Commission for Racial Equality, the Disability Rights Commission and Stonewall Scotland. We gradually developed a project idea and then presented it to the Scottish Executive, which graciously funded most, although not all, of the work—a couple of the commissions put up some of the money. We included 40 questions on the 2002 Scottish social attitudes survey, which means that our research is based on interviews with just over 1,600 people who were interviewed in the summer of 2002.
The Scottish social attitudes survey is an annual survey that was started and is in a sense owned by the National Centre for Social Research and particularly the National Centre for Social Research Scotland, which has its own offices in Edinburgh. Its inspiration is the British social attitudes survey, with which members might be familiar. The British social attitudes survey, which started in 1983, has two principal objectives. One is to facilitate the academic study of public opinion and how it changes. The second is to provide policy-relevant information about attitudes. Our assumption is that, although public policy should not necessarily simply follow public opinion, it needs to be informed by an understanding of public opinion. We therefore cover a range of topics of social and political interest.
When devolution was on the horizon in 1999, we thought that, given that the number of people in Scotland interviewed by the British social attitudes survey was in line with Scotland's proportion of the British population, the sample sizes would be too small for us to talk about Scotland in particular. Moreover, of course, the British survey would not focus on topics of particular interest in Scotland. For those reasons, we started the Scottish social attitudes survey as a similar annual survey with similar objectives.
The project was designed to examine the extent of discriminatory attitudes with respect to women, disabled people, ethnic minorities and gay men and lesbians—the four groups on which the partner organisations are focused. It was also designed to investigate three areas. The first was how much Scots believe that there are discriminatory attitudes in Scotland. The second was the extent and character of discriminatory attitudes, in so far as they exist in Scotland. The third was in a sense the most academic part of the project—trying to understand why people hold discriminatory attitudes, including the extent to which the reasons why people hold discriminatory attitudes towards each of the four groups are similar.
Obviously, to study so-called discriminatory attitudes, it is necessary to have a definition of what a discriminatory attitude is. The next slide shows the definition that we applied in the research. Essentially, it is the belief that certain social groups should not be involved in something that we would regard as something that most people in society—if not all of society—would be expected to do. That is what, colloquially, we were trying to get at. Another way of putting it is to say that a discriminatory attitude is the belief that certain forms of social exclusion should occur.
It is crucial to understand that the study is of attitudes, not of experience. Some members will have read the press reports about the extent to which people from English backgrounds feel that they have been discriminated against in Scotland. We are not into that game. We are not into the game of measuring the extent to which people from ethnic minorities or women feel that they are discriminated against. We are trying to tap and understand attitudes that may be regarded as discriminatory according to the definition that we have applied. I will briefly go through the broad answers to the three questions that we tried to address in the research.
The first question concerned how much discrimination people thought existed in Scotland. Most people thought that there was some discrimination, but not everybody thought that there was a great deal of it. The second headline was that some groups were thought to be more likely to be the subject of discrimination than others. As the chart shows, the degree of perceived discrimination is less as we move from left to right. The biggest bar, on the left-hand side, is for ethnic minorities. Ethnic minorities comprised the group that was thought to be most likely to be subjected to prejudice. In contrast, women comprised the group that was thought to be least likely to be subjected to prejudice.
However, if we put the furthest-left bar and the one next to it together—which gives us the figure for the number of people who thought that there was a great deal, quite a lot or some discrimination—we see that most people thought that some discrimination existed in respect of all the groups. Another way of saying that is indicated in the next slide, which shows that around 50 per cent of people thought that all four groups experience a little, quite a lot or a great deal of prejudice. Perhaps more striking is the fact that only 5 per cent of people thought that none of the four groups experienced prejudice in Scottish society. Nevertheless, people did not think that there was a great deal of prejudice. Only one in 10 people thought that all four groups suffered a great deal or quite a lot of prejudice.
Certainly, prejudice is thought to exist and the extent of it varies according to the group in question. However, it is not necessarily thought to be great. People seemed to think that it existed to some degree, but they did not all think that it was extensive. The next slide shows some of the groups that were thought to be most likely to think that discriminatory attitudes exist: younger people, people living in urban areas, women and people with high levels of education.
Let us turn to the second question. The next slide shows us the simple, summary headline measure of what we found. The question was a general one that was not tied to any specific group, unlike the other questions in the survey. We asked people which of two views came closest to their own. The first was: "Scotland should do all it can to get rid of prejudice". The second was: "Sometimes good reason to be prejudiced" against a particular group.
As you can see, just over a quarter of people gave the answer that we would regard as being an indication of a discriminatory viewpoint. Two thirds of people felt that all kinds of prejudice should be got rid of. In a sense, that is typical of a lot of the figures that I will show. Prejudicial or discriminatory attitudes exist. They are minority viewpoints, but they are held by a not insubstantial number of people.
