While colleges and universities no longer formally prohibit women and people of color from joining the faculty, recent research shows that those groups continue to face institutional barriers. Slow progress on integrating the professoriate, moreover, is undermining confidence in the university as a meritocratic institution. Meanwhile, the paucity of diverse role models in the professoriate discourages women and people of color from pursuing careers in academia and undermines the academic confidence of students from those groups.
NIAS Fellow Frank Dobbin is Henry Ford II Professor of the Social Sciences at Harvard University. He and his team have studied annual federal data on the composition of faculty at U.S. universities and on individual career histories of over 10,000 faculty. They have furthermore surveyed 670 universities on the history of their hiring, promotion, diversity, and work-life programs and policies. With this data in hand, they are analyzing the effects of dozens of different policies on the professoriate.
Their analyses suggest that many of the most popular programs, such as diversity and harassment training and grievance processes, have null or negative effects, leading to decreases in faculty diversity. But other simple measures, such as targeted hiring, formal mentoring, diversity task forces, and work-life initiatives show robust positive effects on faculty diversity.
Research and presentation: Annick van Rinsum
Annick van Rinsum
Welcome to this podcast by NIAS, the Netherlands Institute for Advanced Study in the Humanities and Social Sciences. A place for researchers to find the time and space to fully pursue their curiosity and ask new questions. My name is Annick van Rinsum, and in this episode I’ll be speaking with sociologist Frank Dobbin. His research shows that a lot of the programmes in place to increase diversity in both corporations and in academia do not work or even have adverse effects as people in powerful positions often don’t consider themselves to be biassed, and explain what they see in the world as if the world is meritocratic and fair. Measures aimed at combating individual level bias or at punishing bad actors often do not protect those who are being harassed or discriminated against. What does work is to change the career experiences of those people, for example through mentorship programmes with live programmes, retention programmes and through ombuds office. Dear Frank, welcome in the NIAS library.
Frank Dobbin
Thanks very much. Nice to be here.
Annick van Rinsum
How has your experience at NIAS been?
Frank Dobbin
It has just been a fantastic, very generative few months. The scholars are extremely interesting, from all different places, very intellectually engaged. We have a very dynamic seminar and very interesting engaging lunch talks. It’s just it’s academic heaven. Over the course of my career, I’ve had quite a few sabbaticals and I’ve been at a number of institutes for Advanced Studies. In different places, sometimes in Europe, more of them in the US. Yes, and this group is just the most diverse. There are a number of people from the global south, which is unusual. It’s been great to be out of the United States and in the context where I can talk to people from lots of different countries about the project I’m now working on, which is “Diversity programmes in universities that are designed to diversify the faculty”.
Because people from different countries have such different experiences. Of course the race, religious and immigrant composition of their populations and of their faculties are very different. So the questions about which groups are being excluded are quite different, and the strategies for being more inclusive are quite different, and even what they know is quite different. You know, in some European countries people know exactly what the racial composition of the faculty is. And in some, like the Netherlands, those data aren’t collected, so people don’t really even know whether their countries are making progress on, for example, racial diversity of the professoriate. So it’s very stimulating to see the kinds of challenges that people have in other countries, people who are trying to diversify the faculties. And since most of the people are faculty members, somewhere, I feel like I have a very great international sample of people to get ideas from.
Annick van Rinsum
What was the most dissimilar idea or the new stumbled upon?
Frank Dobbin
To me, it continues to be striking that in quite a few countries administrators know how racially and religiously diverse their students are, but they don’t know how diverse their faculty are in terms of race or religion. And in every country I can think of, they know their gender diversity. It poses a very different kind of challenge if you don’t know whether you even have a problem.
Annick van Rinsum
Do you know why this data is not being kept or monitored?
Frank Dobbin
I think it’s a little bit different from country to country. France doesn’t collect racial data in employment records, including in university employment. For historical reasons: it’s one Republic, so race shouldn’t matter. Of course, race does matter. I still don’t quite have my head around why the Netherlands doesn’t collect data on race or immigrant background, or at least doesn’t publish data on the race immigrant, religious backgrounds of their faculty. In the US we don’t publish data on the religious backgrounds because that is protected. So I I get that, you know, different countries have different things that are hot button items that they don’t feel comfortable collecting or reporting.
Annick van Rinsum
Yeah, because you also use the word protected. So in a way, there’s a sense of protection by not collecting those data.
Frank Dobbin
Right, there’s a kind of tension, at least in the US context, that racial categories are protected by law. And we collect data on the racial composition of the workforce, including the academic workforce. So religious minorities are also protected by the same civil rights laws that protect racial groups from discrimination, but we don’t collect religious affiliation or even religious heritage. I think we didn’t do that in part because of what happened during World War 2, where in places where there was a list of where Jews lived in countries that that the Germans invaded, it was much more easier to round people up and send them to concentration camps. But I don’t really know exactly why we’re not collecting religious background at the workplace where we do collect racial, race and ethnic background. So it’s interesting that the countries are different on this because some countries do collect religion.
