Monday, November 2, 2015

Guest post: Bad letters

Letters of recommendation for the NSF Graduate Research Fellowship Program are due soon. This year I'm writing letters of recommendation for applicants and it has me thinking about how these letters are written, and who is responsible for bad letters. It also has me thinking about the other side, how can applicants be proactive to prevent bad letters and help with strong letters?  Along those lines, would I ever write a negative letter? A couple years ago, I would have confidently said, "no." Now, I'd like to think I'd refuse to write a letter, rather than write an unsupportive letter. But I haven't been tested with that yet. 

Talking with a colleague got me thinking more about this. 

Below is an anonymous guest post by a colleague, written about an experience with a student after last year's round of applications and responses. 

  Bad Letters

The day after Meg Duffy’s great post about crying in science came out, I was in my office with a student who was crying. She was embarrassed that she had been brought to tears but sadly there had been several stresses encroaching on her life and reading the reviews from her NSF GFRP* was enough to push her over the edge. 

The reason why was that her proposal reviewers had included a comment that her letters did not indicate she had a strong potential for success.  She felt betrayed and utterly at a loss as to what to do.

Although she wasn’t my student, I have an open door policy and often find myself as a faculty member that students go to when they have problems. I’m honored by this, and I take this responsibility very seriously. I try to give good advice, or at least to not give bad advice. So when this student came to me asking what she should do I was at a bit of a loss.

She simply asked, “What should I do?”  She is an early career graduate student, with a strong and diverse undergraduate record, and good grades in a top program. Her letter writers were her committee – the individuals who were most familiar with her work. In theory she did everything right, yet still somehow had gotten these bad letters. She was worried. As she progresses she will need these people to write her letters for fellowships, graduate opportunities, jobs etc. She was feeling like she didn’t know who to trust.

I told her I didn’t know offhand, but I’d be willing to ask around to people I know, trust, and respect and get back to her.  I reached out to several friends and got really good advice, and it boils down to this...

This student has been having trouble getting in touch with her committee. They had only been meeting once or twice a semester as a group and she only saw her PI about once a month.  This lack of communication has brought forth several problems.  First, the student did not really have a clear idea of what her committee wanted.  This means that she was going along her own road, and while she is talented, this may mean that she wasn’t doing the things that her committee wanted, simply because she didn’t hear that from them.  Second, this student wasn’t able to communicate what she needed from the committee. She wasn’t able to advocate for herself, to share her successes, and to craft a plan of attack for her thesis based on their advice.  I don’t know the parties well enough to know who was more at fault.  Basically everyone is busy – I get that. But it was sad to see.

I suspect this lack of communication ultimately lead to the poor letter(s). She probably didn’t have a chance to let her committee know what she’s capable of doing, and she didn’t impress the committee because she did not have a clear idea of what they needed her to do. Communication is variable and important - some students need mentoring to be more than a once a month email.

So given this what can she do?  Has the milk been spilled and are we at a situation where the damage has been done? To some extent yes, but because the advisor/student relationship lasts beyond graduation it is important for students to have a group of solid letter writers who they know they can count on.  After getting good advice from my friends I suggested she do the following:

She should email the committee and try to get a time to talk about the GFRP review, both good and bad.  Walk through it with them and take their advice on how to improve the project moving forward. Also, and importantly, come to the committee and say that she knows they’ve had some trouble meeting and that may have resulted in her not always getting the chance to update the committee on what she was up to, and she didn’t always get to hear from them what it was they needed.  Given that she has X months left, she should ask how she could work with them to get to be where they want her to be.  What is she doing well and what are the areas for improvement?

By approaching this in the context of the grant, with explicit comments to address, rather than confronting the committee with “Why did you write me a bad letter?” the student circumvents a contentious encounter with her committee.  Instead, she comes to them in a framework where she can clearly communicate her needs, and in the same breath, admit that there is work to be done. By showing that she is willing to grow and learn as a student, by showing that she wants to be a better scientist, she is demonstrating, at least to me, the indicators of future success.

However I fully admit I could be misreading the situation.  I’d love to hear what you have to say.