Just as we were interested in differences in the degree to which the various different groups were thought to be subjected to prejudice, we were equally interested in finding out towards which of those groups discriminatory attitudes were most common. In order to do that, we tried, where possible, to ask exactly the same question in relation to all four groups.
The next slide, which illustrates that kind of question, is headed "People's preferred kind of MSP". We asked people whether they would prefer to have a gay or lesbian as an MSP, or somebody who is not gay or lesbian, or whether that would not make any difference. Most people said that it would not make any difference but, as the slide shows, nearly one in five respondents said that they would prefer not to have somebody who was openly gay or lesbian as their MSP. In contrast, the proportion of people who did not want a female MSP or somebody who was disabled as their MSP was much lower, at 4 per cent. As we saw with an earlier question, gays and lesbians and black people and Asians appear to be more likely to be the subject of discriminatory viewpoints than either women or disabled people.
The survey dealt with the degree of discriminatory attitudes with respect to relationships and family life. Some of those matters are currently the subject of public policy proposals and other developments. The next slide shows the proportion of people who agree and disagree with various statements. The first is that it is the man's job to earn the money and the woman's job to stay at home and look after the kids. The second question is about the degree to which people would feel unhappy if a relative of theirs were to marry somebody of a different ethnic or racial background. The third question is whether marriage between gays should be allowed.
The big black bar on each of the graphs represents the element that disagrees with the discriminatory viewpoint. Disagreement with the discriminatory viewpoint is the more common but, in each case, some people share that viewpoint. Again, it is on the issue concerning gay men and lesbians that attitudes are most divided and discriminatory attitudes are most common.
The survey examines potential sources of discrimination with respect to employment and the labour market. A similar story emerges on the following slide, which shows that, although more people agree than disagree that people from ethnic minorities provide Scotland with much-needed skills, 18 per cent disagree. As far as ethnic minorities are concerned, discriminatory attitudes appear to be most common with regard to questions of employment.
Another question that allowed us to compare groups was who would make a suitable primary school teacher. That enabled us to get at the question of the gender stereotyping of occupations. Members can see from the next slide that gender stereotyping is still alive and well, in that, whereas 67 per cent of people thought that women were very suitable for becoming primary school teachers, only 41 per cent thought that men were. About a quarter of respondents thought that women were more suitable for that post than men were. The slide also shows that there is not a high level of perception that either wheelchair users or gay men and lesbians are necessarily suitable as primary school teachers.
We considered the degree to which people felt that existing public policy towards the four groups was or was not going far enough in trying to reduce discrimination. The hierarchy of groups shown in previous slides is replicated, the important consequence of which is that, although gay men and lesbians are thought to be relatively likely to be subjected to discrimination and prejudicial viewpoints, and although we know that the incidence of discriminatory attitudes towards gay men and lesbians is relatively high, that is also the group for which support for equal opportunities policy to go further is lowest. Only 26 per cent of people think that equal opportunities for gay men and lesbians have not gone far enough. In contrast, that figure is 58 per cent for disabled people.
The next slide is headed "% who say a great deal/a lot of prejudice exists". Although the people who think that more should be done for the various groups are least likely to hold prejudicial viewpoints, around 22 per cent of the people who think that discrimination exists in respect of gay men and lesbians think that equal opportunities for that group have gone too far.
In the third section of the report, we try in fairly extensive detail to disentangle the relative importance of three possible explanations of why discriminatory attitudes exist where they do. The first explanation is sociological and essentially claims that discriminatory attitudes are the product of different positions in the social structure, different experiences of the life course and social influences to which people have been subjected. For example, we would expect younger people who have been brought up in a society in which discrimination towards ethnic minorities has been for the most part socially less acceptable to be less likely to hold discriminatory views in respect of ethnic minorities than older people who have had a different experience.
The second possible explanation centres on a competition for resources. For example, someone might hold discriminatory views because they feel that they are competing for economic resources or jobs in the labour market with various groups that are different from them.
The third explanation for the existence of discriminatory viewpoints is a social-psychological one. People have identities. When they start to feel that others are different from them and that they do not have much in common with those groups, there is a sense of social-psychological difference and distance.
I should point out that those explanations are not completely contradictory. For example, it is perfectly possible to believe that different social experiences will lead to different psychological outcomes. However, the statistical analysis in the report indicates that the explanations are to some degree independent.
Members might be asking, "How the hell do we disentangle the various possible explanations?" First, I should make it clear that we did not ask the respondents to choose which of the three models explained their attitudes. Instead, we tried to explain those attitudes by asking a range of questions that tapped different social positions, economic outlooks and social-psychological outlooks. As the subsequent charts in my presentation show, we then said that, if for example age mattered, we would find that older people would be more likely than younger people to hold discriminatory viewpoints. We could use that as an indication that different social experiences by age are crucial.
