Diversity in research teams: more = better ?

By Joern Fischer

Anyone working in teams knows that different people contribute in different ways. We’re all different, and working in a team means appreciating these differences and making the most of them. But there’s something implicit in this that I have often struggled with – the assumption that diversity is necessarily good, while homogeneity is necessarily bad.

Some of you may have had “experts” work with their teams to look at different working styles. When you do this, a commonly used tool is the Myers-Briggs personality test (a shortened version is available here). This recognizes that people can be more introverted or extroverted; more intuitive or sensing; more thinking or feeling; and more perceiving or judging. In combination, you find 16 different stereotypes of people, constructed by four axes of two poles each. These people, so the logic goes, function in different ways, and therefore can contribute different things. (Of course individual people are often very much a mix of stereotypes.)

From this, the logic often goes on to say that good teams need diversity; that we ought to appreciate our differences, and that by appreciating them, we get better at working together. So far so good – but there is an underlying assumption in this that more diversity is necessarily better.

My personal experience and opinion is that, in research teams, this is not necessarily true, or only to a point. Rather, I would argue for the “right” kind of diversity, balanced with the “right” kind of homogeneity. What do I mean by this?

To me, diversity is valuable in that it brings different skills and perspectives to the table. Someone may be more inclined to have an intuitive sense about a hot research topic, and someone else may be good at organizing fieldwork, or crunching statistics after data collection. Indeed, to this point, I agree that a certain level of diversity is useful, because different people in a team can complement one another.

But I think the reverse – that diversity is good, and therefore homogeneity is bad – is not necessarily true. Instead, I would argue that a certain level of homogeneity is actually extremely valuable! When personal differences within a team become very large, it may become increasingly difficult to bridge working styles. In a research context, in particular, output-oriented (J-type) people often clash with process-oriented (P-type) people. Getting these types to be truly happy in a collaboration is very difficult, and sometimes, I would argue simply a waste of time (showing I’m a J-type, I guess…). Similarly, most academics are N-type people, because this is the kind of natural inclination that is associated with wanting to generalize and see the big picture. And last but not least, many scientists are introverted. It’s no surprise that one nickname for an INTJ type is “the scientist” – this particular combination of natural inclinations is very common in academia.

Is it true then that such people “need” others to make up for their weaknesses? Frankly, in an academic context, I don’t necessarily think so. Multiple people who have similar personalities and working styles can still complement one another by knowing about different topics, or for example having a background in different disciplines. Differences in training, backgrounds and expertise, in turn, can be quite readily overcome if the personalities fit together. In contrast, highly diverse, interdisciplinary teams with a lot of differences in both personality types and disciplinary backgrounds, in my experience, suffer from too much diversity. Not only are their disciplinary divides to be bridged but also working styles, including very basic questions as to whether we’re here to have a nice discussion or to get a paper published.

Finally, there’s one more dimension to all this that I have not yet mentioned, and that is the dimension of a shared goal or general normative direction. If things become so diverse that the goal is no longer the same among team members, things can really fall apart in research teams.

So, yes, we can learn from diversity. But too much diversity in a research team, to my mind, does not lead to better insights, but instead can get in the way of achieving any insights at all, and indeed, foster frustration instead. Hence, I’d argue for balance in research teams, not for diversity per se.

11 thoughts on “Diversity in research teams: more = better ?

  1. I agree – synergies are created by a diversity of capabilities relevant to the goal, not by diversity per se. I guess the same holds for BEF relationships in ecology. It seems unlikely that one can find one diversity metric that works for all, or even a large range of goals, for either humans or ecology.

  2. As with all such tests they are extremely weak as they stem from middle class, interlectual perseptions based on a modal, reductionist base.First it is obvious that diversity is superior to a restricted team base – common quote:” variety is the spice of life” Dosn’t sount so good but means the same thing. Secondly because the people who set tests have little any real experience of life outside a certain social educational background they faill to a address pluralisitc (unforseen factors) and like the kaleidoscope just one change and the whole picture changes. I tried this with your test and with only the slightest change to two questions scored significntly diffent results – results that could alter the direction of one’d life.

