Sunday, May 07, 2006

Rejection and Rejecting

Luca and Oded talk about rejection. Here are some thoughts.

Rejection hurts. Academics thrive on earning the respect of their peers and it's tough to think that someone doesn't want you. So go ahead and be depressed for a day or two and then move on.

Stanford is my ultimate rejecter, having turned me down for undergrad, grad, junior and senior faculty positions without the least bit of interest. But I got my revenge–I once got a parking ticket at Stanford and I never paid it. Ha!

A few people have complained to me about how rejections letters are written. A rejection letter contains exactly one bit of useful information. The rest is irrelevant and you should not let it get to you.

Suppose Alice sends email to her friend Bob asking if Bob's department would be interested in her. In academics, Bob usually won't give his real thoughts ("We are looking for strong candidates and you are not one of them"), instead he'll find some property P such that Alice has P but they won't hire in P, for example "Unfortunately we are not looking for any cryptographers this year." A couple of warnings for Bob:

  1. Be sure P is not illegal, i.e., based on religion, race, gender, etc. Even if you don't discriminate, saying that you do is not a smart thing.
  2. Bob's department might end up hiring a cryptographer. Then Alice will realize that Bob didn't want Alice because she was a cryptographer, rather he just didn't want Alice.


  1. #2 seems backwards.

  2. Your point that "Academics thrive on earning the respect of their peers and it's tough to think that someone doesn't want you" seems to agree with an essay I recently read (I can't find the link now, unfortunately).

    The essay basically said that science and research pay dramatically less according to the skills required (IQ, training, etc) than most other professions. The essay posited that men are more likely to have science professions because approval from other scientists is important to them. On the other side, women have a broader concept of a peer group and don't need to thrive on the approval. Thus, women are less interested in taking a relatively low paying job than men.

    In other words, women are smarter, and instead get jobs that reward their talents more, like being doctors, lawyers, and vice presidents and CEOs. Men are a little more quirky and enjoy impressing the narrower peer group more.

    This I found is the best reason to explain the "gap" of women in mathematics and science. The real question is why so many men are interested in it, not why are so few women not interested.

  3. Well, call me Alice.

    Bob is being a wimp. He should stop worrying about slighting Alice's feelings and provide some positive feedback that might help with her future.

    Bob would never take marks off a student's exam and then tell the student "oh, we were looking for something else... the grading season was crazy this year... better luck next time... keep in touch...". A junior colleague at painful and confusing crossroads deserves at least that much respect and honesty.

    If Bob is so non-plused with Alice, it is likely that Bob could say something useful... "too many publons, not enough hard problems", "high quality work when you do publish but you just don't publish often enough", "you are too specialized and few people in the department can see themselves collaborating with you", "we don't see your specialty as where the field is headed", etc. Maybe Bob is afraid of offending Alice (more likely, afraid of offending her advisor) but after a bad job search, Alice would appreciate some real feedback.

  4. Actually, Bob should just use a blanket response for all applicants: "Our department likes Destiny's Child, the Apprentice, and people working at the intersection of game theory and CS."

    (It's sad when, for your list of the "hot things" this year, the best you can do is remember pop culture from 2001.)

  5. Hi, this is the parking office at Stanford. Please pay parking ticket for $400. Wacka wacka

  6. I agree with my friend Alice. I wonder why there's a preconception that giving the real reasons would be offensive... On the contrary, the candidate probably has some inkling of them anyway, it gives closure to see them confirmed. What aggravates is uncertainty.

    The uniformity of rejection letters is somewhat understandable because of the abundance of candidates - there is a resource issue... But I believe search committees owe more of an explanation to candidates who have gotten to the interview stage. A friend of mine contacted a liberal-arts college for feedback after he was rejected, and was given the courtesy of a frank and detailed assessment. As far as I'm aware, this is the exception rather than the norm.

  7. Given the litigious state of the U.S. one should not expect the formal rejection letter from a department to have any content in it (other than a polite rejection).

    On the other hand if you know people at an institution if you ask about your application (particularly before the formal letter has been sent) you'll often get an honest appraisal that probably doesn't contain any negative comments about you but probably includes a description of why other candidates were preferred, which can also be helpful.

    The place where feedback is most useful is earlier in the process. If you've just bombed your interview talk then you may have blown your chance at that position, but it would be good if someone where you are interviewing would give you some suggestions for improving it. This might help you at the next interview and makes you indebted to those who've helped so everyone benefits.

    Application packets are much harder to help with since they typically are sent out all at once and any issues are not in the open.

  8. There seem to be two separate issues that could lead to vague answers by Bob. First, if Alice asks Bob before sending in his application, then he may hold back for fear of hurting her feelings, but there are other possible reasons. He might also hold back, for example, because he doesn't know how the rest of his department would regard Alice, and he doesn't want to encourage or discourage her until she has more information.

