17

I am applying for academic promotion and as part of my case I would like to argue that having published x FOCS, y STOC and z SODA papers is impressive. However, I have been told by our head of department that those on the promotions panel will never have heard of FOCS, STOC and SODA and so this will just be compared with the number papers people publish in any other venue. I have further been told that I should provide checkable evidence that these really are important conferences.

How can I do that?

I could quote acceptance rates except I don't believe they are particularly low in these conferences due to self selection.

One might also have thought that recommendation letters would be the best way to judge the quality of someone's work. Sadly we first have to pass a first filtering round in the promotions process before they are requested.

7
  • 4
    "Sadly we first have to pass a first filtering round in the promotions process before they are requested." So your institution's policy is to wait on getting expert opinions until they give their own inexpert opinions, which may lead them not to solicit the expert opinions? You're right, that's sad. I might consider asserting that the conferences are prestigious and giving names of eminent faculty members at other institutions that can corroborate that. There really is no way to check academic prestige without asking other academics, and I would hope that at some level they know that. – Pete L. Clark Jan 9 '17 at 15:42
  • 9
    For those not in CS, it is worth mentioning that FOCS, STOC, and SODA are widely recognized as top publication venues in CS theory, perhaps the top. – Tom Church Jan 9 '17 at 17:46
  • 3
    What role does your head of department play in the promotion process? Typically he or she has a key role to play in advocating for the candidate and communicating to the administration various nuances about the candidate's area and achievements that the administration would not be able to be aware of otherwise - precisely the kind of advocacy that you seem to be in need of. I recommend going back to the head and trying to shake him/her out of their apparent apathy. – Dan Romik Jan 9 '17 at 20:48
  • 7
    Umm... Ask your department head to call me? – JeffE Jan 9 '17 at 21:28
  • 2
    One potential issue is that the "panel" is not so much looking for truth, etc., but is looking for ways to diminish candidates. Some of this can be arranged to be procedural, obviously-but-sadly. And there does often seem to be a sort of xenophobic clannish-ness so that people in other fields willfully misunderstand the context of other fields, so as to be able to "diss" them whenever politically convenient. In that context, you need strong support from your dept chair and faculty, or "the others" may just ride rough-shod over your case. – paul garrett Jan 10 '17 at 0:23
17

I have been told by our head of department that those on the promotions panel will never have heard of FOCS, STOC and SODA and so this will just be compared with the number papers people publish in any other venue.

I find this stance outrageous. It is the promotion committee's responsibility to understand—at least at a high level—the publication landscape in the fields in which candidates publish. Moreover, if the committee is outside your department, it is your department chair's responsibility to educate the committee on the publication culture in your field.

Aside from the conference rankings that Henry mentions, there are other less official rankings of CS departments or just theory groups, based on publication output weighted by prestige. (The rankings themselves should not be taken seriously, but they do provide a window on what the community thinks of the conferences.)

Another method that might be helpful is to gather evidence about where strong theoreticians actually publish. Select a handful of successful theoretical computer scientists in top CS departments and show the committee where those people publish. To pick out a few random DBLP pages off the top of my head: Avrim Blum, Chandra Chekuri, Satish Rao, Santosh Vempala. Choose whatever experts you consider to be your aspirational peers in your subsubfield.

Alternatively, bypass the "prestige" question and argue directly that your work is strong. I assume you are provide the review committee with a research statement. That statement should provide direct evidence that your work is important, visible, and highly regarded. How often is your work cited, and by whom? (Only your subsubsubfield, or a broader set of researchers?) What followup work has it inspired by other researchers? What do other people's papers say about your work or the problems it addresses? Have you received any research awards (for example, Best Papers or an NSF CAREER) or more indirect accolades (for example, invitations to program committees, journal editorial boards, or Dagstuhl workshops)? Are you in Oded Goldreich's list, or in David Eppstein's annual top 10?

I'd also recommend asking your department head whether the pre-letter evaluation is merely a sanity check, or if it has real teeth. My university's promotion process includes a preliminary stage, where the Promotions and Tenure Committee decides whether to request recommendation letters, or not to put the case forward at all. But at least for tenure cases, this stage is essentially pro forma; a decision not to request letters requires a compelling case for failure.

If that's not the case at your institution, I'd strongly recommend pointing the committee to CRA's Best Practice Memo on Evaluating Computer Scientists and Engineers For Promotion and Tenure. In particular:

The primary direct means of assessing impact —to document items (a) and (b) above —is by letters of evaluation from peers. Peers understand the contribution as well as its significance.

3
  • 4
    it is your department chair's responsibility to educate the committee on the publication culture in your field is really the key to the whole problem. This is especially problematic in CS where primary publication venues are conferences rather than journals. Your chair should be telling the committee that these are top-tier places and should be giving the names of a few eminent faculty to corroborate that. – David Jan 9 '17 at 22:18
  • 1
    This is an exemplary answer. – Dan Romik Jan 10 '17 at 1:02
  • In the end I found out what the system was. The first pass filtering does indeed receive no expert advice. The Head of Department is however meant to write a letter of recommendation that should be taken into account by the panel. – donald Feb 23 '17 at 16:01
10

The site Conference Ranks helpfully compiles conference rankings from three sources. Two are "objective" (based on the citations) and the third is "subjective" (based on the opinions of people in the field). (Personally, I think the latter is more meaningful in this case, but your institution might disagree, so it's nice to have both.)

More generally, there's something strange about your situation. Your institution is, surely, aware that publication venues vary widely in quality and must have dealt with this problem in the past. This may be a case where asking a question just the right way will get a more helpful answer: instead of asking what you should do, it might help to ask what previous people have done, and how publication venues have been evaluated in the past.

1
  • 3
    I think this is a good answer. On the face of it, the institution's process is unrasonable. In fact, it is so unreasonable that one presumes there must be workarounds. (Or maybe promotions there are nonexistent or completely random. One would hope the OP would have found that out before now.) – Pete L. Clark Jan 9 '17 at 20:00

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.