As the report contains a large number of potential indicators of the three models, I will simply focus on those that appear to have most value and analytic purchase. In the sociological model, the indicators are primarily age and education; in the economic model, the indicators are differences by income and subjective perceptions of whether people are coping on their income; and in the social-psychological model, we asked people how much they felt they had in common with gay men and lesbians, how much men and women have in common with each other and so on and whether the respondents would prefer to live in an area where there were different kinds of people or where people were much the same as one another—again, we were trying to tap the underlying social-psychological orientation.
The subsequent charts in my presentation give members a taste of the evidence that we received. Although I do not propose to discuss the charts in detail, I will give an indication of how to read them. For example, on the right-hand side of the chart headed "Sociological 1", we examine the differences in attitudes towards same-sex relationships and find out whether people feel that those relationships are always wrong. First, we compared people according to their educational background and then compared the attitudes of younger and older people. Members will see that only 12 per cent of people who have a university degree believe that same-sex relationships are always wrong, whereas 41 per cent of people who do not have any educational qualifications take the view that such relationships are always wrong. We would therefore conclude initially that educational background appears to make a difference in that respect and that the more highly educated someone is, the less likely they are to hold discriminatory views.
There is a similar and even starker difference with respect to age. Among younger people—those aged 18 to 24—only 16 per cent think that same-sex relationships are always wrong. By contrast, 51 per cent of those aged over 65 take that view. There is clearly a substantial generational difference in attitudes to that question.
We are basically looking at the size of the differences in the various charts. Just skating through the charts, we can see that there are similar—although not such big ones—differences in the degree to which people think that women should be homemakers. Interestingly, which is perhaps the exception that proves the rule, the question of whether a wheelchair user would be suitable as a primary teacher is one of the rare examples in which the general rule, that better-educated people and younger people are less likely to hold a discriminatory point of view, is broken. In this case, younger people are less likely to think that a wheelchair user would be a suitable primary teacher.
On the economic question, I shall pick out an example that, unsurprisingly, shows that economic situation makes a difference. It is the question of whether ethnic minorities take jobs away from people who are already in Scotland. Of the people who feel that they live comfortably on their income, only 16 per cent take that view, but among those who find it difficult to live on their income, 34 per cent take that view.
On the chart headed "Psychological - 1", we can see some quite large differences, in particular with respect to attitudes towards homosexuality. Members will notice that, for example, people who say that they would like to live with similar kinds of people are much less likely to regard same-sex relationships as acceptable. People who would like to live in an area where there are the same kind of people as them are much less tolerant of homosexual relationships than are those who are happy to live in an area where there are different kinds of people. We can see that those psychological differences exist to some degree in all four of the examples shown on the charts, but they are weakest—perhaps unsurprisingly—with respect to disabled people.
If we compare the two psychological charts, our general conclusion is that, just as discriminatory attitudes are most common in respect of gay men and lesbians and least common in respect of disabled people, with ethnic minorities coming somewhere in the middle, the same pattern is equally true of the differences that we found. For the most part, we found that different sections of Scottish society are most divided in their views on homosexuality and least divided on their views on disabled people. To that degree at least, the position of the various groups is rather different from one another.
I would like to give the committee some idea of how we concluded that it is the social-psychological explanation that seems to provide the most immediate explanation of discriminatory attitudes, where they exist. The chart entitled "Comparing Explanations" illustrates that by bringing together some of the information on same-sex relationships from earlier charts. The bars on the left hand side compare younger with older people, the bars in the middle compare people on low incomes with those on high incomes, and the bars on the right hand side compare people who feel that they have a lot in common with gay men and lesbians with those who do not.
I want members to notice that the differences are largest with respect to age and psychology and that those differences are bigger than the differences with respect to economic background. For the most part, the economic explanation appears to be the least important, but it is often quite difficult to disentangle the relative importance of social backgrounds such as age or education from social-psychological measures.
The underlying measure that I introduced to members at the beginning of my presentation was prejudice. Sixty-eight per cent of people said that we should get rid of all kinds of prejudice and 26 per cent said that it is sometimes okay to be prejudiced. The next chart demonstrates that for the most part that question provides us with an indication of underlying prejudicial attitudes. In other words—surprise, surprise—those who said that there was sometimes good reason to be prejudiced were most likely to take a discriminatory viewpoint on the questions of which I have given members a taste. For example, 60 per cent of those who think that there is sometimes good reason to be prejudiced think that same-sex relationships are always wrong. That figure falls to 34 per cent among those who think that we should get rid of all kinds of prejudice. As well as giving us an indication of underlying discriminatory attitudes, the question indicates the degree to which discriminatory attitudes vary by group.
If, as in the chart headed "Who says prejudice OK?" we examine the degree to which groups differ on the question, we find that social-psychological questions seem particularly to identify those who think that prejudice is okay and those who do not. Of those who prefer to live in an area with different kinds of people, only 15 per cent think that there is sometimes good reason to be prejudiced. Among those who like to live in an area with similar kinds of people, the figure is 36 per cent. The other types of explanation also show differences, but for the most part they are smaller.