    It should be recalled that experience (outside the saftey zone) will trump apparent theoretical äcademic”wisdom”nearly everytime. See Dave Snowden’s Cynefin Decision Making Frame work and the difference between the Compicated and Complex Domain.

    Thank you for the work that you are doing, the platform you provide and your patience. I believe after 36 years in African Development from being in complete isolation hundereds of kms from civilisation. to working in slums, for NGOs, Remedial Schools, Universities, working for Aid Agencies, and Ministerial Committees that one must speak out and ADD to the debate – what is needed is what Sir Ken Robinson termed “Divergent Thinking” not merely placiidl concurrence – thus we will engender Sustainability. Your curtisy appreciated

    • Thanks Michael, I appreciate your view — and I like your kicking our universal butts to think outside the box 🙂
      Good luck with your work — Joern

      • Dear Joern , Sorry I alway do that always push the “enter “before completion. . Thank you for your broadmindedness.I’m just 70 but still active in communities, research (a little) lecturing and the develpment of governance at policy level.

        I attach a reflection, based and on many years and has been on the social media, hopefully you might enjoy!. Kind regards – MikeAVARICE OR TWO EARS OF CORN

        Dear Friends, over the years (40+ working) I have had an eclectic life from builder’s labourer, and bus conductor to consultant and educationalist. I am particularly interested in “development issues” or whatever they call it this week “upliftment” “sustainable livelihoods” etc. Most of the time I have been in the role of mentoring people, from which of course you always receive much more than you give, and invariably it is the poorest who are the most generous of spirit and have the greatest intellect. For as humanity claws up the ladder of apparent success, so does it become more “covertness” respecting only those from whom further advancement might be received. Slowly they become narrower and less human, dried husks, self-centred dull and uninteresting.

        Recently I had the excitement of doing some development work which I hope may contribute eventually to the improvement of somebody’s lot and about the same time attend a policy meeting. For a change I was very impressed by the commitment of the notaries present. And yet with all the splendour and good intentions of the occasion I recall something that a man, much better than I thought and then put pen to paper.

        The person in question is a Rolland Bunch who’s work with aging memory I am about to paraphrase. His book is probably the only really good book on how to conduct community development – it is called “Two Ears of Corn”.

        The work if I recall was conducted by World Neighbours in Latin America (with apologies if my memory is playing ticks) and he notes that “people who write book don’t do, and people who do don’t (have the time) to write books but they had decided to put down our experiences.”

        In the opening of the book I seem to recall that he quotes Johnathan Swift who wrote the book “Gulliver’s Travels” and here again I rely on fading memory, my volume of his work long misplaced with constant moving, Bunch starts his book quoting Johnathan Swift and here again I paraphrase slightly:

        “If I have passed this way but once and planted but two ears of corn, then I have done more for the promotion of humanity that a plethora of politicians”

        I have thought a lot about this of late, and hope that even if not in exact words, that certainly in the spirit I have encapsulated intention of Mr Bunch’s work. Where I wonder have I progressed to in three score years and nine, have I really done anything, I hope so, certainly I have not made any money? Am I slipping into the abyss (is that the correct spelling?) of clawing my way up the ladder of society; the retribution for which mindlessness is now beginning to rapidly manifest itself in the form of climate change, which I suspect is mother nature’s form of punishment for our endless and needless greed.

        Perhaps in these troubled times its appropriate for us all to take stock and try and make the difference before it is too late

        Go well

  3. Hi Joern,

    I looked a fair bit at this when I was doing my PhD, and found research indicated a mid-range of diversity seemed best for interdisciplinary teams. Sources are getting a bit old now, but here’s a relevant excerpt from my dissertation: “Teams must be carefully engineered. According to Barjak (2006), a curvilinear relationship exists between diversity and performance; teams work best when only two (small teams) or three fields (large) are being combined at most, and when only 20-25% of team members did their doctorate in another country. Melin (2000, p. 36) similarly notes that his interviewees “want their [collaborator’s] minds to think differently [to their own], but still work along the same lines in order to understand each other”.

