Now that I'm on the other side of the table as a faculty member, it's been equal parts fascinating and horrifying to see the job search from the opposite end. It's been intriguing to compare notes between the searches (both inside and outside of MIT), not least because they are each so dramatically different. This is the main takeaway I have, really, about the search process from the candidate's side: I cannot underscore enough how much this academic sifting has nothing -- nothing -- to do with the candidate. This is part of the reason I feel so useless when grad students or postdocs ask me about the job market (which I've written more about here): while there are some tweaks that make for more persuasive writing, no amount of editing or strategizing can change the stark realities of academic inequality and scarcity. I say this to illustrate, perhaps in futility, the capricious nature of tenure-track job searches. These are things that candidates cannot and will never be able to control.
However, there are things that departments can absolutely control to make this grueling process a little more humane. Here is a list of those things that I have collected based on my own experience on search committees and as an interviewer. These are not ranked in any way; they're just in the order as I thought about them.
Initial evaluation and sorting
Include salary range in the ad. This is probably superseded by university policy, so there might not be anything departments can really do about it, but I’m a huge proponent of salary transparency (see more here). Including salary range has many benefits which I won't elaborate on too much, but here's a good starter guide from Salary Transparent Street. For more research-focused guides with broader policy implications, there are also detailed studies from the National Bureau of Economics, Harvard Business Review, and *Nature Human Behavior*.
Consider a lighter initial application packet and ask for letters and additional materials in later stages. Based on my conversations with a wide swathe of colleagues, this is a controversial one! I think this is partially because of discipline -- in computer science, for example, there isn't the same kind of customization that candidates do for their job materials compared to someone in the humanities or social sciences (hereafter HSS). In computer science and economics (the only two cases I have some detailed knowledge about), there's a slightly more standardized approach where departments are generally looking for something relatively broad (and are often hiring even in consecutive years). At MIT, for example, the latest job ad for a position in EECS asks for "outstanding applicants with research and teaching interests in any area of electrical engineering, computer science, and artificial intelligence and decision making." This is emphatically not the case in HSS, where job ads are extraordinarily specific and materials are not so easily recycled. Letter requests at these early stages lead to thousands of labor-hours that ultimately result in extremely personalized but quickly skimmed letters. The stakes of letter-writing in HSS, in my opinion, are very dramatic and structural: each letter feels like a fight against the precarity and rampant inequality that is HSS academia, and not just a detailed reflection on the candidate. Recommenders are essentially also writing job applications on behalf of the candidate for the meager job offerings that are on offer each year. If there are 1000 candidates who each had 3 letters which were just 1 page each, that is still three thousand pages -- just letters! -- for the search committee to read. At the beginning stage, there is simply no way to carefully read that much material. Lighter (and in my opinion more humane) HSS searches I've seen ask for cover letter, CV, and research statement; letters, teaching statements, and writing samples are for the second round.
Reflect + document thoughts independently before coming together to discuss. In industry hiring committees I’ve been on, we’ve had to fill out scorecards individually that we couldn’t see until after submitting individual scores. There were also clear guidelines on what we were looking for based on the job ad and we had to rate based on each of those dimensions (not under consideration, thumbs down, neutral, thumbs up, fast track to hiring). More on using rubrics below, but committee discussion is much more generative when everyone has independently done some evaluation, rather than be influenced at the outset.
Use a rubric to ensure some degree of fairness across readers. This is especially important for early stages, in my opinion, when committees are cutting as many candidates as they can to get to a shortlist. While this does introduce a degree of rigidity and arbitrariness to the process (what's the real difference between an average score of 4 vs. 4.5?), I do think being clear about what the committee is valuing makes for more consistent (and faster) reading. As with qualitative research or grading, you could also do a small sample of applications and do some inter-coder reliability checks. This is especially useful since people have radically different approaches to winnowing down candidates -- in some searches, I've heard of committee members only looking at candidate CVs to determine fit with the search description, after which they reject any candidate with less than 5 publications (never touching the cover letter). Others read the cover letter and research statement very closely, but completely disregard the letters of recommendation since everyone is always the best thing since sliced bread. Some put the most weight on the writing sample; others read that last. Being clear about these expectations is especially important for standardizing the committee's approach, especially since there is also deeply coded language to interpret in recommendation letters. Some folks interpret 1 page letters as lukewarm support or take phrases like "they are very dedicated to their students" to imply a weaker research profile. Others disagree. Given this range, it's useful to get everyone on the same page before easily preventable rifts arise.