The conclusion that we reach on the third question is that we cannot discard any of the explanations—all are valuable in certain situations—but that the social-psychological explanation appears to be the most important in explaining underlying discriminatory viewpoints. For example, when people think that they have something in common with a group—that a group is similar to them—they are less likely to hold a discriminatory viewpoint. Equally, if people are happy about living in a society in which there is diversity and with people who are different from them they are less likely to hold a discriminatory viewpoint.
We came to the conclusion that if the committee is interested in reducing the incidence of discriminatory attitudes two strategies are potentially open to you. Those strategies are not necessarily contradictory. The first is to encourage people to believe that groups have something in common and are an integral part of Scottish society. The second is to say that it is perfectly acceptable in our society for people to be different. Both strategies may be of interest to those who are interested in changing the picture that we have tried to paint for you in our report.
My colleagues, who are all female MSPs, would like to ask you some questions.
I will not comment on whether that is evidence of discrimination.
They have got me.
Clearly, the survey was about attitudes. The results suggest that in respect of attitudes to discrimination against disabled people and to some extent women, the battle for hearts and minds is on the way to being won. However, they also suggest that there is a need for more targeted work on the psychology of discrimination in general and on attitudes towards black and minority ethnic communities and lesbian, gay, bisexual and transsexual communities in particular. What follow-up work—research and action—would you advise in respect of the psychology of discrimination in general? What would you advise in respect of ethnic minority groups and the gay community?
That is a good question. It may be true that women and disabled people are not necessarily subject to discriminatory attitudes, but that does not mean that they are not subject to discrimination.
One of the things to be aware of is that, in effect, we are considering attitudes that might result in behaviour that is directly discriminatory—although that needs to be unpacked. There is also evidence that discrimination occurs indirectly, through organisational practices. Our research says nothing about the degree to which any of the groups in question are subject to indirect discrimination that might not even be intended and might not be related to attitudes.
With that caveat, I will take your question head on. There are two things that one might want to do. First, a possible follow-up would be to go down a qualitative route. In a sense, we were painting a picture of what appeared to be the incidence of discriminatory attitudes and what appeared to underlie them. If one wanted to engage in a campaign or an activity that was designed to reduce discrimination, it would be helpful to identify a group of people who, in various ways, were willing to articulate discriminatory viewpoints. Focus groups could be organised to analyse those people in greater detail and to try to work with them. Various messages could be put out to deal with their viewpoints. That would be closer to testing activity in that it would be designed to try and change attitudes, if that was the aim. That is probably what one wants to do.
Secondly, if one is interested in changing attitudes and one regards it as an aim of public policy to reduce the existence of such attitudes, we would say that those questions should be revisited every three or four years or so to test the degree to which attitudes are changing. That would be an argument for saying that we should replicate such research two or three years down the track, to assess the degree to which attitudes are changing.
In the past week or two, we have been discussing mainstreaming equalities quite a lot; in fact, we did that this morning in the context of our consideration of the Executive's budget. Given that the indication is that further work needs to be done in relation to ethnic minority groups and the lesbian, gay, bisexual and transgender groups in society, would you go so far as to say that, as well as come back in three or four years to assess whether attitudes have changed, we should have an indication of the use of the Executive's budget for those groups? If nothing has been done in policy terms in the three or four intervening years, why would attitudes change, other than in as much as some older people will no longer be around? If nothing is changed, attitudes will not change.
If you are asking whether an implication of our research is that attention needs to be focused particularly on ethnic minorities and on gay men and lesbians, rather than on women and people with disabilities, the answer is that it is clear that that is a potential implication of the research as far as attitudes—though not necessarily discriminatory practices—are concerned.
The material that I have presented today and some of the material that is buried deep in the report suggest not only that the incidence of discriminatory attitudes to ethnic minorities and to gay men and lesbians is higher, but that our questions, which were designed to tap discriminatory attitudes, did so in a much clearer way with those groups than they did with women and the disabled.
Although it would undoubtedly be preferable if the attitudes in the answers to some of the questions that we asked about women and disabled people did not exist, those attitudes did not seem to be derived from a clear belief that certain kinds of prejudice were okay. They might have been derived from other things, such as a failure to appreciate that it might be possible for someone in a wheelchair to be an effective primary teacher, as it should be perfectly possible to organise a school to allow that to happen. It might be desirable to make people aware that it is possible for disabled people to achieve things, but that is rather different from saying that a perception exists that those people are different and should therefore be excluded.
In other words, the report provides evidence that, with ethnic minorities and gay men and lesbians, we are looking at prejudice. We found people who said that those groups were different to them and that they should be treated differently. Although, as I have said, there is some evidence of gender stereotyping, it is more the case that, with women and disabled people, there are questions about the potential of the people in those groups and a failure to understand their potential.
The other aspect is that if you decide to change policy in order to change attitudes, you must have information about people's needs and wants and what they are currently coping with.