Annick van Rinsum
I would imagine that you would be happy if countries kept those data points.
Frank Dobbin
Yeah. Well, it’s very useful for social scientists to be able to look at the data. And I think it’s useful for administrators to be able to know where they stand in terms of racially, religious or immigrant diversity. If their faculties are to figure out whether they should be doing something about that, or how serious the problem is.
Annick van Rinsum
So what do we know about the diversity of academic staff in the US?
Frank Dobbin
I would say the main thing we know is about the academic staff, faculty. Their diversity, particularly the representation of black men and women and Hispanic men and women on the faculty, has grown very slowly. Much more slowly than the representation of those groups in the undergraduate body and the Masters level programmes and doctoral programmes. So in particular, the tenured faculty are have been very slow to diversify, and that’s one of the problems we’re trying to address and better understand in our research. What we thought 20 years ago was that there was a pipeline problem. So there just weren’t, for example, black women going in, going to college and going on to get PHD’s. We kind of solved that problem. So now there are black women who go through all of that and get pH D’s and start academic careers. But they are more likely than white men to leave those academic careers. So part of the question we’re asking now is “what’s going on?” After they get academic jobs what causes them to leave in higher numbers than, say, white men.
Annick van Rinsum
Yeah. And what are some of the findings?
Frank Dobbin
One of the things we found concerns people of colour, especially black and Hispanic men and women, but certainly also Asian American men and women. They get PHD’s and start as either postdocs or junior faculty members (in our country those are assistant professors). They tend to lack mentoring. They tend not to get voluntary mentoring from senior faculty in most fields. Most of the senior faculty are white men. And when mentoring is left just to volunteering, often people volunteer to mentor people like themselves. And people will even say it’s better if you mentor somebody like yourself. They think you might understand their problems better. So often white men won’t step up and volunteer to mentor people from other groups. But if almost all of the tenured faculty members are white men in a field like chemistry or physics, it means that the growing number of undergraduates who were black or Hispanic, women or men, are less likely to have mentors. Mentoring is usually effective.
Work-life programmes are very effective in particularly for women of colour. So for black, Hispanic and Asian American women work-life programmes, do a lot. Even when they don’t costs very much money. Because they signal that the university would like to keep you around. One of the things that’s very effective is an elder care programme that just helps you find a place for your ageing, infirm parent to live or to go to daycare. There’s there’s a growing trend in adult daycare in the US. Meant for older people who are not able to stay on their own all day, but are happy to go off and be with other people during the day. So that’s one of the things that’s surprisingly effective. But work-life programmes are very effective whether it’s childcare or you need to shift your teaching around so you can pick up your child, whether it’s parental leaves or extending the tenure clock after you have a child, or two children, or three children in the course of being an assistant professor. One of the issues in the United States is that after you get your PhD, you’re kind of on probation for six or eight years (if you go immediately to an assistant professorship) or longer if you go to a postdoc. Because it takes that long for you to come up for tenure in most universities, it’s usually either six or eight years after the postdoc. So you’re on probation for a long time. Anything the universities do to signal, we want you to have a family life and we’ll help you with that. Come to us if you have a problem helps. So, the more programmes universities set up to address work-life issues, the better women of colour do. White women as well, but women of colour are particularly helped by those programmes, so those are some of the things that are effective.
Annick van Rinsum
So to take account of the actual lives people are leading and specifically certain groups who are previously overlooked, that actually helps diversify. But there’s also lots of initiatives to increase diversity that actually do not work at all, or even have negative effects. But some are still quite persistent and popular. What are the kind of measures, ideas or programmes that are actually unhelpful?
Frank Dobbin
Things I’ve been talking about that actually help people in their careers are getting a mentor: somebody who can help you to figure out how you would get tenure sometime or at some point in your life. Someone to help think about what you might need to do next in your career, or who would just be there for you when you have a bad experience with the chair of your department and you need somebody to talk to. And somebody who could say, oh, it doesn’t matter, your chair is not going to determine whether you get tenure or not. The whole department will. And really, the people outside your university will make that decision because they write letters, people in your specialty. So. And then work-life programmes are also providing some sort of resource to help you stay on the faculty. Both of those things are really oriented to keeping people from leaving academic jobs before they come up for tenure. And so those are effective. Unfortunately, those are pretty rare compared to diversity training or sexual harassment training. Also diversity and harassment grievance procedures are common. Grievance procedures are the most common diversity programmes. They are the procedures you can start if you have a claim of racial discrimination, say against your department, because you feel like you were denied a promotion in a timely manner.
