Recommendation letter conundrum

In our newest "how can we help you?" thread, a reader writes:

I have a question about the ethics of recommendation letter writing. Some time ago a former student approached me and asked whether I would consider writing a letter of recommendation for his masters application. At the time I told him that I was willing to write a letter, but that I would expect for the student to work with me through the application process. By this I mean, I wanted to see drafts of the student's writing sample and statement of interest. This was in part to improve the student's application by providing feedback, and in part to give myself more exposure to the student's writing which would, in turn, improve my letter.

Since then, the student's revisions have been consistently very poor and hastily thrown together, and I am starting to wonder whether I can, in good conscience, write a strong recommendation. It's too late in the process for me to expect the student to seek out an alternative letter writer. My question, then, is whether anybody has advice on how I should conduct myself: should I be as honest as possible, to the point of (presumably) damaging the student's prospects, or should I perhaps restrict the focus of my letter to the student's course work that they wrote for me over a year ago (which is somewhat better than what they've since shown me)?

I should add that this would be my first time writing a letter of recommendation for a student.

I'm very curious to hear what everyone else thinks, but here are my two cents. It seems to me the thing to do here is to probably sit the student down and be kind but forthright with them, letting them know of your reservations and the reasons for them. See what they say. Maybe they just weren't very aware of how hastily thrown together their work seems--and maybe they'll take it as a big wake-up call to do better. Who knows, maybe they'll go back and put together some very good revisions. That, I think, would tell you something about them (e.g. about their resilience and willingness to work hard). On the other hand, if they respond poorly, then that tells you something about them too. Either way, you have more to go on in making a further decision about whether to still write a letter, and if so what kind of letter to write.

But these are just my two cents. What are yours?

Comments

[I am with Marcus]

I agree with Marcus. You need to sit down with the student.
Twice I was helping students, and the students were getting advice from a colleague who was clueless about graduate school applications because they had gone to grad school so long ago. In both cases I had to tell them that their A level paper from the other professor was going to kill their application. In one case it was too late; but the student did get into a crappy PhD program despite their writing sample. In the second case, the student left my office thinking I was an a-hole, and then applied to law school instead.

[anon]

When I agree to write letters, I've already seen enough work to write the student a strong letter. It seems to me that your agreement here was a bit more conditional than that - it was partly on the basis of how the student would perform in revisions.

I don't think that students typically expect this sort of conditional agreement, and recommend making that sort of conditional agreement very clear whenever that's what's happening.

(For any of my students, if they were revising a paper badly, I'd just say to stop working with this incomplete paper, and use some of your excellent previous work, otherwise you'll harm your application.)

I agree with Marcus - sitting down with the student and talking about your reservations about the paper sounds like a good idea. Make it clear that you want to say good things about the paper in your letter, but cannot at present really do so.

[Amanda]

Related issue: Is it common to get very, very short, recommendation letters for PHD programs. I was surprised when revising applicants this year that a number of students have very short letters. Like 4 lines for one and 6 lines for the others. There were a lot of letters that were half a page. My letters are always 2 pages, and a pretty full 2 pages.

One of the shortest letters I got I was sure was a "not recommendation" recommendation. But then the last line was something about how the student was better than all of their own grad students. With job applicants there is so much to work with that I feel I can trust the other materials. But with PHD applicants there often isn't a lot to go on. It's hard when we are only admitting a very small percentage of applicants and I want to evaluate fairly. I don't want to hold something against a student because they above a lazy letter writer. And these letters were not from Europe or the UK, they were US letter writers. Actually the UK letters I got were all at least one page and were well written.

Also, how many people ask to see lots of, or at least some, material from students they write letters for, i.e. transcripts, a paper sample, etc. I thought this was standard but a student wrote to me and asked for a last minute letter for transferring schools and insisted that none of her other letter writers asked for any material. I also asked a student who wanted a law recommendation letter to send me some material and they never replied. Granted most of the letter I don't talk about the other material, but it can be helpful. But maybe most people don't ask for that.

To the OP: hmm I would feel I have the obligation to write a letter if I already agreed, and I would write it based on their class performance. But clearly I'm an outlier. This depends, though, on the agreement. If it was clear your agreement was conditional, then I guess you don't have to and you can ask the student to turn in better work if they want a good letter. But it seems you should have done this at the first signs of shoddy work, not when the deadline is approaching.