Page 8 of the report talks about the need for a survey of Scotland's minority ethnic communities and mentions the fact that a scoping study has been published. What are the needs within those communities? Are you aware of any similar initiatives relating to considering the attitudes and experiences of the LGBT community? Earlier, we heard about the need for disaggregated data across all the equality strands. Practically every form that one fills in asks whether one is a man or a woman, but no one is going to ask a member of the LGBT community who is not out to fill in a form that asks who they are or what they need. Similarly, it is possible that some people who are disabled might not want to say how disabled they are.
I do not want to deny the value of what you are saying, but I point out that what you suggest involves considering the question from the other side. Our research was into discriminatory attitudes in Scotland in general—by implication, the attitudes of the majority community—towards those groups. As you noted, it was easy for us to differentiate between men and women in the survey. We also asked some questions that were designed to find out whether the respondents or their family members were disabled. Furthermore, we asked a standard question that allowed us to identify respondents from ethnic minorities.
You might have noticed that we did not ask whether anyone was gay or lesbian. That was a quite deliberate decision based on the fact that not everyone would want to answer that question, as you said. We felt that there was a danger that the research would suffer if that question were asked. Meanwhile, as you will be aware, given the size of the ethnic minority population in Scotland, it is inevitable that a general sample of the population will contain too few of that group.
If you want to consider the question from the other side, you are—I presume—interested in the degree to which members of those groups feel that they are subject to explicitly discriminatory attitudes that might be articulated to them, or the extent to which they are experiencing discrimination. If you were to turn the research around, that would be the focus of the study.
The Scottish Executive has been conducting a number of scoping studies to try to work out how to approach a study of Scotland's ethnic minorities. However, there is a basic problem, which is that because the ethnic minority population is relatively small, it is not that easy to sample. Also, there is no list of all the members of ethnic minorities in Scotland.
There are substantial methodological challenges to coming up with a representative sample of ethnic minorities. The techniques that have been used in Britain essentially rely on the fact that, for the most part, ethnic minorities tend to live in geographic concentrations. That allows strategies to be employed that allow those areas to be sampled.
In Scotland, however, that becomes more difficult because there are, with one obvious exception, no locations where there is a heavy concentration of ethnic minorities. That is not to deny that the work would be valuable, but one must be aware that a study of ethnic minorities in Scotland will be a considerable methodological challenge.
When it comes to lesbians and gay people, we come back to the question of whether it will be possible to do a study. Are members of the population who fall into those categories going to feel comfortable, able and willing to participate in survey research of that kind? How are we to ensure that we have a representative sample? Both of those questions represent significant challenges.
The results suggest that there is a hierarchy of discrimination with ethnic minorities and LGBT communities at the wrong end of that hierarchy. As that does not appear to reflect the development of equalities legislation, what does that tell us about what we are doing to tackle prejudice?
Can you unpack what you mean?
Much discrimination is against people from ethnic minorities, although for some time there have been statutory reasons why people should not discriminate against people because of their race. That does not seem to have brought about the required shift in public attitudes.
I will interrupt you there. You should be aware that the survey provides a snapshot—it does not show how attitudes have changed over time. It shows that discriminatory attitudes vis-à-vis ethnic minorities and gay men and lesbians are more common than they are vis-à-vis women and disabled people. It does not show whether discriminatory attitudes vis-à-vis ethnic minorities and gay men and lesbians are more or less common than was the case 20 years ago.
We know from other research, particularly the British social attitudes surveys, that attitudes towards homosexuality have changed. Twenty years ago, the majority of people would have said that homosexuality is always wrong. We have now reached a point where Scottish society is divided down the middle on the subject.
For example, on the questions in the survey about gay men and lesbians being parents, or on the question of gay marriage, the split is not a million miles away from 50:50. That is arguably why those issues are now becoming issues of public policy. It is no longer the case that society clearly believes that such things should not happen and that they are wrong; but equally we are not at the point where society believes that such things should happen, if that is the point that we are heading for. We are at the point where the matter has become an issue.
As far as ethnic minorities are concerned, the evidence is that explicitly articulated racial prejudice is less common now than it was 20 years ago, although there might still be more of it than we want. We cannot draw the conclusion that public policy has been ineffective: the evidence indicates that having public policy that says that discrimination against ethnic minorities is wrong might help to persuade opinion, although a lot of other things have happened during the past 20 years. However, it will not necessarily eliminate that discrimination entirely, which is arguably true of all law. If all that we had to do was pass a law to stop people doing things, we would never put anyone in prison, so I am not sure that we should be terribly surprised.
The survey makes fascinating reading, and I am sure that there is a huge amount of follow-up work to do.
Have you had any responses or feedback on the results from the equalities groups that collaborated in the research? It would be useful to know the extent to which the attitudes that you found tally with their experiences.
