Social Selection Nature’s snapshot of science on social media

Scientists hunting for academic jobs got a rare glimpse into the mysterious tenure-track hiring process. A blog post written by computational genomicist Sean Eddy at Harvard University in Cambridge, Massachusetts, outlined the steps that he and his colleagues have taken since November to evaluate nearly 200 applicants for a Harvard faculty position. Interviews for six candidates begin this week. A tweet by Eddy on 9 January attracted fresh attention to the blog post, with commenters applauding his efforts to lift the veil on the selection process. Holly Bik, a genomics and bioinformatics researcher at New York University who is applying for jobs, tweeted:

Eddy is a co-chair of the hiring committee for the faculty position at Harvard’s FAS Center for Systems Biology, and only joined the faculty there in July 2015. He says that he wrote the blog post to clarify the hiring process, adding that he is not commenting on Harvard’s recruitment policy. “People don’t get a lot of information about what happens in one of these searches and what the selection criteria are,” he says. “I’m worried about people not applying because they think we’re not going to hire minorities or women or people who don’t have Harvard degrees.”

Eddy described the first step as triage, in which three faculty members (including the two hiring committee chairs) review every application, spending about ten minutes on each one. He wrote that he is looking for a clear research question in the research proposal, and looks through the publication history and attached publications — not to scrutinize journal impact factors or citation counts, but to assess the quality, creativity and trajectory of a candidate’s scientific contributions. “I think a lot of the angsty gnashing of teeth about needing Nature/Science/Cell papers is self-inflicted by the candidates,” Eddy writes. Honours, grants and letters of recommendation count at this stage, too.

Successful applications are then re-read by three faculty members randomly chosen from the entire eight-member committee. They take a more thorough look at the research aims and the publications. In the end, however, Eddy admits: “A lot of it comes down to intangibles, like whether people in the department get excited about a candidate’s research question,” he writes.

Fishing in a self-selected pond

Eddy noted that the applicant group was made up of only 21% female candidates and 5% from underrepresented minorities. Many — Eddy didn’t count the exact number — had done at least some of their training at Harvard.

Bik responded to the blog post, noting that she avoided applying for Harvard positions because of the sheer volume of other jobs that she was also applying for. She decided to place her bets on positions where she thought she had the best chance of success. “I said: ‘I’m not going to apply for this one because I have other applications that I really need to focus more of my time on’,” she said in an interview.

On Twitter, Eddy asked Bik’s advice on how to avoid applicants taking themselves out of the running so that employers can recruit from a broader pool of applicants:

She replied:

Bik also suggested in her blog comment that crowd-sourced job wikis — on which job applicants anonymously compile information about available faculty positions — would be good places for schools to encourage applications from people who might need a nudge.

Being open about bias

In his post, Eddy also wrote that he tries to consider his own implicit biases — such as those against women and minorities in science — during the shortlisting process. One way to measure these biases is with an online test designed by Harvard social ethicist Mahzarin Banaji and her colleagues. Eddy says that he has taken this test a few times — after starting at Harvard and again after reading the hiring committee’s guidelines.

“I used to think that I don’t have such biases. … Now I know I have implicit biases,” he wrote in the blog post. To counteract them, he said he initially evaluated applications from women and minority candidates separately from those from men, and then combined the shortlists — but not to create quotas, he added. “It’s one of the few concrete things I can think of to do in a process like this, to force people including myself to have a conscious, slow, second look at their decisions,” Eddy said. “It’s a work in progress. These are tough issues.”

Bioengineer Ian Holmes, an associate professor at the University of California, Berkeley, who has worked with Eddy in the past, commended him in a tweet:

Holmes acknowledged his own bias in subsequent tweets, adding in an interview: “I find it regrettable that I am biased, but I think there is more shame in not acknowledging one’s bias than in having the bias.”

Eddy’s post resonated with other faculty members who have been involved in recruitment. Joan Strassmann, an evolutionary biologist and professor at Washington University in St. Louis, tweeted:

Strassmann has written for years about the challenges of hiring faculty members on her Sociobiology blog. “I don’t want anyone to not go into [academia] because it seems like a club with secret rules,” she said in an interview. In November, she wrote how the process is inherently unfair. “Our job is to hire an excellent scientist, colleague, and teacher,” she wrote. “There are likely to be others even better in the pool, but not discoverable by our imperfect techniques.”

Bik says that increased transparency in hiring is good for applicants. Whereas some might have insider information about a position because of well-connected mentors, others may have to rely only on what they can find online. “I think that transparency and availability of information are extremely valuable to evening out the playing field.”