If not diversity training, then what?

It can be frustrating reading the literature on implicit bias and wondering what to do about it. It’s one thing to know it exists and quite another to know how to effectively combat bias. As someone who regularly chairs gatekeeping committees of various sorts (from admissions to appointments) I struggle with this one. For sure, I talk to committee members about the issues. Our university’s equity guide is pretty good on these points as well. But I was disheartened to read a Boston Globe story that diversity training has little effect on bias. The Boston Globe story, “Who’s Still Biased?” is here.

Here’s a piece of their story: “Now a few social scientists are taking a hard look at these programs, and, so far, what they’re finding is that there’s little evidence that diversity training works. A paper published last year by the psychologist Elizabeth Levy Paluck of Princeton University’s
Woodrow Wilson School and the Yale University political scientist Donald Green comprehensively surveyed the literature on prejudice reduction measures and found no empirical support for the idea that diversity training programs change attitudes or behavior. Similarly, a 2008 literature review paper by Carol Kulik of the University of South Australia and Loriann Roberson of Columbia University found that, on the question of changing behavior, there were few trustworthy studies – and decidedly mixed results among those. And research by a team of sociologists on more than 800 companies over three decades has found that the best diversity training programs make little difference in who gets hired and promoted, and many programs actually decrease the number of women and minorities in management.”

There were some dissenting voices in the story who argued that effects can be subtle and that small changes can add up and make a difference over time. What do you think? Does standard issue diversity training make a difference when it comes to bias? What concrete steps can we take to combat bias in settings where we can’t simply evaluate without knowing candidates’ gender?

13 thoughts on “If not diversity training, then what?

  1. This recent episode of Talk of the Nation featured Elizabeth Levy Paluck. The contrast between her empirically-informed perspective and that of the other guest, someone from a diversity training service, was interesting.

  2. Really interesting and important. The most important bits, though, seem to me buried– e.g. that programmes emphasising lawsuits are the least effective but most popular.

  3. I found the Boston Globe article both interesting and worrying to read. And I’ll declare an interest, in that one of the things I do is “deliver” what one might call “training” on topics which are loosely “diversity”-related.

    So first, I’m sceptical about the generalised nature of the claims (at least, as represented in the BG article – I haven’t yet listened to Talk of the Nation, or read the published paper, though the topic is one I’ve read various other published papers about). Jender’s point is that there are variations between the content of different training programmes, and I’d add that the design of the learning activities and the competence of the trainer are also crucial.

    And that, of course, is true of any topic that can be taught – I mean, some people are *really* bad at teaching philosophy!

    But there is something additional here, which has to do with affective or attitudinal learning outcomes.

    First, I think it’s just harder to design learning activities which *both* respect the learners as autonomous and reflective people *and* also encourage them and enable them to think about their emotions and attitudes in ways which might result in change. It takes a lot of courage, because you have to provide rich material for them to wrestle with, and you have to be honest about what you’re trying to do (otherwise I think you’re failing to respect them properly). I’m not sure I manage it successfully on a regular basis, and my experience of attending other “diversity training” sessions is that other people really struggle with it, too.

    Second, I think the emphasis on “diversity” is unhelpful. It immediately encourages people to focus on (a) other people, rather than on themselves, and (b) on differences rather than similarities. Trainers (in my experience) often fall back on exercises which make people feel comfortable by allowing them to laugh about silly stereotypes. They may then go on to complicate and disrupt those stereotypes, but I don’t think that’s what then sticks in the mind.

    So I think it’s useful – at least to start with – to get people thinking about themselves, about how they form their own beliefs and attitudes, and what they think influences those things. (So we might look at social attitude surveys which show correlations between socioeconomic class, newspaper reading habits and particular sets of attitudes/beliefs, and talk about how and where we look for evidence to support or disconfirm those attitudes/beliefs. And we’ll look at implicit bias stuff.)

    In that context, subsequent discussion can be seen as trying to expand people’s options in deciding what to think, rather than on trying to impose one particular correct thing to think. That does, of course, depend on the trainer accepting that people can then change their minds in all sorts of directions, and not only the approved one. I don’t know how many “diversity trainers” are willing to accept that.

    Writing about this has made me realise how much my approach to this sort of work has been informed by my experience of teaching philosophy…

    Anyway, the reason I found the article worrying is that it could provide an easy excuse for institutions which never wanted to invest in diversity training in the first place to stop doing so. I mean, instead of making the training better, or at least, seeing if it’s possible to make the training better. But just to be clear, I’m still glad it’s been published – I wouldn’t want to pretend things were working if they weren’t.