You can bring that claim to the university and a panel will hear your claim. Usually the person or the group, whether it’s, say, your department chair (the person) or your entire senior faculty, will have somebody to represent them in the proceeding. And you’ll present your case, and they’ll present theirs. I think the problem with both diversity training and harassment grievance procedures is that they tend to backfire. They do not lead to more women and people of colour staying on the faculty but to more people leaving the faculty.
We see that in our own statistical analysis where we look at 600 universities over about 25 years. When we look at all of the programmes they’ve adopted, that might affect diversity and all of the policy changes they have in terms of how they hire people or promote people. And with a lot of data in a single model, we can figure out what the effects of different practises are on each of eight different groups we look at: white, black, Hispanic and Asian American men and also women.
So we have pretty good sense of what exactly is helping to promote diversity and what isn’t. So the problem with diversity training and these grievance processes is they’re oriented to blaming somebody for slow progress. Usually somebody who’s in some kind of a position of power, because usually the people who face complaints in these processes are department chairs or senior faculty or sometimes Deans. And the idea is that those people hold the power to make career decisions, but they might discriminate. In the case of civil rights grievance procedures, those are usually used for complaints of racial or sex discrimination in the hiring and promotion or salary setting processes. So when universities put them in place we see no increases for any of the eight groups. They’re supposed to protect those groups but we see some deep, significant decreases. The most striking effect is for black men. It seems that putting in that process can actually decrease the number of black men. I think that’s because, if you bring a complaint, usually you have to leave your job. Because mostly the people who face complaints, like the chair or the Dean, are unhappy that a complaint has been brought against them and they can make it difficult for the person who’s brought the complaint to stay on.
And that’s true not only in academia, but it’s true in the corporate world too. So civil rights grievance procedures and sexual harassment grievance procedures, about 95% of universities have these in place for faculty to use for complaints. But they’re not showing a pattern of positive effects, they’re showing a pattern of some negative effects or no effects. For some groups they’re just not fixing the problem. But universities believe that they will fix the problem so they’re kind of committed to them.
And then about 95% of universities have sexual harassment training. That mostly doesn’t work either. It has negative effects on some groups of women in in the faculty. It’s better if it’s voluntary. It has some positive effects if it’s voluntary. So if you make sexual harassment training voluntary, you send the message Please come to our training if you would like to help. That’s a very different message than making it mandatory, where you’re kind of saying we think you might be engaging in harassment. So you we want you to come learn what’s against our rules and learn what the punishment is if you break our rules.
Why Diversity Training Doesn’t Work? Diversity training often doesn’t work. It’s even given in universities, where there are plenty of social scientists who could advise the president and deans on how to best design and implement it.Most diversity trainings are kind of accusatory. They usually begin by trying to convince faculty that they are biased. There’s often an implicit bias component at the beginning, which is meant to explain what bias is and to convince people that they themselves are biased.Then, it usually moves on to detail the punishments for acting on those biases—punishments that are typically handed out through grievance procedures.This combination—of trying to prove people are biased and then warning them of the consequences—is where a lot of universities have put most of their attention, and spent most of their time and money structuring, at least in the U.S.But it doesn’t show positive effects. Mostly, it shows a pattern of negative effects—like actual decreases in the diversity of tenured faculty.
Now, there are better and worse ways to do these trainings and complaint processes. But I think the solution to the problems we have isn’t in trying to fix individual bias in people’s minds, or in focusing on individual propensities to mistreat others through harassment.
Annick van Rinsum
Maybe not in this way—because, as you’ve said before, this approach really stems from ideas in psychology: the belief that if you become aware of your biases or your problems, that awareness can lead to a solution. But in this case, when you present it that way, people often become very defensive, or even angry, because they feel like they’re being blamed. They feel like it’s their fault—and that just doesn’t work here.
It’s not that people aren’t biased; they are. But these explicit ways of telling them that they’re biased don’t actually help change anything. That’s part of why the problem is so persistent.
You also pointed out that the training doesn’t really protect the groups it’s meant to protect. Instead, it tends to protect the university or the corporation—legally. Maybe you could say a bit more about how that works.
Frank Dobbin
Well, in the United States, both corporations and universities tend to take the same approach: they focus heavily on trying to stop individual-level bias or harassment. They do this through training sessions and through grievance processes that are meant to punish the so-called “bad actor.”
And you’re absolutely right—these kinds of trainings often make people angry. They usually include a section about the grievance process, laying out the punishments you might face if you say the wrong thing or do something that could be construed as discriminatory.
Now, it’s not that individual-level discrimination doesn’t exist—there’s plenty of it. The issue is that these interventions don’t actually stop it. Instead, they tend to provoke anger and resistance. I think we’re seeing that play out at the macro level now, in the backlash against diversity, equity, and inclusion (DEI) efforts in the U.S.
