How Networks Affect Searches
The University of Houston has a vast trove of data on faculty searches, a byproduct of an aggressive effort to diversify the university’s professoriate. A team of the university’s researchers mined that data — 13,750 applications for 156 faculty positions over three years — and came to a striking conclusion.
The study showed that search committees chaired by women or underrepresented-minority faculty members led to significantly more diverse applicant pools. Committees led by women can be expected to have 23 percent more female applicants, and those led by underrepresented-minority faculty members can be expected to have 118 percent more applicants from such groups.
What caused the increased diversity in the applicant pool? The researchers point to homophily, a psychological theory that says people prefer others who share a characteristic with them to those who do not. This means the people leading faculty searches tend to tap into their own networks, in which people are more likely to share their same characteristics. This, the researchers say, partly explains why the professoriate continues to be disproportionately white and male, but it also offers a roadmap for change.
The Chronicle spoke with Christiane Spitzmueller, a psychology professor and one of the study’s authors. The following interview has been edited for length and clarity.
What’s the genesis of this study?
Like most universities, we have struggled to create a faculty that mirrors the diversity of our student body. There are a lot of benefits to having a diverse faculty, but the actual how-to, the what works, wasn’t quite as well understood. So we redesigned our faculty-search processes, and we collected data and evidence every step of the way. And one key piece of data was the applicant pool. We decided to take a very close look at these searches with some very rigorous modeling. We have a provost who really encouraged us to look at every single piece of data, even data that nobody has looked at, like the external review letters for promotion and tenure cases. She gave us access to everything.
The Chronicle wrote about the University of Houston’s effort to diversify its faculty last year. It sounds like the campus has been paying attention to faculty diversity for years.
Yes. The recruiting part has been critical. What’s neat about this study is that appointing women and faculty of color as search-committee chairs is not that hard to do. We can talk about how it might increase the service load of people who already have a high service load, but it’s not hard to accomplish if you free up time otherwise.
Some would argue that it is, in fact, hard to do. In some disciplines in the sciences, about 1 or 2 percent of doctorate recipients in any given year are Black, and women are also heavily underrepresented. How much of a challenge would this be?
You’re right when it comes to some disciplines, but there is much more diversity in almost all disciplines than there has been. Even in some subfields of computer science, you have much better diversity and options. For a lot of institutions that have not paid attention to this, it’s very hard starting from zero. Once you’ve built critical mass, and faculty of color recognize that it’s a good place to be and people are retained and promoted and supported and there’s infrastructure, then it becomes much easier. I also believe there’s an opportunity to partner with other related disciplines in appointing people to chair searches or even participate as committee members.
Could the discipline be determining what both the search-committee composition and applicant pool looks like? So, for instance, I would expect it to be more likely that a search for a Black-studies professor would have both a Black committee head and more Black applicants. That doesn’t mean the committee chair caused the diversity in the applicant pool.
That’s something that we absolutely had to deal with here. Because otherwise we’re looking at spurious relationships based on how diverse fields are to begin with. To deal with that we used multi-level modeling, and we controlled for the diversity of the discipline statistically. So that’s basically modeled out of it. So the effects that you see control for the amount of diversity in the discipline.
You recommend integrating homophily-theory tenets in training the faculty to do searches. How can colleges and departments go about doing that?
In my opinion — and this is not substantiated in the paper — there are too many of these search-committee trainings, like anti-bias training and implicit-bias training. Unfortunately, the research evidence tells us that’s not particularly effective. Oftentimes when you tell people about their biases you sort of get backlash, and people feel like they’re being personally attacked. It’s a much more effective strategy to teach people techniques, things they can actually do to make a difference. Teaching people about these principles is not very hard. Having them write down the names of 15 people in their discipline, and then afterward having them write down the demographics of those people, the homophily principle becomes immediately evident. Because you see how we all know people who are like us. I’m an immigrant to America, so I know lots of other immigrants who do research in the same space.
What surprised you about this study?
We had thought we would see a stronger effect for a signaling- theory kind of process. So we had expected that putting a person of color or woman in charge of a search committee sends the message to potential applicants that, as an institution, we are putting people who are traditionally underrepresented in positions of power. That sends a signal that you, yourself, could potentially be promoted, that you could climb up the ranks, and this might be a good place for you to be. We did not find very much support for that signaling-theory effect.
What questions do you want to dig into more?
There are a million questions left. Right now, what I’m most interested in is, we’ve done phenomenally well in terms of recruiting, but for the academy to ultimately become more diverse, we need to make sure faculty of color are promoted, retained, and have the same opportunities as everyone else. We just got a $2 million National Science Foundation grant to examine bias in promotion and tenure processes. Our grant looks at external review letters. A single sentence in an external review letter can tank your career. Unfortunately, there are more of those negative sentences in letters for women and faculty of color. —Vimal Patel