I am pleased to say that our relationships with the representatives of those groups are as good now as they were when they started the research. [Laughter.] When I talk to them, they tell me that they are pleased that the research has been done, although not all of the results came as a surprise—I would not pretend to suggest that they did. On the other hand, it provides a firm base of evidence.
They are all aware of the research and all made their own statements when it was published last week. I invite the committee to talk to them about what they think the implications are. They have all welcomed the research and have said what they think should follow as a result of it.
In your report you state that the questions do not cover bisexuals and the transgender community—that you limited your questions to cover gay men and lesbians—because you doubted whether people would necessarily understand those terms or have enough knowledge about them to be able to answer questions on them. How confident are you that the respondents had enough knowledge about gay men and lesbians, or indeed any of the other groups, to be able to answer questions on them?
It depends what you think is the requirement in respect of knowledge. We felt that if we use the terms "gay men" or "lesbian" most people know what we are talking about. Our concern was that if we were to talk about transsexual and transgender, the meaning of the terms might not be immediately obvious to a significant proportion of respondents.
One of the things that we had to be careful about in the research was to ensure that we used language that was understandable to the general population, but which was not at the same time potentially offensive to various groups. There is clearly a potential tension because there is undoubtedly a language that people who work in equal opportunities use; it is sometimes relatively technical language that will not necessarily be understood by the wider community. That is one of the issues that we had constantly to consider and be careful about.
The judgment that we made—it is fair to say that the representative of Stonewall who was on our steering group agreed—was, rightly or wrongly, that we could ask about gay men and lesbians and we would get answers; people understood the question, so to that degree at least they had knowledge. I am, of course, not saying that the attitudes that were expressed are necessarily grounded in a full understanding of the psychology or whatever of homosexuality: I am simply saying that if we use certain terms people know what we are talking about and will have an attitude towards that group.
I have a similar question about the term "disabled" and whether that was taken to apply only to wheelchair users. When you used the term, to what extent did people show understanding of disability in general?
You pick up an important point. For the most part we tried to ask questions using the generic term, but when it came to the question about a primary teacher the obvious danger that we faced was that if we asked, "Is a disabled person suitable to be a primary school teacher?" we would get the response, "Well, it depends on the disability." That is why in that case we had to be more specific.
We did not dream up those questions in the corner of a room then launch them on to the world without testing them. The Scottish social attitudes survey goes through an extensive process of testing, which includes conducting two rounds of piloting. We get our interviewers to go out and interview about 40 or 50 people. We ask the interviewers to administer drafts of the questionnaire—often a much longer draft than that which we intend to use—and the interviewers get back to us about how many blank faces they got in response, and whether people said that they did not understand certain questions or showed hesitancy. We use that kind of test to establish whether people appear to understand our questions. This module, like all our modules, went through crafting, recrafting and re-recrafting, but at the end of the day we are as confident as we can reasonably be that they are questions that the general population in Scotland can understand.
I have two questions on women. The survey seems to suggest that people generally feel that women do not suffer a great deal of discrimination, but there is still clear evidence of gender stereotyping and there is recognition of a need to help women into work. Is that a mixed message, or am I not understanding the subtlety of the results?
It is a mixed message, and I think that you understand perfectly.
A small proportion of people now take the strong gender-stereotypical view that women should be in the home. One way of reading what we have come up with would be to say that the gender division of labour is perhaps no longer between the home and the labour market, but within certain sectors of the labour market. Clearly, the perception persists that certain kinds of jobs are still likely to be more suitable—whatever that means—for women than they are for men. It is not that many people think that women should stay at home; it is that we are not quite sure that we want to let them loose on all aspects of the labour market.
I have one more question. Forty years ago, I was a feminist student fighting for women's rights. We have not come very far since then. The gap between men's and women's pay is still about 19 per cent, and the Equal Opportunities Commission still handles around 24,000 discrimination cases a year. Relevant legislation has been in place for a significant length of time, so how would you suggest that we try to bridge the gap between the perception and reality?
I am not quite sure. You are right in a sense. Society is perhaps more sanguine about the position than it should be. The question that we asked people probably invited them to think about the degree to which there is overt, direct discrimination rather than about indirect discrimination. The pay gap exists. We could argue about whether that is direct or indirect discrimination, but it is, at least in part, the product of indirect discrimination.
What was interesting about the research was the degree to which older women were more likely than younger women to think that there is discrimination. That may reflect a difference in experience. We discovered that the objective position of younger women was rather different from that of older women and that the process is partly gradual. That may be part of what is going on. However, I am not going to dissent from the suggestion that one may need to persuade people that there is more discrimination going on than there actually is.
Good afternoon. I am not an academic. To a layperson, your sample of 1,600 people seems quite small in the context of the whole of Scotland. Perhaps you can comment on that. Do you feel that the survey results are robust enough to serve as an effective baseline against which to measure progress in the future? Do you know of any plans to carry out a repeat of the exercise in two or three years' time—or, as you suggested earlier, following whatever time gap might be appropriate?