For example, the Supreme Court struck down the use of race in college admissions, and that decision helped trigger a torrent of anti-DEI activity. A lot of it is happening on social media, mostly driven by right-wing pundits, who argue that we’ve gone too far in favoring certain groups—usually they mean Black and Hispanic Americans—and that this needs to stop.
As a result, some states—Florida and Texas in particular—have ended diversity programming in public universities and public agencies. They’ve even fired staff who were working on DEI initiatives. So we’re seeing a really significant political backlash.
During his first term, President Trump used executive power to forbid any company with a federal contract—which includes pretty much every large company in the U.S.—from teaching critical race theory. Now, in the social sciences, “critical race theory” has a very specific meaning, but he was using the term more broadly, essentially to refer to teaching anything about implicit bias or racial inequality.
He issued that order late in his presidency, and Biden overturned it before it could have much of an effect. But Trump is expected to reinstate that order as soon as he gets back into the White House. That could really reshape what a lot of companies are doing.
That said, it shouldn’t affect the kinds of programs that actually work—like mentoring or work-life balance initiatives. These aren’t even technically DEI programs. In fact, it shouldn’t affect many of the things that have been effective for university faculties either.
Still, there is a real movement against these DEI programs—which is ironic, because they haven’t even been very effective at increasing diversity in the first place.
Annick van Rinsum
So they’re saying we shouldn’t have policies that favor certain groups—but the data actually shows that these groups aren’t even being favored by these measures. And yet, this idea is so persistent that there’s still a lot of pushback against them.
Frank Dobbin
Yeah, exactly. I think the irony of this general backlash against diversity programmes is that they’re pretty ineffective. In most companies and universities, if you got rid of them, it probably wouldn’t change much of anything.
But I think the backlash itself is quite toxic. When big companies like Walmart, Lowe’s, Molson Coors or Budweiser roll back their diversity commitments and say they’re going to stop doing some of the things they were doing, psychologists have produced quite a bit of evidence showing that if a company signals it has changed its mind or removes diversity programming, people of colour will feel less welcome.
So I worry those companies might lose some employees—even if the programmes they’re cancelling didn’t really work to promote diversity—because it just looks bad.
There’s also some evidence that companies that keep sending signals that they want to support diverse workers—even if their programmes aren’t very effective—may still be effective at retaining Black and Hispanic workers in particular, as well as Asian American workers and immigrants, by keeping them from leaving for other jobs and encouraging them to apply.
So, I feel like we’re in a situation where universities have these fairly expensive training and grievance systems that don’t work well. And some big states—Florida and Texas—have disavowed those programmes.
I know quite a few young faculty members who want to apply for jobs in those states now. We have a huge national labour market for academics, and both Texas and Florida have large, very good university systems. If those states have trouble recruiting people of colour—and they’re themselves very diverse—that will be very unfortunate for their students.
Annick van Rinsum
These unhelpful policies and programmes are so persistent because of the legal frameworks within which those universities—especially in the United States—operate. If you have a grievance procedure, you essentially have proof that you are trying to solve the problems. Is that right?
Frank Dobbin
That’s exactly right. Yeah. And your legal counsel—the university’s legal counsel or a corporation’s legal counsel—will say that what we need is a procedure that looks like what happens in court if we were sued for discrimination. That way, we can prove that we investigated ourselves and either there wasn’t a problem or there was a problem and we fixed it.
So that’s one reason I think universities, just like corporations, are kind of stuck on this legalistic approach. The training is part of that legalistic approach, partly because in the late 1990s, in two separate sexual harassment cases brought before the Supreme Court, the court said you might be able to avoid liability for a sexual harassment claim—if you’re a town, a corporation, or any employer really—if you have a clear grievance process. That means there’s someone to go to, there’s a commitment that you will hear their complaint and do something about it, and that everyone is informed about the grievance process, often through training.
So the courts kind of encouraged employers to think that these things would protect them if they were sued. They didn’t explicitly state this as protection, but they encouraged employers to believe it would help.
After that, a lot of universities basically copied the playbook of corporations by putting in these procedures and training programmes.
Annick van Rinsum
And it seems they tend to protect themselves more than the groups these policies are intended to protect.
Frank Dobbin
That’s what they have in mind. In fact, if you get sued, it’s very unlikely that the person who sues you will succeed. Very few race discrimination or sexual harassment lawsuits actually provide any benefit to the person who filed the suit.
But the courts don’t really consider the fact that you’ve had diversity training or that you have a grievance procedure as evidence that you’ve genuinely tried to fix the problem. So, if you do get successfully sued, that’s not really a strong defence.
However, companies think it is, and there’s some logic there: if we investigate ourselves, then we can make sure there isn’t any bad behaviour.
The problem is, people know that if you bring a complaint against your boss for race discrimination, you probably won’t have a job in two weeks. In fact, the most likely thing to happen if you bring a complaint against your boss is that you’ll have to leave your job, while your boss gets to stay.
