The Jobs that Can’t Be Seen: The Problem with Early Hiring Timelines in Undergraduate CS Departments

I am an Associate Prof of Computer Science at Bucknell University. This post is based on a presentation I gave at a ACM SIGCSE 2021 pre-symposium event by the Committee on Computing Education in Liberal Arts Colleges.

In 2013, I had no idea what I wanted to do with my life. I read advice about industry/academia/everything, but choosing a job wasn’t that simple. Blogs don’t capture who exactly I am and who exactly they are.

I didn’t see my next step until I broadly explored the academic landscape, directly comparing values and tradeoffs, and discovering that some undergraduate CS departments aligned with my priorities more than I ever expected. That process of exploration was so valuable to me that I wrote an entire post about it:

The Jobs I Didn’t See: My Misconceptions of the Academic Job Market

But that path is no longer open, and portions of that post are no longer possible. Today, it is not just that there are jobs that people don’t see, but job processes that can’t be seen.

In 2014, my phone interviews, campus visits, and offers roughly aligned across different kinds of institutions. Those poorly-dawn little symbols are supposed to represent phone interviews (📞), campus visits (✈️), and offers ($)

Hidden Paths: You can’t apply to both anymore.

The crux of the issue is this: most CS departments at undergraduate institutions have shifted their hiring timelines so early that there is no longer any overlap with hiring at research institutions.

Let me give an example: scanning my listings of undergraduate CS faculty positions this year reveals that 33/46 institutions list deadlines on November 15th or earlier. 26 of those require applications before the end of October.

Consider that information alongside the timelines of a research-focused search by Dr. Claire Le Goues in 2012 and a more teaching-focused search by Dr. Kevin Angstadt in 2019 (the snapshots below are excerpts from their excellent CS faculty job guide … and included with permission).

The images above are truncated timelines from the excellent CS faculty job guide, which highlights the difference between a search for an undergraduate-focused position and one that is more focused on research.

In the example above, Kevin received 6 offer deadlines from undergraduate institutions almost 50 days before Claire’s first research university offer. Even if R1s are moving a month or more earlier today, it is still impractical to simultaneously explore research and teaching institutions as a candidate.

In today’s job market, phone interviews, visits, and offers for LACs (liberal arts colleges) often set offer deadlines before phone interviews begin for research universities, forcing candidates to choose one.

Why so early — and is it really that bad?

The why is pretty simple — the shift is a response to the highly competitive market for CS Ph.Ds. Not many job candidates directly target undergrad institutions, so everyone scrambles to get the few that do.

And yet…

While early deadlines enable colleges to aggressively compete for top candidates, they negatively impact both candidates and departments in ways that we don’t talk about enough.

It’s bad because it severely limits exploration.

Diverging deadlines harm candidate decision-making by implicitly obscuring information about the academic landscape. In 2014, I had bad misconceptions about the salary, scholarship, and mission of undergraduate institutions. And the prestige bias of R1s + big companies nudged me towards those it-feels-good-to-brag-about names. If forced to choose one path, I would have chosen the R1 (or industry) route. Fortunately, I didn’t have to do that.

But impact doesn’t end once you’re hired. The goal of a tenure-track hire is not for someone to succeed for just the next 3 years or 5 years, but (hopefully) for the next 20. If you want your new faculty to be happy and valued at your institution, is a foundation that is based on obscuring other opportunities is really the way to ensure long-term retention?

It’s bad because you need insider knowledge.

Relying on insider knowledge disadvantages candidates of resources that are aligned with the center-of-gravity created by research university timelines. This means that those job application workshops may be later than you really need, your mentor might not be expecting to work on that recommendation letter quite yet, or your peers just aren’t thinking about applications because your major conference has a September deadline. These are nudges, but they accumulate.

Part of the problem is that the processes and expectations of non-Ph.D.-granting institutions still aren’t known at many research universities. While I’d like to think that things are getting better, Ph.D.-granting institutions are calibrated towards training students for Ph.D.-granting institutions because… well, that’s just who they are.

