Interview with Miles Hewstone


April 24, 2009

Miles Hewstone
is Professor of Social Psychology and Fellow of New College, University of Oxford.

For further information click here.


What does diversity mean to you by way of your work and field of expertise?

Well, let me start with social psychology and question myself. If I were to do a literature search, what would I find on diversity? I would definitely find some work evaluating diversity training – some very practical work there looking at whether it works, or not. You would also find some work linking diversity in small groups, and also in organizations, to creativity. This is an interesting question of whether having a more diverse body can lead to greater creativity. This links up with some work on group brainstorming, and also on minority influence. Contrary to what most people think, group brainstorming doesn't lead people to be more creative than they are as individuals, but the groups do better when they are made up of heterogeneous than homogeneous individuals. So that mixes perspectives. If you like, a mix of identities or knowledge background, which is diversity of another type, tends to be associated with creativity.

We also have been influenced in social psychology by some classic sociological work by Rosabeth Moss Kanter on percentages of minority and majority groups. She wrote about work forces originally. She referred to skewed versus tilted distributions of majority and minority members. You might have, for example, an engineering firm with a vast majority of male employees, and a tiny minority of women. And then within those categories you also have some solo or token females. So you might have a workgroup with mostly men, and a ‘token’ female; the impact of diversity in these different settings is very interesting to look at. For example, can you effect change of stereotypes by having a single female in this group; and the question is, if she is seen as just a ‘solo’ or maybe an exception, or perhaps chosen for a very specific reason, probably the assumption is that she will not have that much impact on stereotypes. So that's certainly a literature in which the d-word, diversity, would be used.

Of course, most social psychologists of my type are interested in group relations in general and diversity is just one aspect of that. But I think this possibly merges the first question with other questions, where I think to myself: What does diversity have to offer us? What does the study of diversity have about it that is interesting? And one of the major things – that is theoretical and it's also empirical – is we tend to work in a paradigm with a single in-group and a single out-group. So it's an approach to intergroup relations as if they were dichotomous. And this simplifies matters greatly, makes it much easier to do laboratory experiments. It makes it much easier to run statistical analyses. But of course it's a gross simplification of real-world intergroup relations. And I think when you study diversity it forces you to take a broader theoretical approach and it forces you to take on the complexities of data analyses with a multi-group structure.

So diversity means more than a simple distinction between two groups?

Exactly. So we've just recently been doing some field research in Malaysia. You have three different groups there, which are referred to, locally, as ‘racial’ groups. You have Malays, Chinese and Indians. This is just a huge addition in complexity to the way you have to think about the intergroup structure and how you analyze it. Diversity is also investigated in social psychology by research on factors that are associated with the salience of identities. So, does a female feel that her gender identity is more salient when she is the sole female in a group with five males, or when there are three females and three males? And people have looked at these kinds of studies with a variety of different groups, including race groups and gender groups and so on. So diversity can be used there very specifically as a contextual variable, which, at least in experiments, can be manipulated.

Is ‘diversity’ just a Zeitgeist term – a post-multiculturalism policy catch phrase - as in ‘integration and diversity’ policy - a corporate tool - as in ‘diversity management’ - or can it be a concept that can help structure and advance social scientific analysis?

At the moment, I don't see it carrying a whole lot of baggage. By this I mean there is the stuff that we talked about at the meeting, diversity training, and now the consultants are in there and so on, to help you achieve a more diverse organization, get your diverse organization to work effectively and so on. But it links up with what I was just saying. I think about moving from the two-group to the multi-group structure. So I think when you engage with the issue of diversity, you have to acknowledge that the two-group structure is too simple. And actually thinking about diversity also has had some impact, I think, on our research, not so much on research methods. There are actually some other variables that we would study. Traditionally, for example, people have studied stereotypes and they've studied them in a straightforward sense of being central tendencies. Members of group X tend to be quite Y. But more recently people introduced the notion of dispersion, which is the idea that group members vary around that central tendency. So you are immediately forced to acknowledge the diversity of members in any group. Even if you have a contentious issue like a discussion of whether males have better spatial memory than females, or ethnic groups differ in terms of their IQ. Central in any debate like this is to think not just about central tendencies but about the distribution around each central tendency. So it makes a big difference whether you stereotype a group in a particular way but you see the degree of diversity in the group as being very narrow or being very broad, because this affects how confident you can be that any member of the group will have those traits. So this is a sort of dependent measure aspect of diversity, I think. So for me personally the word ‘diversity’ doesn't have this baggage, it just makes me think of a wide variety of research. I don't see it so much shaping the agenda in my discipline, social psychology, but I do see us responding to some of those issues that are being thrown up and undoubtedly I suppose it will lead to some developments of our theory. Whether that'll be accidental or deliberate, I'm not sure.

At MPI-MMG, we are looking to develop research and theory spanning contemporary immigration societies – especially in Europe - and longstanding multi-ethnic and multi-religious societies, such as South-Africa, India and Malaysia. How do you see the concept of 'diversity' shaping this agenda - or not?

Again, probably not. I think Steve said at this meeting that this Max Planck Institute wants to avoid a narrow focus, a focus exclusively on ethnicity. For me as a social psychologist, diversity means breaking up these homogeneous perceptions of groups, recognizing that in any kind of group, for example chosen in terms of ethnicity, there will be variation in terms of class, age, gender, sexual orientation. So we have a diversity of subcategories that we have to think about as well. I totally agree with the idea that we should not get caught up explicitly on the notion of ethnicity.

From your perspective, from your expertise, what are a few of the key empirical, theoretical or methodological challenges currently facing 'diversity'-related research?

We encountered lots of them in writing our grant together. If you're interested in a notion like identity or contact you have many more potential questions to ask once you go out of the comfort zone of the two-group structure. So you have to ask potentially about contacts with a multiplicity of the other groups and then you will also have to think about the multiple identities of the person that you're interacting with. For example, are you interacting with someone as a male or a female, or as a member of this country or that country of origin, or selling you something in a bazaar, or are you indeed interacting with them, you being the purchaser and them being the seller in terms of market roles? So I definitely am fully convinced by the notion that we'll have multiple identities. And these are switched on, we would say they become situationally salient as a function of what we're doing, where we're doing it, with whom we're doing it. And of course the key factor in all that is who else is present and what group memberships do they belong to.

So these are all challenges that make our job as a researcher much more difficult and specifically in our own project. I think a big challenge there, which at the moment I don't know enough about but I need to know more about, is the technique of multilevel modeling. It's no good anymore just to collect a sample, however big the sample is, and just analyze it. We have to start to look for the interesting subgroups within. If you look at schools for example: each school should be seen as, potentially at least and statistically by necessity, having its own micro-culture or ‘context’. Or looking at classes within the school, some of which may have a high mix, some of which have a lower mix of students belonging to different groups, and undoubtedly we are individuals nested within categories, nested within other groupings. So we have to develop our expertise methodologically in dealing with these kinds of things. Take an idea like ‘intergroup contact’, which I'm especially interested in. We have had a primarily dyadic approach there. So you ask an individual whether he or she has had degrees of contact with members of another group. But I think the whole notion of diversity is going to make all that much more complicated. Lots of people are now spending more and more time talking about networks, which we traditionally haven't done in the discipline of social psychology. So again we have to explore new tools of doing that.

Thank you very much.

Interviewer: Sören Petermann

Go to Editor View