Annick van Rinsum
Also in the Netherlands, we don’t tend to go to court as often to resolve these kinds of problems, but I see very similar procedures in place—like diversity trainings and, I think, also grievance-like procedures. So I think it’s maybe not just about the law, but also a kind of moralistic idea of: this is how we should be.
Frank Dobbin
I mean, if someone is mistreated, we should protect them—that makes sense. But this whole set of solutions also suggests that everything else is fine. That there’s nothing wrong with, say, the academic career system. Nothing wrong with the fact that most people have to work 60 hours a week between the ages of 28 and 35 in order to have a successful academic career. That this doesn’t conflict with any gender norms in society, or that it shouldn’t lead to any inequality.
The suggestion seems to be that the only issue is a few bad actors—and that if we could either change them through training or punish them so they stop discriminating, we could solve the whole problem. And I just don’t think—as a sociologist—I don’t know why anyone would believe that. I don’t know what kind of theory of the world you’d have to think that’s how we could actually improve the lives of people trying to succeed in academia.
Annick van Rinsum
So on the one hand, there’s the “bad apples” narrative, but there’s also the systemic, “everyone has biases” kind of narrative. How do these two fit into the way our diversity programmes are structured?
Frank Dobbin
I think the idea is that everyone has biases—that’s essentially what implicit bias training is based on. And, just as you were saying earlier, if you simply make people aware that they’re biased, then they’ll be able to correct themselves, because they didn’t mean to be biased in the first place.
But people tend to react quite negatively to being told they’re biased—because they don’t believe they are. And it’s very hard to convince people that they are biased. The people who are most explicitly biased will just say things like, “Well, people from this group over here aren’t as smart, and that’s why they’re not as successful in academia.” Or they’ll say—as many people have said to me, as a white man—things like, “Men are better at maths, so of course men do better in academia.”
And I’ll say, “Well, actually, the research doesn’t show that men are better at maths.” And then they’ll point to certain individuals who’ve won Nobel Prizes. And then I’ll point out that only men received Nobel Prizes—apart from two exceptions—until about 20 years ago. So Nobel Prizes don’t actually tell you very much.
I do think people tend to interpret what they see in the world as though the world is meritocratic and fair, rather than asking whether there are ways in which it isn’t. And instead of thinking about how we might change those structural issues, we focus on trying to change the minds of people who don’t think they’re biased—though they might be.
But it turns out that it’s incredibly hard to change bias. There have been over a thousand experiments—mostly by psychologists—trying to reduce bias, and on average, they have no effect, or only very short-term effects on implicit or unconscious bias. It’s extremely difficult to change through any kind of training or intervention.
Annick van Rinsum
Do you think the policies we see in place actually support the idea that the world isn’t meritocratic and fair?
Frank Dobbin
Well, let me give you one example of a policy I’ve mentioned, and one I haven’t. We carry out a lot of interviews in addition to analysing years of quantitative data. One thing we’ve consistently seen in interviews is that when we speak to people involved in mentoring programmes—specifically those assigned as mentors—we uncover something quite telling.
The most effective mentoring programmes in academia typically operate at the departmental level. Each department is responsible for setting up and designing its own scheme, and the main requirement is that everyone in the six to eight years leading up to tenure has a mentor. Everyone should have one.
What we’ve observed, as a consequence of these programmes—and we see similar effects in corporate environments—is that white men who become mentors often say something like: “One of the things I learned through mentoring is that our system isn’t entirely fair. I always thought it was, because I always saw people like me succeeding through merit. I succeeded because I’m good at what I do—at least, that’s what I believed.”
But then they’re assigned a protégé—perhaps a Hispanic man, or an Asian-American woman—and that person comes to them with challenges they’ve never encountered before. Things like: no one listens to her ideas during curriculum development meetings, or colleagues repeatedly telling her she’ll never win a grant and shouldn’t even bother applying. Or even being told she has to choose between having children and staying in academia. And the mentor is shocked, because no one has ever said anything like that to him.
And often, the protégé turns out to be brilliant—frequently the strongest person in the group. Mentors tend to claim the success of their protégés, even when they were randomly assigned to them. There’s a little cognitive dissonance there—they like to believe they personally chose that protégé.
Then they say something like, “Actually, this system isn’t fair for everyone. It might be meritocratic—for some—but not for all.” Which is striking, because if a system is truly meritocratic, it should be fair for everyone. And so they become champions of mentoring schemes, particularly those that support individuals from groups who historically haven’t had access to mentorship—essentially, anyone who isn’t a white man.
Sometimes they’ll even say, “I didn’t think this programme was a good idea initially, but now I realise that she might not have had a mentor at all if she hadn’t been randomly assigned to me.” That kind of experience genuinely opens people’s eyes to structural discrimination.
