Skip to content

How many minutes should you spend writing a letter of recommendation?

Helen DeWitt, commenting on about friends/colleagues/acquaintances who ask her for reference letters, writes of “a mythical entity: a reference that can just be dashed off in half an hour and popped in the post / fired off in an e-mail. There is no such thing.”

Jenny Davidson follows up with:

I [Jenny] do not know why someone thinks that it is possible to write a good letter of recommendation without a HUGE amount of supplementary paperwork . . .

What’s my experience? I get asked for a fair number of letters of recommendation or evaluation, and I take about 15 minutes to write such a letter and email it off (to someone who prints it on letterhead paper and mails it). From the remarks above, I suspect that it’s considered a norm to spend more time than that, but I think it’s a bit of an arms race: your letter has to be long so it can compete with other people’s letters. So by writing short letters, I’m doing my part to make the process more sane. There’s a well-known statistician who always writes letters for his students saying essentially that they’re the second coming of Cauchy; he’s recognized for doing this. As long as people have the expectation that my letters will be short, everything should work out fine.


  1. dani says:

    once I read about John who got a letter from a professor at CMU when looking for admission for a Phd in Math.

    The professor did not take long to write a good letter.

    he just wrote one sentence: "This man is a genius." … in that case no exaggeration.

  2. dani says:

    I mean John Nash.

    sorry for the confusion.

  3. derek says:

    I think a letter of reference has two requirements: it should show that the applicant isn't lying about having been a student of Andrew Gelman, and it should show that Professor Gelman would not be positively embarrassed to have helped persuade the hirer to take the applicant on.

    A short letter, provided it's on the right headed notepaper, fills both those requirements fine. I suppose a longer one would show that the writer cared enough for the student to have put the time in to help the student's career, but then it might just show the writer is a soft touch.

    One annoying thing letter writers do that they shouldn't, though, is identify the student by a generic first name only. They should take care to identify the student uniquely by name.

  4. Helen DeWitt says:

    I don't think it's a question of length. When a grant-awarding body takes up a reference, they ask things like: How long have you known the candidate? How much do you know about the candidate's work?

    If this is a grant for writing fiction, I may well think that X's e-mails and blog show a remarkable talent for writing and be willing to write a reference on that basis. If X has 10 partial novels and 50 stories on his hard drive, though, it would be much better if I could say that I have seen a generous selection of the kind of work X might actually submit for publication. (At the risk of stating the obvious, X doesn't need a ref from me to write a blog.) If X declines to send this material, I can a) state that I have only seen samples of X's writing on his blog and in his e-mails, or b) lie. a) is, to my mind, weak. b) is, to my mind, unethical.

    It may be, for all I know, that nobody who understands the process actually cares about these things. I don't need to fret about the gap between the quality of evidence they're asking for and what I actually have; I can just sail right in and testify to the genius of X.

    The problem is, though, that this actually takes up a lot of mental space. There is a SIMPLE procedure in which a grant-awarding body asks how much relevant evidence I have and the candidate provides relevant evidence and I state that I have relevant evidence. Then there's some thing, some process under which those in the know know they don't have to worry about evidence, they can ignore whatever it is the GAB said it wanted and just rave. But I don't want to allocate mental space to the possible gap between what the GAB actually needs and what it says it needs. I want something very very simple. They ask a question. I answer the question. I get back to work.

  5. Andrew Gelman says:


    But couldn't you just spend 15 minutes writing that you think X's e-mails and blog show a remarkable talent for writing, and that X also has 10 partial novels and 50 stories on his hard drive which you have not had time to read? As long as everyone knew this was coming, I'd think this would be fine.

    Of course, if the committee is paying you to do the evaluation, or you feel a personal obligation, that's another story.

  6. Jenny says:

    I don't spend an excessive amount of time actually _writing_ the text of the letter – it's more the cluster of obligations/responsibilities around it that make it an unusually stressful and attention-consuming job. I think what I really need to do is put together a very concrete list of what I must have in order to write a good letter. The trouble is that for the pattern/prototype I have in mind (which doesn't differ as much as one might think based on the stage of life the recommendee's in, except that grad students on the job market need significantly longer and fuller accounts of the dissertation project), there is no substitute for 2 things: (1) very full and complete first-hand knowledge of the student's comportment in classroom, paper-writing and preferably department/college/university as well and (2) CV and personal statement so that one is able to be precise and accurate in details and sensible in emphasis. Once I'm in the position of writing for someone I haven't taught in a couple of classes, the whole thing becomes much more difficult, and in fact these are the situations in which I am also likely not to have such a clear idea of the person's prior educational record, etc. etc.

    Andrew, the thing about your suggestion ("So and so's e-mails are brilliant, I have not had time to read bulk of other work") is that this is a letter that actively undermines the application. I write ones like this now and again, but only after making it clear that the student should find someone else who knows them better if humanly possible – when I am at the other end of the process, I am partly looking to see that the applicant has built suitable relationships with letter-writers (preferably involving knowledge over an extended period of time, and real mentorship of some kind), and a letter like this is a red flag. I'm talking here about grad school or job applications, though, and I understand the issues may be somewhat different for grants – not, I think, though, as different as all that…

  7. Doug says:

    The coolest recommendation I ever received was when my advisor sat me down at his computer and dictated it to me!!! Only took a few minutes. Then I took the file and did what was necessary with it.