It all rests on statistical theory. Statisticians tell us that if we take a random sample of 1,000 people in a population, as long as we avoid the various pitfalls of bias—of which there are many—if an attitude exists in the population which would give the true value of a yes response 50 per cent of the time and a no response 50 per cent of the time, 95 per cent of the time we will get the answer right within 3 per cent. In other words, if we kept taking a sample of 1,000 people and conducting the survey properly, 95 per cent of the time between 47 and 53 per of the responses would be a yes response. That gives you an idea of the degree of robustness of any of this kind of research.
Often, we are not interested in the attitudes of the whole of Scotland. We may be interested in the attitudes of men, women, ethnic minorities, and so on. As I have said, this survey is inadequate as a survey of ethnic minorities, as they are not sufficiently represented in it. We used a sample of 1,600 people rather than 1,000 people partly to try to ensure that we had at least reasonably large sub-groups for the more common groups in the population. That is also why we over-sampled rural parts of Scotland.
One of the clear lessons that we have learned is that in Scotland there is a particular policy interest in considering the position of rural people. For example, this project was particularly concerned with finding out whether there might be greater evidence of discriminatory attitudes in respect of gay men and women in rural areas. For the most part, the research does not substantiate that. There are some signs of such discrimination, but there is not as much evidence of it as we expected. We deliberately over-sampled in that way to ensure that rural Scotland could be considered on its own. We would like always to include more people in surveys, as that would mean that the sub-groups would have more people in them.
That gives you a partial answer to your second question. Certainly, in so far as measuring attitudes in Scotland in general is concerned, this is high-quality research that is conducted to pretty much the best standards that there are. I will give you an idea of the kind of efforts to which we go to ensure that. I do not mean to criticise what other organisations do—one gets what one pays for. Nonetheless, let us consider the opinion poll that was published in The Herald this morning, which was conducted by System 3. That is a standard, omnibus piece of survey research that was conducted over the course of a week. Inevitably, that kind of research is focused on those people whom one can manage to contact in the course of a week. The fieldwork for our survey started in June 2002 and continued through to October. We tried hard to get hold of somebody at all the selected addresses to ensure that the sample was as representative as possible. Even so, with the best will in the world, we can succeed in getting hold of only about 60 per cent of the people whom we would like to get hold of. We would like that figure to be higher.
The survey is about as solid a piece of evidence on attitudes as we can get. However, every last figure should not be read as being exact. If I say that the figure is 37 per cent, it means that it is between 34 and 40 per cent, but we are definitely not talking about 60 per cent. The 37 per cent figure provides a baseline against which to measure future progress.
On whether there are plans to repeat the survey, there are certainly no concrete plans to do so but, as the organisation that initiated the survey, we are keen to repeat it. Those with whom we dealt in the Executive are pleased with the research. I hope that we might be able to persuade them to repeat it in two or three years. If the committee were to suggest that that might be a good idea, it might give extra strength to our elbow.
We will consider it.
The paper suggests that this is the first time that a public attitude survey on discrimination has been carried out on this scale. Are there similar studies from other areas or countries? How could they help us to compare different policies?
Let me explain the basis of the paper's claim. We are not aware of a previous attempt to look across the range of groups. Previous research, including some with which we were involved, has tended to examine specific groups such as ethnic minorities, gay men and lesbians and women. The various commissions regularly do research that enables them to consider their particular area. For example, the CRE got MORI to do research last year on discrimination.
Our project is attractive to the various commissions because it compares across groups and ascertains the extent to which something is a common attitude or an attitude that varies according to the group. Only by pulling together information on different groups can we begin to address that question.
Those concerned with different groups do research on their own groups. For example, the British social attitudes survey has individual material on discrimination against ethnic minorities. We have covered other forms of discrimination in the Scottish social attitudes survey. For example, if you look at the 2001 survey, you will see that there is some material on religious discrimination and, this year, in collaboration with Professor Miller of the University of Glasgow—another aspect of whose research got some publicity yesterday—we are examining attitudes towards English-born people and people of Muslim background. Classically, such surveys are done on their own and they are not integrated within the report.
I have a final question. You have pulled together all the strands from the different equality groups. What key benefit can policy makers derive from the results of your survey?
The key benefit, as I suggested towards the conclusion of my presentation, is that the survey suggests that if you want to tap discriminatory attitudes and change them, the thing that you can most immediately get at is people's social psychology—their sense of difference and feeling comfortable with difference.
To elaborate, we could have said that once you have looked at people's educational background and age you can forget it, because there is no other form of explanation. Basically, we would have had to say to you, "Well, terribly sorry, but it is going to be a long process. You will have to ensure that more and more people go to university. You will have to wait for generational change to result in the replacement of attitudes, but eventually, in 30 or 40 or 50 years' time, attitudes will change." That process is going on. It is part of the process that explains the change in attitudes over the past 20 years, and it is likely to mean that attitudes will continue to change over the next 20 years.