I’ve been talking about individual-level discrimination—where women, for instance, might be discouraged from pursuing success—but there are also structural barriers. Take parental leave, for example. Most universities now offer a term or a year without teaching for new parents. But some still say, “No, you can’t have leave in your child’s first year—we didn’t plan for that in the curriculum. You can take the second year instead.” But for a new mother who’s breastfeeding, that’s not very helpful.
So even though the rule says you’re entitled to a term or a year without teaching, it doesn’t always allow for flexibility. Ideally, the parent should be able to choose the term or year that suits them. Maybe the mother wants the second term because the father is taking the first one, or because the other parent is a lawyer whose workload is lighter at a certain time of year. In any case, structural barriers like that still exist—and mentors often come to recognise them only once they’re exposed to these situations.
Another example: suppose you run a large lab, and you need to take a year off because you’ve had twins or a child with a medical condition. Most major funders don’t have any way for you to put your grant on hold for that year. These are the sorts of structural problems mentors become aware of through the experience.
Now, a second example—another practice through which people, often white men, come to understand the challenges faced by others—is participation on diversity task forces. These are among the most effective tools universities have to promote faculty diversity. The same applies in the corporate sector, based on our broader research.
When university presidents or deans form such task forces, they often ask department heads to either join the group or nominate someone from their department. If the task force itself lacks diversity, they usually add a few members to ensure broader representation.
Given that universities are full of scientists, these committees tend to examine data—hiring figures, promotion trends, leave requests, and staff departures—typically twice a year. They try to pinpoint where problems lie.
One of the first patterns they tend to spot is that women are more likely to leave academia entirely—especially when compared to white men. So they begin asking why.
Many people tell me that, prior to serving on one of these committees, they thought: “We don’t need this. Our department doesn’t even have many women or people of colour, so we don’t need to be involved in a diversity task force.” Which is an interesting way to look at it.
But then they attend the meetings, examine the data, and as scientists, start to notice patterns. For example, women were leaving lab-based sciences after their second child—sometimes even after their first—whereas men were not. Women were often leaving for industry instead. So the question became: why is this happening?
They tried to solve the issue—initially through work-life balance schemes or counselling. But those didn’t work. After a couple of years, though, they found approaches that did help—and managed to halve the number of women leaving.
These same people then say, “I didn’t want to be on this committee, but we actually did some good. We solved a problem I didn’t even know existed. And now I’m staying on voluntarily, because I can see the value in it.”
So, whether through the personal lens of mentoring or the institutional focus of a task force, people come to see that their institutions aren’t entirely meritocratic—or that they present significant barriers for women and people of colour. They also begin to see themselves as part of the solution.
Rather than feeling blamed in an implicit bias training session, or threatened by grievance procedures for harassment or discrimination, they feel like they can actually make a difference. Of course, not everyone’s mind is changed—but it’s surprising how many people we spoke to in interviews told us those experiences shifted their perspective, convinced them there really is a problem, and made them want to help fix it.
Annick van Rinsum
So in a way, making people aware does help — but it has to be done in a way where they feel like they’ve discovered it themselves, and that they’re not being positioned as the perpetrator of the injustice that’s happening.
Frank Dobbin
I think it would be quite similar to bystander intervention training for sexual harassment, which we’ve seen on university campuses — when undergraduates are trained — and also in the US military, when new recruits are trained. What we’ve seen is that, in bystander intervention training, people are taught how to deal with situations where sexual harassment might be beginning to occur.
That kind of training automatically places men in the role of helping to solve the problem, rather than being the problem themselves. And people tend to prefer being in that role. Even if, from a feminist theoretical perspective, putting men in that role might be seen as problematic, it does appear to be more effective in making progress than casting them as the culprit — even if, in some ways, they might be culpable.
Annick van Rinsum
Yeah. So in a way, it is unfair because the emotions and feelings of people in positions of power have to be taken into account quite a lot in order to actually bring about change — in order to help the groups that are structurally discriminated against.
Frank Dobbin
Yeah, I mean, some people object to some of our findings because they’ll say, well, why are you treating those people so gently?
Annick van Rinsum
And the response is: this is just evidence-based — it’s simply what works better if you actually want to bring about change.
Frank Dobbin
it’s quite simple psychology — if you give people a positive message and offer them a positive role to play, they’re more likely to join your cause than if you send the signal that the problem with the institution is that people like them discriminate or harass others.
Annick van Rinsum
And it’s a good thing that, in the majority of cases, this works well. But I’m also thinking, for example, if you’re mentoring someone who isn’t from your own group, and you’re, say, a white man, and you hear your mentee talk about how she isn’t being heard, or how people talk over her, and your reaction isn’t, “Oh, there’s a problem in the system,” but maybe, “She isn’t that smart,” or “Maybe she’s just like that” — if that doesn’t change the mentor’s mind, then it’s still very problematic. So, yeah — do you have any thoughts on how we can deal with that?
