How do you hire for success?

I don’t know.

Actually let me qualify that statement (as you would expect from an academic). Except in one category with a limited data set, I don’t know how to predict with a high probability whether a given candidate will be successful, at least when it comes to hiring new assistant professors at liberal arts colleges that emphasize both teaching and research. In the lingo, these are often referred to as SLACs (Selective Liberal Arts Colleges). When I use the acronym, I will refer to the wider net of SLAC environments and not necessarily small prestigious liberal arts colleges.

Okay, okay. I thought I knew once, but now I’m not so sure. I naively thought I would figure out the secret tried-and-true formula after being involved in hiring many new faculty members over the years. Many years and hires later, I haven't figured it out. However I’ve been thinking about this issue recently after reading several essays from Malcolm Gladwell’s What the Dog Saw. The book is a collection of Gladwell’s articles from The New Yorker magazine. His pieces are well written, engaging, and they make you pause and think.

The first essay, Most Likely to Succeed, discusses the vagaries of picking quarterbacks from college-level American football who could be successful in the professional league (the NFL). This turns out to be a tricky proposition because the pace and style of the game in the professional level can be quite different from the college level, and therefore a successful college quarterback could do miserably in the NFL (and there are many such examples). In his essay, Gladwell links this issue to finding good teachers in schools. His argument: Potentially the best way to find the best people is to put them in the actual job and see how they perform. This requires having a much larger pool to start, keep the good ones, and let go the weak ones.

The second essay, The Talent Myth, discusses the dangers of the McKinsey and Enron philosophy of “hiring top talent at all levels”. It’s not clear what counts as top-talent, and often the prestige of the institution and degree program attended by the candidate is used as a proxy. Gladwell describes this as the War for Talent and that the strategy used is “differentiation and affirmation”. Essentially, pay the top performers exorbitantly, and push out the lower performers. HOW one decides how to differentiate is much more problematic, and Gladwell describes how “you end up doing performance evaluations that aren’t based on performance” at least in the McKinsey/Enron world. (Gladwell contrasts this approach by comparing it to companies that do not employ a “star system” and does not rely on the wunderkind.

The third essay, The New-Boy Network, is subtitled “What do job interviews really tell us?” While providing examples from the tech-hiring world, the essay is mainly about what we have learned from psychology research. Specifically, it discusses how we make unconscious snap judgments as to how capable someone is from as little as two seconds of videotape that correlate well with a short meeting you might expect in a job interview. Quoting one of the psychologists, Gladwell writes: “The basis of the illusion is that we are somehow confident that we are getting what is [really] there [from the interview].” Gladwell also narrates his experience with a human resources consultant on “how to extract meaning from face-to-face encounters”. Apparently, the only method that actually has some success in predicting subsequent performance is a highly scripted process. However, most employers choose not to use it. Gladwell dramatically explains why: “We are looking for someone with whom we have certain chemistry… the unlimited promise of a love affair. The structured interview, by contrast, seems to offer only the dry logic and practicality of an arranged marriage.”

Let’s compare all of this to hiring college professors. Since my experience is in a liberal arts college that emphasizes teaching and research, that will be my benchmark. Being in the sciences also has its unique challenges when one is in a SLAC environment. A candidate aspiring to land a tenure track position in chemistry, in this very tough job market, needs at minimum to have a Ph.D. in hand and some postdoctoral experience. One needs to be productive research-wise measured in contributions to research publications and presentations. Some teaching experience is needed, the more extensive the better; and it helps if one is the instructor of record and not just a teaching assistant. The tricky problem here is that if you’re teaching extensively as a graduate student, this takes time away from your research. Many postdoctoral research advisers are also not supportive of their postdocs moonlighting as teachers on the side.

Teaching is a large component at a SLAC, but one also needs to keep up a research program competitive enough to get external funding (albeit at smaller dollar amounts than needed at a research university). At top-flight institutions, one is actually discouraged from spending too much time teaching as a graduate student. One is also usually plugged into a project to produce results, which have to be of a certain quality and quantity to make it into a top-flight journal. This allows the P.I. (primary investigator, or one’s boss) to keep writing grants to support the laboratory research. More publications and grants lead to a higher profile, academic fame and prizes. The typical graduate student does not learn how to manage a lab, or time, in a way that prepares him/her to be a liberal arts college professor. (Neither does the typical postdoc for that matter.) There might be parallels between the situation of scouting out successful postdocs (who were also successful graduate students) to be SLAC faculty members, and scouting out top college quarterbacks for the NFL. It’s difficult to predict success.

Liberal arts colleges are smaller institutions, and therefore it is very costly (not just in monetary terms), to hire faculty who prove unsuccessful and don’t make tenure. It’s a small tight-knit community, and people are hired for their specific complementary teaching and research contributions. (I've chosen to omit discussing service, the third leg of the academic's stool, in this article.) It’s very painful when someone comes and struggles and then has to leave. Mentoring and pre-tenure reviews can be time-consuming, and being in a smaller department means most people have a reasonable degree of involvement in the process. There is also the nebulous process of finding “fit” to the department and institutional culture. One often hears the claim that this is a crucial part of the interview. After shortlisting all the candidates who look great on paper, the phone/Skype interview, (and for those who “fit” best) the subsequent on-campus interview, then lead to a final ranking of the candidates and who might be offered a position.

The winnowing process inevitably involves some degree of considering the candidates’ academic pedigrees (which institutions they went to, and if they worked in the lab of someone famous or a rising star). There is plenty of speculation about the potential trajectory of each candidate and what “problems” might be encountered on the tenure track. We’d like to avoid the problems of The Talent Myth, and I’d like to think that institutional snobbery is not present, but it’s hard to tell. It’s not always easy to differentiate the candidates appropriately.

My department’s phone interview process has a structured component to it, although there is also some degree of flexibility that a seasoned interviewee would know how to navigate well. The on-campus visits are much more open-ended in the sense that the candidate has many individual meetings with the faculty, all of whom have very different styles in “interviewing” the candidate. They are highly unstructured and therefore we can have very different impressions of the candidate (for good or ill). In my younger days, I used to think that I could get a good feel about whether a candidate might do well long-term from the in-person meetings, the research talk, and the one class that the candidate teaches. I’m not so sure anymore.

We have had good success in hiring staff members who prep our lab courses, partly because we have a skills test that involves what they will be doing in their job. I think this strategy works well for a more narrowly and clearly defined job description. On the other hand, our success in hiring faculty who go on to receive tenure and be successful is a mixed bag. When we hired faculty who were on the tenure track at other liberal arts colleges AND had already shown success in both teaching and obtaining external grants, they successfully earned tenure (and continue to do well). Perhaps that is not surprising because they had already shown they can be successful in the relevant environment. (Most of them had been on the tenure track for 3-5 years at a previous institution.) That is my one useful predictor with a limited data set.

When we hire fresh postdocs or those with experience at different institutional types, some do great, but some do not. All had potential and promise (or so we thought), but it’s not so easy to pick out those who would go on to be successful, especially as the performance “bar” continues to be raised. (SLACs are in a ratrace to increase their research profile.) It seems easy in retrospect to point to the early warning signs and struggles, and things that gave us pause when we interviewed the candidate. We think we learn something from these “mistakes”, but how much we do learn is debatable. I have a mental catalog of these, but I’m not sure how useful they are overall. Each case seems unique and different from the others.

What I do know I’m going to do about all of this is learn a bit more about structured interviewing and how to use it effectively.