All of this means that many advisors implicitly nudge away from undergraduate institutions because it’s not a path or process they are familiar with. And it’s not just senior faculty. It’s also well-meaning new faculty hoping to pass on their job-hunting experience to their colleagues. Remember, you need your statements and recommendation letters months earlier for many undergraduate institutions. The next time you are looking at faculty application guides, reflect at how many of them suggest writing applications a month (or two!) after they are needed for PUIs.

While the institutional bias here is a messy enough issue to grapple with, the immediate consequence is that candidates miss deadlines that they can’t get back. I can’t count the number of conversations I’ve had over the last three years with fantastic, caring advisors who were floored by the early and aggressive timelines of undergraduate institutions.

It’s bad because you can’t negotiate very well.

Early, fast-exploding offers limit competition. If you look back at Kevin’s timeline from 2019, many of the offers dissolved within 10 days or less (!!). This is increasingly commonplace. Listen, being put into challenging, time-sensitive decisions is the norm for job-searching. I get it. But stripping negotiating power is harmful to candidates in terms of leverage and calibration, and more likely to amplify existing negotiation disparities.

After sharing my concerns at a gathering of liberal arts CS faculty at ACM SIGCSE 2021, a colleague surfaced another concern — the early + fast-exploding offers make it nearly impossible for candidates to seek arrangements for partners.

I’ve heard some rational explanations for all of these decisions, but to me, this combination of fast+early offers manipulate the power dynamic in hiring to favor the institution and deliberately disadvantage candidates. We need to do better.

It’s bad because it limits the application pool.

So yes, it’s bad for candidates, but can I flip to the other side for a moment? I worry that we are also limiting our pool of potential candidates. Again, if forced to choose 1 path amid great uncertainty, won’t most job-seekers travel down the well-trodden, familiar paths of their peers and advisors? I suspect so.

It seems to me that somewhere along the line, we implicitly made a decision to prioritize the small number of people who already know they want to teach at an undergrad institution (because we view them as the most desirable?). But within that prioritization, we decided that we were willing to disregard the many applicants who might love it if we gave them a chance to explore it.

Have we thought carefully enough about the tradeoffs and implications of that decision? My sneaking suspicion is that it limits who applies and lands faculty positions in ways which we would not enjoy hearing about.

Change is hard, but we can set expectations.

If I could have it my way, offers would be much later (we shouldn’t be giving out offers before winter break) and slower (no 1-week exploding deadlines).

But change is hard.

Departments are highly incentivized to maintain a competitive advantage. Even more challenging, many departments are bound by institutional policies surrounding timelines… and the cost of not hiring is much higher in a department with 6 faculty than a department with 30 faculty.

So what can we do about it?

1. We should keep identifying bad processes as bad processes. Universities very likely aren’t going to feel a pull to change unless that gravity is created by an external force. Changing the norms will involve changing how we talk about them.
2. We should point to universities that develop more candidate-centered hiring processes. Celebrate, retweet, and share positions that enable candidates to make more informed, deliberate decisions about their next job (I’m willing to use https://cs-pui.github.io/ as a vehicle for this in the future). Candidates, you should interpret hiring processes as a long-term signal of how the university might support (or not) you in other hard, career-defining processes.
3. We should discourage fast exploding offers. Candidates often feel too powerless to push back when given a short fuse (even when they aren’t!). Departments are so terrified of losing their next 2 or 3 candidates that we are willing to put people in extremely challenging situations just moments after parading the quality-of-life at our institutions. It just feels disingenuous to me. We can do better.

There is work to be done here — defining what exactly a candidate-centered hiring process in CS means (that is understanding of the constraints that some institutions might face), deciding how short is too short for offer timelines, and how early is too early for applications. But given the disadvantages, I believe that these are exactly the kind of structures that we can’t just sit back and say well, that’s how it is. They need reimagining and redesign.

If you’d like to chat about this more, please don’t hesitate to reach out to me via email at evan.peck@bucknell.edu or on twitter at https://twitter.com/EvanMPeck.