Frank Dobbin
Well, I think it’s important to keep in mind that, at the macro level in our datasets, both for corporate managers and university faculties, mentoring programmes are among the most effective tools for promoting diversity among managers in corporations and among tenured faculty. So, on average, they work. That doesn’t mean there aren’t many cases where a woman is assigned a mentor who is a habitual harasser and who mistreats or abuses her. And it doesn’t mean that beneath the overall positive statistical effects there aren’t a number of failed mentoring relationships.
One strategy that I think has been effective in many universities is to develop a mentoring team for each junior faculty member, so they don’t just have the one person they’re assigned to. They might have someone in their department, perhaps in their field, someone outside the department, and perhaps a peer mentor as well.
Another approach that both firms and universities use is to assign someone new every year—not as a replacement, but as an addition. So if the first mentor you’re assigned proves problematic, you’ll have someone else to turn to.
One thing universities have picked up from corporations is the idea of employee resource groups. These are groups of people, usually from an identity group—such as Black employees or women. Often people prefer to have someone from their own group to talk to about certain issues, but also perhaps someone from the dominant group—usually white men in most university departments—who could act as a sponsor or help connect them with Deans or others who might support their careers.
So, I do think it’s important that people who are assigned a mentor who isn’t working for them have a way out. And if you start by talking about mentoring networks, it becomes easier for the person with a problematic mentor to approach whoever runs the mentoring programme and simply ask for someone else straight away.
But yes, you’ve touched on a very important problem.
Annick van Rinsum
I wanted to ask you as well, because some of the things that work well aren’t officially labelled as diversity programmes, which is kind of a good thing—especially if those programmes are being banned or restricted again. If they fall under work-life balance categories, there’s a much better chance they can still continue.
But, for example, another thing that works well is diversity task forces. So, what would your recommendations be for deciding whether measures should explicitly fall under diversity programmes or not, especially in the context of Trump’s policies?
Frank Dobbin
Well, most of the backlash has been against diversity training in universities and diversity positions in universities. So, I think it’s a bit difficult to know what this will look like under a Trump administration. In his first term, he didn’t always keep the promises—or threats, if you like—that he made during the campaign. Some things he did, some things he never got around to. He didn’t do much in higher education, but he might do more this time, so it’s a bit hard to know what to worry about.
What I’m already seeing in corporations—because quite a few have rolled back some of their diversity language and commitments—is that sometimes they continue similar programmes, but just call them something else. So, for example, a diversity task force might simply become a task force on retention. Retaining people in the academic workforce. And it would be undertaking similar programmes, like mentoring, or encouraging work-life balance initiatives.
Annick van Rinsum
It’s also quite obvious, isn’t it? If you call them retention programmes, people will assume that you are specifically focusing on certain groups who may be underrepresented.
Frank Dobbin
Well, I mean, there’s something of a brain drain from academic science to industrial science. So I think it’s quite easy to say we need to retain more scientists in academia. And ironically, I think most—even most university administrators—are unaware that women are much more likely to quit their academic jobs than men. Unless they’ve been on the task force, they simply don’t know that. So it seems unlikely to me that right-wing pundits will immediately recognise that a retention task force or, say, a faculty excellence task force, is addressing these issues.
One of its main goals ought to be to retain faculty—to stop them leaving for other jobs. So I think something like that could be rebranded quite easily, and even the practices might not change very much. Certainly, mentoring is already not labelled as a diversity programme in most universities. It works very well just to have departments run their own mentoring schemes, because then Black, Hispanic, Asian American faculty and women all get mentors—and that’s what helps. Work-life programmes generally aren’t seen as diversity programmes now; they’ve become both a Republican and a Democratic issue. Both sides want paid parental leave, for example, which very few states currently provide. Both sides want more childcare. So I think that would be fairly easy to sustain. A lot of the programmes that work could survive.
I’m not trying to paint an overly rosy picture though, because, again, if states ban any diversity programmes—as Florida and Texas have done—I think we’ll start to see some effects. People of colour may not apply for jobs there, and as a consequence, they may leave academia, since those two states are very large employers of faculty. More states may follow suit. So I don’t mean to be Pollyannaish; it’s not a good time for universities when questions about diversity arise. On the other hand, there are still things universities can do, I think, over the next few years.
Annick van Rinsum
Entering programmes retention task force, what else?
Frank Dobbin
Work-life programmes of all sorts—so parental leave, paid parental leave or paid family leave—universities can usually afford those quite easily. Childcare, whether it’s subsidised childcare or services that help you find childcare in the community. As I mentioned before, elder care, where the university doesn’t pay for it but helps you find a place to care for your elderly parent. So all kinds of work-life programmes are effective.