  4. Paluck’s conclusion seems to be different from the article’s. She emphasizes that the research into the programs is inadequate, so we just don’t know what’s working.

    A lot of programs are mandated; all state agencies in my state have to have them, and this includes the public universities. My university is quite up front that its program is designed to meet the requirement without any regard to whether it works or not. It is unlikely to make much difference, by the way; it consists in online training and tests, and colleagues send around lists of the correct answers.

    It’s also the case that it is very, very hard to change people’s behavior.

  5. I’m thinking the claim “diversity training doesn’t work” makes about as much sense as “sex education doesn’t work”. You might well make that claim, looking at sex education in the US overall. But that would be because of the recent dominance of abstinence-only education, so very misleading about the prospects for educating about sex.

  6. My only experience with diversity training has been as a learner. At my institution, much of such training is compulsory. As a result, I have sat in workshops where some people were visibly irate about having to be there, and ready to quarrel all the way through with anything that was “rammed down their throats”.

    Yet at the same time, to some extent, equity issues have become somewhat normalized, as the faculty has become more diverse and as the collective agreement has made equity provisions non-negotiable.

    The one thing that worries me now is a kind of backsliding, resulting from the (false) impression that all gender, racial, sexual orientation, and ability issues have been resolved.

  7. There is an interesting article in the current issue of “Yes Magazine” on What white people fear. The author basically says that diversity training has little impact because us white folks aren’t questioning the underlying power structures that benefit us. Instead of diversity training, he suggests figuring out ways that we can all work together on dismantling these power structures. It’s not quite clear to me exactly how he suggests doing that but his point is interesting, especially when viewed together with the research on implicit bias. There are some powerful structural forces at play that a little diversity training can hardly undermine.

  8. Remember Guys ~ These are learned behaviors…and there is positive reinforcement for maintenance of those behaviors. Consider our current keyboard style. Now consider that someone switched just 3 of the keys around – and this switch becomes your new “suggested” reality. How difficult would that be…and you are trying your danmest to learn the new sequence! In that analogy, I am talking about technical issues – a keyboard. Technical shifts are more easily tolerated than social and emotional changes. We do technical shifts all the time. Social, emotional, and cultural shifts take a long time, and lots of exposure. Here you are dealing with age and levels of malleability. You are dealing with re-teaching learned behaviors that have supported a population for generations. Consider (very briefly) that you are standing in the middle of a business that has, arguably, been built and established on the positive reinforcement of the very behaviors that you are now suggestng must be re-visited, re-viewed, re-learned. …and you expect to do that in two days? You cannot potty-train your child in two days! Of course it does not work. Here is what you do for me: The minute you can learn a new language in two days – – – come talk to me. It is necessary, however, to keep those at the extreme left of the bell curve, engaged. It will change, and our country will need you. …but not in a two-day seminar.

  9. Dr AF, Thanks for visiting the blog; I’ve just bookmarked your blog.
    You are right that this will take a long time. Improvement may in some ways be getting harder because so many people believe the problems are really over.

    I liked your comment about finding out whether your chancellor is a manager or a leader. I think one thing that’s also making things difficult is that the levels below the chancellor are getting filled with people who are not even good managers, though they may have a great deal of power.

  10. One of the problems is that diversity _training_ should be diversity _education_ which I have found is highly effective. I have campaigned for this distinction because I think it is a significant one. In my view, education is far more effective as consciousness shifting than is training. A second issue is whether the diversity workshops are done as a contribution to a more inclusive human community or as a way of increasing profits. I presented a paper at a diversity conference nearly 10 years ago arguing that the latter does little to shift the community values but the former one has a lot of potential to change our world for the better.

  11. Great question. I think diversity training is only as effective as a company’s commitment to making it work. Companies that “get it” about any type of training tend to develop a plan to address the issue and then commit to providing training opportunities over time. Behaviors don’t change after one exposure to any given concept, it takes ongoing and deliberate reinforcement of the new behaviors.

    It would be interesting to see what types of impacts ongoing diversity training has in companies who offer it continually to all their staff, beginning with their leadership. The company’s leadership is frequently the reason for the success or demise of any training initiative. If you have sustained buy-in from the top, it tends to magnify and strengthen the efficacy of the training program.

    Take care,


Comments are closed.