However, we are also able to say that, even once you take into account people's education, age and various other influences—and there is some detailed statistical research in the report that shows this—psychological orientation does make a difference. For example, the "One Scotland. Many Cultures" campaign is clearly designed to encourage people to think that it is okay for there to be different kinds of people in Scotland, because that is part of Scottish society. It is clearly tapping into at least one aspect of the psychology that we point out. However, we indicate that there is another potential approach—which is summed up in a phrase of which I am aware because it comes from the area of learning difficulties in which my wife works—and that is "the same as you". That phrase picks up another way of saying to people that there is no need to be prejudiced in the way that they appear to be.
You addressed many of the points that I was going to raise. I will focus on education from two sides. The results seem to say that the better educated you are, the less likely you are to have discriminatory attitudes. Do you have any thoughts on why that might be? Is it to do with intelligence, is it to do with being in a diverse institution where mixing is going on, or is it to do with the content of education, such as citizenship? Have you given any thought to that?
Yes. The finding that education is related to discriminatory attitudes fits into a wider body of evidence that people who are better educated tend generally to be more liberal in a whole range of areas. They are more likely to hold the underlying attitude that it is okay for people to do whatever they want to do, and it is up to people to decide for themselves what their own social and moral frameworks are. Of the various arguments that have been put as to why that finding exists, the one that I argue for is that more advanced forms of education require you to take on board different arguments.
Students are required to be aware that there are different theories out there, to be able to critically evaluate them, and to be able to engage in debate about them. The crucial thing is not intelligence, and it is not necessarily the content of the education, because in some senses the content is irrelevant. The issue is the diversity of viewpoints and the diversity of content. It is about learning that you have to be able to engage in dialogue and that you have to be able to critically evaluate all points of view. That is the underlying part of the educational process that probably results in people coming to the view that, "Well, of course there are different religions, there are different ethnic backgrounds, and there are different sexual preferences, but that is part of diversity." That is what people have learned to deal with through the educational process.
Specifically on education, do you think that there is a need for targeted education and training to combat some of the attitudes that are held just now, working alongside some of the adverts that you mentioned? Where would we start with such an approach? I want to pick up on something that Margaret Smith said and something that you said about it taking a long time for attitudes to change. I am a wee bit concerned that as people get older their attitudes change as well. It might not just be a matter of waiting. Do we have to tackle children's attitudes? Zero Tolerance ran its "Respect" project through pilots and it is now talking to the Executive about rolling it out, because it made a big difference to the children whom it targeted.
There is no doubt that if we want to influence attitudes, the educational process is one way of doing so. That is a slightly different argument from the one that I gave you. I gave you the argument that, if we expose people to a diversity of views, they come to an understanding that all views are arguably wrong, because all views are partial truths and we should not therefore try to exclude people.
You are suggesting something slightly different—and I am not disagreeing with you. You are saying that within the wider schooling process we should try to encourage people explicitly to recognise that there are people of different ethnic and racial backgrounds and sexual orientations and in some cases say explicitly that that is okay. That needs to be underpinned by various forms of educational content so that people have an understanding of what are common social and cultural practices within a Pakistani or other ethnic community.
I do not disagree with you, but it is also true that that sort of process is slow. If you want to change attitudes within a reasonable period, you will not want simply to target younger people who are going through the formal education process.
I have been reading recent work on racial prejudice. Although we can see the generational processes at work, we can see that older people have also changed their attitudes. Older people are not immune to changes in the wider climate in society. That is why I would say that, although education is, of course, a good place to start, I would not confine myself to it.
I am not sure whether you have touched on the role of the media, apart from the Executive's advertising campaign. Did you see any sign of media influence? If you conducted an attitude survey on ethnic minorities both before a media campaign against asylum seekers and again after such a campaign, I assume that you would see differences in attitudes. Did you investigate how the media affects people's attitudes at all? Obviously legislation affects attitudes to an extent, but to what extent does the media affect attitudes?
The answer to your question is no, and you gave the reason for that in your question. If we want to understand the impact of the media on attitudes, we have to have an over-time research design and ideally the same people have to be interviewed over time. I would not even bother to run the analysis in our survey. In our survey we asked people what newspaper they read. If you wanted to, you could examine the relationship between attitudes and which newspaper they read. I have no idea whether there are differences, because I have not bothered to look. That is because even if I were to find out that readers of newspaper X are more likely to hold a discriminatory viewpoint, I am still left with the conundrum whether they hold those viewpoints because they read that newspaper, or whether they read that newspaper because they already hold those viewpoints. That cannot be unpacked with the sort of cross-sectional research design that we used.
Thank you very much, Professor Curtice. What you said was very interesting. I have asked the clerks to produce a paper on today's discussion to consider what process members would like to follow.
Thank you very much for your time and attention.