Measuring the Effects of a Diverse Workforce

Q&A with Professor Judith White.

Why did you choose diversity as a Research-to-Practice Seminar topic?

There are no courses taught here specifically on diversity. When we talk about the Research-to-Practice Seminar I teach, we're really talking about a gap between the research on a subject and the practice of that subject. And I really couldn't think of any area where there was as big a gap as diversity. There's an awful lot of practice in diversity, and there is some research, but the two have different languages and ideas and don't really talk to each other. This was an opportunity to create a synergy by putting them together—to discuss how the research is translated into practice and to identify the pitfalls that may come up in that translation.

How did you prepare for this new way of teaching and structure the course?

I planned this as a linear narrative, divided into discrete, sequential discussions—similar to an academic review paper. There was a heavy reading workload—the research papers are 20 to 50 pages long, with very dense statistics and theories—but there were no lectures and no PowerPoint, except in the final presentations. Students took turns leading each week's discussion, and I met with them ahead of time to review the readings. The entire group would then meet to discuss the individual papers and synthesize the research and diversity practices to see whether we could come up with a better way of evaluating those practices. I wrote the discussion summaries, and the students gave final presentations at the end of the seminar, applying the research and discussions to specific companies' diversity-management practices.

Why did you think it was important not to lecture, as you would in a core course or elective?

When I teach my elective, the idea is that I'm imparting what I know to the students. They come in assuming that I know something they don't know. In the seminar, it's a very different feeling. In my mind, I have one problem that I'm trying to solve, which is, What is the research-to-practice gap, and how can that inform my research? The students have a different problem that they're trying to solve. They're asking, What is the research-to-practice gap, and how can that improve my practice? I'm the one who designed the course and structured how it would unfold, but the students were the ones who brought the questions, the experience, and the context to it. Each one of us had a certain set of information that the others didn't have, and we combined all that to create a true learning experience together, which was so exciting for me.

In the seminar, you discussed "evidence-based" management. What does that term mean?

It comes from medicine. It's the idea that practice should be based on evidence, not on past practice or opinions or anecdotal information. So in medicine, when they say something is a best practice—like how long you should wash your hands—they don't mean what the best hospitals do; they mean which practice has been shown to maximally reduce the number of infections. They collect the actual data, review it, and determine from it what the best practices are, and those are what they emulate. There's a movement in management to do the same thing.

Speaking of definitions, how do you define diversity?

We started the seminar by trying to define it from our readings and discussions. For some, it's a political term that means affirmative action; for others, it's a research term that means differences. And companies talk about diversity in different ways. In some places it's race-based; in others, it's based on caste or gender; and in some it's based on beliefs or training. It was clear that there are different definitions, and that's important. That's why the title of the course isn't Diversity Management, but Measuring the Effects of a Diverse Workforce.

So how do you measure those effects?

Early on, we discovered that people actually don't measure the effects. They do a lot of diversity-related things, but they don't actually measure what impact those things have. To do that, you first have to have a theoretical model, and there are two basic models in the literature. The first is a commonality model, based on demographic diversity and inter-group relations, which says that people are attracted to and comfortable with people who are similar to them in some way. But there's a more recent model of diversity that comes out of biology, which suggests that diversity is essential to keeping an organism healthy, resistant to threat, and competitive. So, following the biological model, should we measure things like the costs of employee turnover, productivity, recruiting, and training; or overall firm productivity, return on investment, or some financial metric; or something else, maybe innovation, longevity, or the ability to change? How do we know which things are a causal link? There are so many intervening variables between the composition of a workforce and how well a company performs, and we have to develop better models.

Would you say that these seminars are a groundbreaking way of learning?

I would, at least in the area of management. As far as I know, this is the only course like this that has ever been taught. I canvassed all the diversity classes I could find among MBA programs around the country with colleagues of mine, and all the syllabi were on one side or the other of the research-to-practice divide. Either they were pure Ph.D.-style courses designed to give students research theories and information, or they were MBA-toolkit kinds of courses, designed to give students information about practices via cases or textbooks. But when you teach a course like that, the students never have to ask or answer the questions, Why does this work so well? What are the fundamentals in my field? What are the fundamentals of human behavior that result in this being the right thing to do?

And that's that synergy between the research and practice.

Yes. What works for me in management is to get people accustomed to making decisions based on the evidence, and the evidence is based on objective gathering of data with the understanding that there are some underlying rules: if you collect enough data, you'll get the rules, and if you get the rules, you'll know what to do.