On the first day of September, I welcomed a new member of my team in the dean’s office via Zoom. I was on the campus, but our new hire — let’s call her C., so as not to put her on the spot — was still more than 1,000 miles away.
C. had asked to start work remotely for three weeks so she could have time to pack up, find a place to live, and move, all of which would take more time than usual given the restrictions of pandemic life. Just that morning, she had closed on a new house. Had she come to New Orleans to look at our housing market? No, she replied, a realtor had showed her houses via FaceTime. “I’m committed!” she said.
The entire job search had been done during the pandemic. We had screened applicants, gone through three rounds of interviews, and recruited our candidate of choice — all without shaking a hand, being in the same room, or hosting finalists. Even local contenders for the job never crossed beyond the digital threshold.
Speaking with her that first day on the job, it struck me how unusual all of this would have seemed only six or eight months ago.
Her success remains to be tested, of course. Still, I am confident that our hiring process not only led to the best choice, but that it screened out aspects of academic hiring that have long bothered me for how they replicate forms of privilege and implicit bias.
As a dean and professor, I’ve done my share of hiring, and chaired or been involved with dozens of faculty, staff, and administrative searches. In this case, C. holds a senior administrative position. But what we learned from this virtual search seems relevant to both faculty and administrative hiring going forward.
Driven out of necessity, the way we structured our search process helped us focus more on the talents of the candidates and the likelihood of their success in the job than on the superficial aspects of traditional interviews which tend to leave cultural self-replication unchallenged. If we are serious about transforming academic culture and leaving behind the pernicious category of “fit,” we need to reconsider how we go about hiring.
I wasn’t thinking about behavioral economics when I started this job search, nor the deep tension between intuition and judgment that can plague in-person interviews. In fact, the job description itself had been crafted before the pandemic forced us to take a new approach, and I thought little of the means by which we’d go from approval to hiring.
We were about to post the job ad when Covid-19 hit, which led to a hiring freeze for all but essential personnel. A significant portion of this position is essential to our educational mission, so we got approval to restart. But now we were in pandemic mode and had to improvise.
First, I asked a larger number of colleagues than usual to play a role on the search — not just people in the dean’s office but also faculty and staff members — and divided them into teams of two or three. The first team screened applications based on CVs, focusing on qualifications. Then a second team conducted Round 1 interviews, with questions and interview duration standardized. Yet another team interviewed those who advanced to Round 2. And I recruited another dean to interview the final two candidates in Round 3. My executive assistant coordinated the process and asked everyone to write up their ratings of the candidates at each stage. I asked my executive assistant not to tell me what the teams were reporting until the end of each round.
I also interviewed the candidates at each stage — briefly for Round 1, and more extensively as we went forward. I compared my notes against those of that round’s interview team, before deciding which candidates would move into the next round. This allowed for dissent throughout the process, and heightened the likelihood that judgment, rather than intuition, would take precedence in our decision. As it turned out, there was agreement at each stage. The final decision was mine, but in making it, I had maximized the amount of input I had to draw on and purposely slowed down the process.
Why so much structure?
Job interviews are one of those tasks in which the two competing aspects of decision making — intuition and judgment — are particularly likely to be thrown out of balance. In his 2011 book, Thinking, Fast and Slow, Daniel Kahneman, an emeritus professor of psychology at Princeton University and winner of the Nobel Prize in economics, argues that there are two aspects to how we make decisions. He calls them “System 1” — driven by intuition, speed, and emotions — and “System 2” —more deliberative and slow, associated with judgment.
System 1 is highly valued in our society. We often celebrate leaders for their quick decisions and “expert intuition.” Intuitive decision making frequently carries the day, particularly in job interviews. Even those committed to working against implicit biases or the so-called halo effect (the influence of positive first impressions) have a hard time resisting the impulse to rely on intuition.
Anyone who has served on a faculty search committee, working through dozens or hundreds of applications in order to derive a short list, has witnessed how conference interviews or campus visits can turn the tables. We argue against fetishizing the brilliant job talk but find ourselves unable to resist the candidate who “interviews well.”
In his work on interviews, Kahneman tried to temper the powerful charisma of System 1 thinking by developing System 2 mechanisms. He asked colleagues to decide in advance which aspects of doing a particular job mattered most, and then to use interviews to rate candidates numerically on how well they fulfilled each characteristic. Such simple algorithms, he argues, are better than so-called “expert” judgment. “Because interviewers are overconfident in their intuitions,” he writes, “they will assign too much weight to their personal impressions and too little weight to other sources of information.”
Even before Covid-19, many academic departments were moving away from using annual scholarly conferences to conduct first-round, in-person interviews for a faculty opening. One of the reasons for that shift is budget cuts.
But another is the clear ethical case against conference interviews: the awkwardness of interviews held in the search chair’s hotel room; the high expense for graduate students or underemployed Ph.D.s to travel to conventions, frequently for a single job interview; and the environmental impact of so much travel. Those are all good reasons to start using Zoom interviews — even after the pandemic is over — to assemble the short list of candidates you plan to bring to town for campus interviews.
Now, with Covid-19 continuing to limit travel this year and to force academic conferences to go virtual or be canceled, the question is: What might we gain by moving all stages of the hiring process online? (Some may object that on-campus interviews allow departments to recruit a candidate and show off their campus, colleagues, or city, but this function can be separated out till after a hiring decision is made. Too often, in my experience, the so-called “wining-and-dining” stage can go awry and privilege social niceties over academic and professional credentials.)
As I look back on our experience running this administrative search, virtual interviews democratized our procedures and minimized aspects of hiring built around intuition and discriminatory assumptions about class, gender, disability status, and race — assumptions that I believe fall into the category of System 1 thinking. By focusing attention on the candidate’s talents and experience, virtual interviews can screen out many if not most of what decision makers, despite their best intentions, may be distracted by during initial in-person encounters.
Sure there are multiple social cues and codes that can come across in a Zoom interview and put implicit bias back into play. Room Rater, a popular Twitter account that evaluates (tongue firmly in cheek) Skype and Zoom backdrops, has taught all of us to look at cues in the background of a room. Being able to captivate during a Zoom interview allows some candidates to thrive. Putting your laptop on a stack of books, aiming a warm light source at your face, and taking care with your background are simple hacks for job candidates that can be surprisingly effective. We need to be on the lookout for the candidate who “interviews well” on Zoom, too, despite lacking the background and training we need for the position.
But the tendency to be influenced by a candidate’s charisma can be countered on the hiring side by a rigorous System 2 process that builds algorithms and rates candidates accordingly — like the process that we put together on the fly. For faculty hires, no less than administrative ones, quick judgments are activated by prestigious letterhead or famous reference writers. Even simple algorithms that isolate the characteristics for success as a faculty member at your particular institution — rather than a generalized image of success — forces interviewers to focus on those traits.
After C. had accepted the position, I told our new chief diversity officer my theory about the all-virtual search: “I feel I know all that I need to know. When C. gets to New Orleans, I’ll find out if she’s short or tall.” That was a joke, of course, but intended to make a serious point.
Like any of the candidates, C. might be short or tall, in a wheelchair, or have no legs at all, but none of that would matter for the job we’d hired her for. And while height, weight, the kinds of shoes you wear, or the briefcase you carry are not at all relevant to the job, so many of those cues do affect decision making in person, however implicitly, which is where the devil lies.
Brian T. Edwards is dean of the School of Liberal Arts at Tulane University, where he is also a professor of English. He is on Twitter @briantedwards.