Writing Strong Letters of Recommendation for Not-So-Strong Students

If you stick around in philosophy for long enough, you'll almost surely be asked to write a letter of recommendation for one of your students at some point. I did this a few times as a graduate student, but recently I've run into a problem with doing this.

When I wrote letters of recommendation as an ABD graduate student, the students who requested a letter provided me with a resume, personal statement, and a description of the programs they were applying to. They were among the very best students I had ever taught, and they discussed various aspects of their application with me in personal meetings prior to my writing their letter. These letters are easy to write and don't require any embellishment to be an asset to the student's application.

But I have recently learned the hard way that a lot of requests for letters of recommendation do not fit this description. In my current position, I have been asked to write letters of recommendation for students that I barely know, students who only did somewhat above average in my class, and students who provide me very little information about the programs to which they are applying and what they're saying in their personal statement.

Now, in one case, I managed to persuade the student to find someone else to write the letter: I stated in friendly but straightforward terms that I simply did not think that I'd be able to write a strong letter of recommendation for her. But some students won't take "no" for an answer. In one case, I even received an automated request to upload a letter without the student even consulting me about it.

So the challenge is this: what do you do when you've been asked to write a letter of recommendation for someone when you really don't think you can write a strong letter? Refusing all such requests is tempting, but it might not be the best tactic in some cases. Students in STEM fields sometimes get degrees where virtually all their classes are large lectures with hundreds of students and almost no personal interaction with the professor, so if your class is smaller and/or has lots of discussion, then you might be one of the only instructors that the student felt they knew at a personal level. Additionally, the student may have still performed above their normal ability in your class, even if this only resulted in a B+. Getting a letter from another instructor might result in them getting a worse letter than the good-but-not-great letter you would write.

So if you decide to write the letter, how do you go about it? I know letters of recommendation are frequently exaggerated, but I refuse to be dishonest in a letter of recommendation: I'm not going to say the student was the best student in a given class unless they in fact were the best student in that class. So within that constraint, I am curious what strategies others in the profession have for writing good -- or at least decent -- letters of recommendation for their students under non-ideal conditions. Thus far, I have just written the best letter I could with what I knew about the student, but in my estimation, that has not produced the best results. Any pointers?

Comments

[Amanda]

"Thus far, I have just written the best letter I could with what I knew about the student, but in my estimation, that has not produced the best results." - I'm not sure what else you can do. If you feel compelled to write the letter, and also compelled to be honest, that leaves little wiggle room. You should just say what is, in fact, positive about this particular student. All students have some positive aspects. I'm not sure what you mean by "best results", but if you mean the letter didn't help the student get into the program they wanted...well that might not be bad if they were not qualified. Also remember letters are only one part of the application, and students have a lot of other evidence to put forth in respect to their qualifications.

I have to say I am a bit surprised you seemed to have wrote letters of recommendation for philosophy grad school when you were still a grad student? I would refuse to write those, as I know my letter would carry almost no weight. Instead I would try to convince a faculty member to do it, and if they all refused, well...maybe.

[Trevor Hedberg]

Hi Amanda,

So first, the letters that I wrote as a grad student were not for philosophy graduate school. In fact, I have never been asked to write a letter for a graduate philosophy program. Letters I wrote as an ABD grad student were for (1) jobs the students were applying for, (2) applications for scholarships or other academic awards, and (3) law school. Some of these job and scholarship apps were successful. The law school letter was written for the best student I ever had: he had taken multiple courses from me, and I wrote a very detailed account of our interactions and the skills he had that would not show up merely in his transcripts, LSAT scores, etc. Anyway, that student got into a bunch of excellent law programs and later listed me as a character reference for the Bar Exam. Maybe my letter made no difference to the outcome, but the evidence would suggest my letter didn't hurt his chances either -- it's not clear to me he would have been better off with a far less detailed letter from a faculty member that barely knew him, and he told me as much in person when we discussed the matter.

By "best results," I just mean that the letters do not read like great endorsements. If I were reading them, I would think that they sent the subtle message that the student wasn't a great pick -- since letters of rec are often embellished, "decent" letters come across as being weak.

[Michael Cholbi]

A few thoughts here:
- Many students don't appreciate that how well the letter writer knows them matters as much as what the letter writer says. A gushing letter from someone with only a passing familiarity with the applicant is probably no better for the applicant than a positive but nuanced letter from someone with very extensive knowledge of the candidate. So if (for instance) the student who requests the letter only took a single lecture course with you, you might say, "I'm open to writing a letter for you, but keep in mind that I have only very limited knowledge of your abilities. Is there someone else who could write for you who knows you in greater depth?" I also thinks it's OK to be candid with a student: "I could probably write you a letter that would help you at Less Prestigious Program, but I'd have trouble making a case for you being admitted to Highly Prestigious Program" This could be an entry point into a discussion about whether the student's ambitions are realistic.
- Don't hesitate to write a positive letter qualified with the caveat that you have little information to go on. Those reading your letter will be able to surmise that your letter is positive but shallow and should thereby be accorded comparatively less weight.
- If the letter is accompanied by one of those rubrics, mark 'NA' 'No basis for judgment', or whatever -- those looking at the application materials will get the hint that you're working from rather limited evidence

[Trevor Hedberg]

Thanks, Michael. That's very helpful.

[Amanda]

Hi Trevor,

Okay that makes sense, I wrote those letters as a grad student too. For some reason I thought you meant philosophy grad school. Law school I would be more hesitation about, but it seemed to work in your case. I also think law school is a lot different than philosophy grad school, and an ABD letter in general would be much more worrisome for philosophy grad school.

As for your letter coming across as weak - it is supposed to, isn't it? If you are going to be honest, anyway. Depending on what the student is applying to, even a weak letter can be helpful. Sometimes the competition isn't stiff, or sometimes they just really need "a" letter for all their other application materials will do the job.

[Mike Titelbaum]

I always require a student who wants a letter from me to schedule a meeting with me first. I tell them to send me all their application materials, and then we discuss what they’re applying for and what that kind of position wants in a recommendation letter. Finally, I tell them what kind of letter I think I can write for them given what I know of them. If I don’t know the student at all, I tell them that my letter will basically describe the course and then report what scores they got. And sometimes I say something to the student like “I am going to report that you got a B, and were at the fiftieth percentile in the class.” Then I let them decide whether they still want a letter from me, being clear that I’m happy with either decision and won’t hold anything against them either way.

[Paul]

First, my general rule of thumb is that you have to have taken two courses from me and received at least a B+ in each of those courses. (Caveat - I teach in an honors program and the courses are rigorous so a B+ is a good grade.) So, I also focus a lot on the program that I teach, its rigor, and how this program prepares them for med school, law school, or whatever program they are applying to. Perhaps you could do the same with your philosophy courses. I sometimes make exceptions to the above stated rule, say, if a student took my upper division philosophy elective and did very well, wrote an excellent research paper etc. The main thing is that you need details to discuss. Before I implemented this rule I wrote a couple of very brief, lukewarm letters which we all know are not actually recommendations. Now I just have a conversation with that student explaining to them why I can't write them a strong letter.

The other side of the issue is that with most grad and professional schools, I don't think that letters count for nearly as much as data - GPA and test scores. Ask the student for that info, and if they don't meet the minimum requirements, your time is probably better spent having a conversation about vocation with them than writing a letter that probably won't matter. Just my two cents...

[anon]

I do roughly the same thing as Mike Titelbaum above. All my letters include explicit quantitative comparisons, and I try to be clear with students about what I could say about them in that part of the letter.