    It could also be the case that Bob honestly believes the department isn't looking for cryptographers, tells Alice, and then the best cryptographer in 20 years shows up on Bob's department's doorstep. In that case, the department may reasonably hire the cryptographer and incidentally make Bob look like a liar.

    Second, Alice could ask Bob for reasons after being rejected. (This isn't the scenario in the original post, but it's also a plausible scenario.) In this case, Bob is in a bad position, because the reasons may have a lot to do with the hiring committee's evaluation of other candidates -- which Alice has no right to know. Not to mention the fact that Bob's colleagues may not want their opinions shared outside the department.

    By the way, Macneil, the article you're thinking of appears to be this

  9. Oded comments that people should keep in mind that the hiring process is not entirely rational. I have to concur. The longer I've been in academia the more unexplicable decisions I've seen. For every cogent decision by a hiring committee I can name a complety mindblowing gaffe hiring/rejection. I'm sure every department has at least one story of the type: "we rejected Turing-medal-calliber X back when he applied for a job here".

    The same can be said on the candidates side: I've seen candidates pass on some top notch offers only to go to not so good academic places instead.

  10. (Thanks David, indeed that was the essay I was talking about!)

    As for getting the reasons for your rejection: It's certainly worthwhile to ask and try, but you might only get a good reason, and not the real reason.

    Overall, your advisor and other faculty at your school should be able to coach you and provide you with the honest answer you need. And often there are ties between the faculty at schools for the real reasons to filter through unofficial channels.

    Admissions is a probabilistic thing, and I'm sure no one wants to go on record saying "your research suggests you probably got lucky and your great result is a fluke" only to later see that it was just the beginning of many great results to come. It's a small world, and you don't want to burn any bridges with people you'll be working with in organizing conferences and editing journals.

    In terms of interacting with people in the long term, we'd much rather see and give face-saving rejection reasons.

    "The same can be said on the candidates side: I've seen candidates pass on some top notch offers only to go to not so good academic places instead."

    I think for this, you have people look at "top notch" school A and get scared. "Oh gee, professor B was just denied tenure at A, and he does work in my area." Of course, it's much easier to work your way down in prestige than it is to work your way up. It's probably better to go to A anyway, try as hard as you can to get tenure, and if you fail, *then* you can go to the lesser school, where you likely will get tenure.

  11. At the risk of being self-indulgent, I'd like to offer my own rejection experience. Coming out of graduate school, I applied for a small number of academic jobs, got a smaller number of interviews, and was rejected by all of them (including Harvard, where I later ended up). I was fortunate to get a position at Digital Systems Research Center, where I had been a summer intern.

    Looking back, this was far and away the best thing for me, and, not coincidentally, probably the right decision for all of the schools. Fresh out of graduate school I was unprepared for a junior faculty position, and I suppose it showed. The experience I gained at the research lab was invaluable, thanks to remarkable colleagues that taught me a great deal.

    The lessons I took away from this:
    1) Don't be in a rush to get your first faculty position. I think most graduates benefit from a postdoc or a research lab position, a point I've seen reinforced in many books on succeeding in graduate school/your academic career.
    2) The effects of rejection need not be permanent.
    3) Your career can be a long and winding road.

  12. Rejections do not change the intrinsic value of your work, so ideally you should be able to take them in stride...

    Also, surely you don't really want to go to a place that does not really want you :)

    Also, as has been pointed out, by and large rejections are not a judgment of the quality of the applicant's work. Departments are looking for a good match between their perceived needs and the applicants, and their judgment of the applicant's research is only one variable in what is often a complicated function.

    Also, how much does it really matter anyway, as long as you end up somewhere reasonable? The work conditions at the top 50 or 80 schools are actually pretty similar, it seems to me, aside from graduate students.

    The difficult rejections are the ones which force someone out of academic research, or which fore someone to take a sequence of temporary, unstable positions. Those are serious because they have a major concrete impact.

  13. While it is true that it is much easier to move down than move up (in terms of reputation of the department), moving down can also be tricky. The administration at some institutions (such as mine) that are mediocre, but like to advertise themselves as top, generally object to hiring faculty that did not get tenure at a top school. It is tough to claim that you are top when you hire other's rejects.

  14. The work conditions at the top 50 or 80 schools are actually pretty similar, it seems to me, aside from graduate students.

    80 is too optimistic a number. The top 30 or so universities are research focused and indeed aside from the "minor" issue of quality of the grad students coniditions are comparable. The lower you go, the more noise you'll find from incompetent administration and staff. In a research intensive university is very clear that the staff are there to assist the faculty, while in universities lower in the totem pole the staff and administrator whims take precedence over the needs of the researchers.

    Lastly do not underestimate the quality of the graduate students (and colleagues). The same person at a top university will publish 4-8 more papers a year than he/she would at a university ranked between 30 and 40.