It’s also quite effective for universities to have a diversity officer. Some have got rid of those posts, like in Texas and Florida. We’ll just have to see in the future whether more states try to cut those positions.
One thing we’ve noticed is that when Deans collect data about different stages in the career process, that tends to have a positive impact on faculty diversity. So when Deans, for example, gather data on who applied for junior positions in the department and who got interviewed, they often look at who applied and review the ranked candidates for interviews. They might say, “I think you should interview candidates 1, 2, 3 and 5 instead of just 1, 2 and 3,” because 5 is a person from an underrepresented group who was almost in your top three anyway. That’s one way they can use data.
We’ll have to see if they can find a way to keep doing that under the threat of state legislatures trying to end diversity and inclusion programmes.
Annick van Rinsum
So one thing we haven’t mentioned yet is the Ombudsman’s office. If things do go wrong, that can also be a helpful way to address these kinds of problems. Maybe we could briefly touch on that as well?
Frank Dobbin
One of the problems universities and corporations in the United States have faced is that whenever there’s a racial discrimination or sexual harassment complaint, it often goes through formal grievance procedures. As I mentioned before, usually if you bring a complaint, you end up having to leave your job — either you get fired, or you face further harassment or discrimination. Usually, the whole process isn’t confidential, so the person you complain about knows, and others find out, including their friends.
What works much, much better, and there’s quite a bit of research on this, is using an Ombudsman’s office. That’s an official whose role is to provide advice to any employee who’s having a problem with management or colleagues. They don’t intervene directly and don’t usually speak to the person accused of racism or harassment, but they help the complainant figure out solutions. For example, if you’re a biochemist, maybe you can switch your main appointment from biology to chemistry, where the problem person isn’t. Or maybe the Ombudsman can find out if the person has caused problems for others and work out ways to stop it. People who use the Ombudsman’s office tend to be satisfied because they get advice on how to handle the situation so it doesn’t continue to affect them.
Another option many universities and corporations have is a dispute resolution process, where you can raise a complaint with a mediator who tries to resolve the issue between the complainant and the accused. Studies show that about 90% of people in large organisations who go through this mediation process are happy with the outcome, and very few leave their jobs. So they avoid the backlash that often happens with formal grievance procedures.
In short, there are better ways to handle complaints than the traditional grievance route.
Annick van Rinsum
Also, I still think that having to switch from biology to chemistry because of a problematic person seems really harsh. But, unfortunately, in some cases, given the current system, that might be the best available solution for that person.
Frank Dobbin
Absolutely not. That’s obviously a suboptimal solution. I mean, the person doesn’t have to accept a solution they’re unhappy with. The good thing about both systems—the ombuds office or the dispute resolution process—is that if you don’t like the outcome, you can always escalate to the formal grievance procedure, since all universities have one. But at least you’ve explored other possibilities first. Most people are satisfied with the solutions reached at those earlier stages. That said, I agree that switching from biology to chemistry, or vice versa, is usually not the outcome anyone wants. Partly because it ends up punishing the person who faced the problem, rather than the person who caused it.
Annick van Rinsum
Yeah, but in reality, if you go through a grievance procedure, you’ll probably end up worse off because it’s unlikely to help you.
Frank Dobbin
Well, the good thing about starting with these other processes is that you can at least take some time to think through your options, and the people you’re consulting with aren’t trying to push you into a grievance procedure—they’re trying to help you find the best possible outcome for you.
Annick van Rinsum
What’s the key message you want people to take away if they want to make the world a bit more meritocratic or inclusive?
Frank Dobbin
Generally, the measures designed to tackle individual-level bias through training and punishment simply haven’t worked. They’ve often been counterproductive. Yet some employers, universities, and corporations are reluctant to let these go. And when a university vice-chancellor says, “We can’t cut the diversity training programme,” I usually respond by saying, just don’t put any more time or energy into it, and instead focus on things that might actually work. Those things involve changing people’s career experiences by ensuring new recruits get hired and receive mentoring, and by providing robust work-life support. It’s during those first six to eight years before securing tenure that people are most likely to leave academia and not return, which is why this is so important.
Another effective strategy in US universities is what’s called “target of opportunity” programmes, where the Deans might say: physics, chemistry, biology — you don’t seem to be hiring enough women, so why not look harder? If you find some women you really want, we won’t make you wait until there’s a vacancy in the very narrow field you’re recruiting for, say organic chemistry — you can hire them now. The same approach applies to Black professors in some fields. This kind of programme is very effective, so it changes the hiring system.
I’ve also spoken about changing the mentoring system and improving work-life provisions. Those things genuinely work well. And, yes, having an excellence task force also helps.
Annick van Rinsum
Thank you very much for sharing all this knowledge today.
Frank Dobbin
My pleasure.