* Terry McGlynn has written very eloquently about inequities in the funding allocation and advantages that students at certain schools get when applying for the GFRP, and I am trying to be mindful of that.


megan said...

I feel like it is good that this student was given feedback by the granting agency that the letters said these things...many students don't know they are getting poor letters. When I have seen unsupportive letters I have often wondered how to ethically get that information back to people. If the student doesn't know s/he is getting poor letters (due to poor communication or whatever the case), how will they ever get funding or a job, or fix the problem?

mathbionerd said...

I agree, Megan. I had a discussion a couple days ago with someone who had a bad letter of rec for jobs. Applied to 50, no bites. One person on one committee gave this person a heads up about the bad letter. The next year, did not ask for that recommender, and had a stunning increase in success rate.

But, I agree. Letters are meant to be confidential, so returning information to a person about those letters would be a breach of ethics (and sometimes, as with NSF, referees can specify that they'll only provide information with the agreement that anything in the letter will be kept in the strictest confidence).

I also don't want to get caught up in thinking all negative letters are bad letters. Sometimes negative things need to be said. The point of confidential letters is to, in my opinion, provide thoughtful and honest feedback about a candidate. If that candidate has faults that pertain to the application, the letter is where someone should be able to point them out. But, parsing that from a vendetta... that is probably nearly impossible.

Mark P said...

Thanks for a thoughtful post on an important issue. I write letters for many, many different reasons, ranging from those for undergraduates in a 200 person class, where I usually can do little more than report a grade, to those for neighboring grad students and postdocs, to those for my own trainees. In the latter cases, I am fortunate to have had a strong group of trainees, and the folks on whose committees I have served are also generally strong, so writing a strong letter has not been difficult. If I feel I cannot write a strong letter, I would begin by suggesting the person consider another letter writer. In most cases, if my opinion suggested the person was a poor fit for the position/fellowship, I could not, in good faith, write a letter.

For my own trainees, it is more complex, as the lack of a letter from a thesis or postdoc advisor in itself would be a negative. In cases like this, I would hopefully already have been in discussion with the trainee about their career goals, and provided advice about positions that seemed appropriate given their mix of skills. For other students (including undergraduates who did poorly in my class or lab), I'd simply decline to write if the letter would not be supportive. I certainly would not write a weak letter for such a person for a job, fellowship or postdoctoral position. Crafting an honest letter in the case where someones record has significant weaknesses is a challenge that each of us are still learning to face.

One other issue worth noting--when faced with a strong set of applicants, reviewers/search committee members are often looking for a reason to remove someone from consideration, and sometimes "read between the lines" in ways the writer didn't expect. This also can come into play with unconscious bias, which lots of data shows influences letter writing for men versus women..

mathbionerd said...

Thanks for your comments, Mark! I agree. I'm new to the letter writing side of the relationship. I'm learning how much more calculated one needs to be when writing these letters.

Anonymous said...

Thanks for this post. There are so many issues around the letters of recommendation (LOR) in academia that need to be discussed, especially because these letters could make or break career of a graduate student or a postdoc. I think there is a lot of influence on letters from the advisors, as there could be, but there are also many many situations where advisors are not fair and give letters of recommendation that represent only their perspective, which could be highly biased. Confidential LORs are fine in an ideal situation but the law does not prevent an applicant from looking at LORs and as long as the hiring person knows about whether the LOR comes confidentially or not, I think LORs should be seen by the applicant. This is sort of akin to peer-reviewers of journal articles having to disclose their identity. The hypothesis here is that an open LOR or an open review 1) will be more responsible one, 2) will give chance to the applicant to voice his/her concern, especially given the gender-based discrimination that can sometimes happen in LORs and 3) it can give real chance for the applicant to see what are the expectations (too high/too low) and 'feelings' of the prospective employer, which is especially useful in the case that the applicant is hired. Perhaps needless to say, I am a victim of a toxic postdoc advisor and hence lack of his recommendation as well as negative recommendation (out of spite) during my job applications and so I have seen the bad side of the process. But I am also someone who got a really good LOR from my PhD committee, so I know the difference that both positive LORs, negative LORs and the lack of LORs made in my case.