Clarify expectations among the search committee about answering questions from and meeting with prospective candidates. While a majority of the questions will be directed at the search chair, it's likely that candidates will email different combinations of people asking the same questions. In this case, it's useful to try and mitigate conflicting answers by coordinating beforehand, and by agreeing on a blanket policy for meeting with prospective candidates about the role. I personally declined any request for meetings as a general rule, as it would be logistically impossible for me to actually do all of them and I didn't want to increase the recency / familiarity bias before reading the applications. For this past search in my department, we also put together an official website for the search and posted answers that candidates asked there so that people could have similar information (and to have a single source of truth that the committee could refer to in terms of how we decided to answer certain questions). It also minimized time spent writing and reading email (every little bit helps!), as you can just point candidates to the site. Depending on how transparent you can be, it could also be used for timeline updates, like denoting when the committee has moved from the first cut to the second read). This is especially important in the absence of timely rejection letters.
Send timely, kind rejection letters!! This can be sent at each stage of the process so people don't hold out hope, and to prevent cases where applicants only hear about their rejection when the successful candidate announces they'd gotten the job online. (I know that being able to send rejection letters is often dictated by university HR policy, where departments can only send them after the entire search is done. You also want to be careful about rejection letters for the finalists, since there's always the possibility that the top candidate declines and you move on to the other ranked finalists.) That said, I've seen too many searches end with no rejection letters at all. This is, in my opinion, unfair to candidates who have poured their heart and soul into these applications, and it's relatively easy to automate so a short note feels like the bare minimum. In these cases, no response can often be the worst response. For bonus points, some departments I know also called the finalists who didn’t get the position to offer positive feedback after the fact. I wouldn’t write the “we can offer feedback on your application," but it might be nice for the finalists, even if it's just a short "your research on XYZ was helpful for helping me rethink ABC."
Zoom and campus interviews
Send questions to applicants in advance for Zoom/phone interviews and use the same set of questions for each applicant (minor clarifying questions to follow up are okay, but the point is to give applicants similar experiences and make it easier for the committee to compare apples to apples). This approach prioritizes content and gives candidates an opportunity to put their best foot forward; there are other opportunities to gauge how well they think on their feet.
For the job talk, instruct the candidates beforehand about what the department hopes to learn from their presentation. I think that there’s a lot of hidden curriculum with job talks that the committee should try to mitigate as much as possible. Yes, the successful candidate needs to know how to play the tenure game, but I think making sure that everyone is on the same page (especially across candidates) is important. It also helps set the faculty's expectations when everyone knows what the main goals of the talk are.
For virtual, late-stage interviews, preemptively offer a small stipend for a hotel room, co-working space, and / or childcare. This is not always possible for a number of reasons, especially given this era of financial austerity, but being able to take an important call uninterrupted is something that all candidates should have access to -- not just the ones who can afford it.
For in-person campus interviews, reduce financial burden by not relying on the reimbursement model. I know that this requires coordination across the unit to make sure that the financial details are squared away, but I think it’s important to make sure that candidates (especially grad students) are not worried about their bank accounts when they should be focusing on interview prep.
Proactively address logistics and (in)formal accommodations. Other equity measures for in-person events include communicating a clear itinerary that includes plenty of time for bathroom breaks and eating (!!). This could include giving the candidate a small space to be alone for a small portion of the visit, especially before the job talk. Other important things could include giving candidates a map of the rooms they'll be in (e.g., rooms for job talk, faculty 1:1s, grad student lunch, etc), with gender neutral restrooms clearly marked so they don't have to preemptively ask. It may also be helpful to note in the email that the candidate can send information about accommodations to a department administrator so that candidates do not feel pressured to reveal anything to the search committee.
Don't put the onus on the candidate to walk around campus to meet faculty. In the case that candidates will have to move across campus to meet with faculty members across the university, resources permitting, enlist someone to walk them to and from their appointments, making sure that there's at least 5-10 minutes between each appointment (so each chat ends up being ~20 min). Having the escort check in to bring the candidate to the next place also keeps each meeting on time. Alternatively, have the candidate stay in one place and have faculty swing by to minimize travel (and potential confusion) time.
Make meals a little less stressful by planning ahead. Asking for dietary information beforehand is pretty standard now, but beyond that, I think it's just as important to make sure that the "lunch" slot actually gives the candidate time to eat. I've been in too many grad student job lunches where the candidate is trying to sneak in bites between questions, but realistically, by the end of lunch, they've had a few bites of a sandwich because they've been busy talking (and then they're shuttled off to yet another faculty 1:1 on an empty stomach). In the logistical email coordinating dinner, I would also include language relieving candidates of any obligations to drink, e.g., "note there is no expectation or judgment concerning alcohol consumption: please order whatever you would like to drink with your meal, alcoholic or not." People have all sorts of reasons not to drink -- they're sober, want to avoid alcohol for health reasons, religious practice, or they just...might not feel like it. As a fellow non-drinker, I'm always grateful to not have to explain why I'm not joining in on the bottle of red. It's one less thing for everyone to worry about :)
I'm sure there are many more things to add, but I'll stop here for now. Please send me a note at crystall [at] mit [dot] edu if you have more practices that I could / should add to this list -- with attribution, of course!