    Barjak, F. 2006, Team diversity and research collaboration in life sciences teams: Does a combination of research cultures pay off? Series A: Discussion Paper 2006-W02, University of Applied Sciences, Switzerland.
    Melin, G. 2000, Pragmatism and self-organisation: research collaboration on the individual level. Research Policy, 29, pp. 31-40.


    • Great — thanks Kate! Very interesting and good to get something more substantial on this than my meandering experience 🙂
      Take care — J.

  4. It’s interesting to think of this from the perspective of a different “outcome”: building communities, connection, and interpersonal understanding. I’d say within the larger society we wish to affect, there is of course a much, much larger diversity than exists within our research groups. Insofar as we wish our work to speak across different groups, and aid in informing a society far more diverse than our own groups may be, it may be that incorporating more diversity into our groups and spending more time figuring out how to deal with it would make us more effective at the bigger outputs we wish to achieve. That is, social change informed by science and practice.

    I’d be *very* curious about measuring diversity vs. *those* outputs, Joern 😉 And of course, these outputs can only be measured at longer time scales, and with less certainty. But it is quite possible that if one adds a third output to “including very basic questions as to whether we’re here to have a nice discussion or to get a paper published,” which is, “effectively building communities for change,” the answers on “how much diversity” might be different. And if we add a fourth output, which is, providing opportunities, exposure, and pedagogy as broadly as possible to interested individuals, that may shift the answer as well.

    I actually agree to an extent with your specific point, that “highly diverse, interdisciplinary teams with a lot of differences in both personality types and disciplinary backgrounds, in my experience, suffer from too much diversity.” By the same token, these kinds of groups may be the most likely to ask “Why are we here, anyway? And are we sure it is to produce papers?” and in that way may be more likely to generate fundamentally different ideas–which itself can be a positive or negative.

    I suppose I see a lack of time dimension here as my core qualm. I think it is likely just unworkable to have too many kinds of diversity at the same time in the same place with too many people. But the same is not necessarily true, to my mind, over a long period of time: incorporating different elements of diversity as time goes on and understanding and ability to deal with diversity (hopefully) goes up. So one may not have all forms of diversity at all moments, but over time, I would hope “more diversity” is better, in that we continue to expose ourselves to and learn to work with more and more different kinds of personalities and disciplines.

    • Hi Jahi — excuse the late reply. I think time is critical, but also scope. Not all of us, all the time, solve all problems. To do our job well, we need to remain functional. I realise this is the same argument that, if you reduce it, is why some people argue that their disciplinary research is “just fine”, so … yes, I see the point. Ultimately, we need a large amount of diversity “in the boat” to bring about social change. But even then, I think it’s not about getting everyone and everything on board, but rather creating a critical mass of a significant minority. And this critical minority, indeed, needs to be well-enough aligned within itself.

      So, I guess the more concrete and short-term the output is one seeks, the more important it is to have a team that is functional straight away — and this means, a certain level of shared goals at least, is likely to be important in many research teams.



  5. I think what is needed is a shared epistemology – ideas about what can be known, what is good science, what the goal of science is/should be. For the rest, divergent working styles are probably not a big problem when people are good communicators, able to explain what they want from their team-mates.

  6. I think that requiring a shared epistemology is potentially quite problematic in the context of integration, and risks such projects missing lots of richness and nuance that come with qualitative methods. I feel that conflicting ideas about what constitutes evidence is, in part, what leads to so much (quant) social science being done by junior members of teams of life scientists. I get asked to review a lot of the resulting papers, and I have noticed a few red flags for such work: http://katesherren.org/index.php/2016/09/20/when-to-call-a-social-scientist-or-how-to-fool-one/
    Research following different epistemologies is more likely, I feel, to reach different conclusions than those following one. In a complex world, that diversity would bring more strength to the overall outcomes than a falsely coherent one.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s