In our December "how can we help you?" thread, a reader ('G') writes:

I am a junior faculty member, and I started to get requests from students (undergraduates) for letters of recommendation for different purposes. I am wondering if people can share some tips, principles, and/or strategies for writing letters. I am especially curious about people's thoughts on the following questions:

1. If a letter only mentions positive things about a student, would it be less credible? If I need to mention some potential concerns, how should I do it? Should I be specific or not?

2. Do you use different strategies for different purposes, such as graduate school, scholarship, etc? I've learned that some people tend to be more positive if a letter is written for a scholarship (some only mention positive things about students). Is this the common practice?

3. How do people write letters for average students (below-average sometimes)? I know that some people just politely decline the request. But if I need to do it, how should I write such a letter?

Great questions! I'm very curious to hear everyone's answers. Here are mine:

On (1): I don't know whether people find a letter less credible if it only mentions positive things. But, for my part, this concern is a reason why I normally do include some discussion of specific areas where I think a student could improve. It has always seemed to me that if I only trumpet an applicant's virtues, then I've done the admissions committee (and applicant) a disservice. No student is perfect, and if my letter pretends they are (by omission), then it seems to me that someone reading it could read the letter very skeptically–particularly if the areas the applicant might need improvement in show up elsewhere in their application (e.g. their writing sample). So, I normally do include some notes in my letters where applicants can improve.  When I do, I normally carefully qualify it something like as follows: "Like all students, X does have a few areas where they could improve. For example… However, I want to emphasize that they are good at X, and an excellent all around candidate for admission to your program…"

On (2): I've only written letters for graduate admissions, so I can't quite comment on this.

On (3): Generally speaking, I won't accept a request to write a recommendation for a student I believe to be below-average. In those cases, I'll usually tactfully suggest that the student seek out someone else. I have written letters for average students, but usually ones who I've seen made substantial strides through hard work. When I write these letters, this is usually what I focus on: the student's upward trajectory. I'll usually also give the applicant some heads up too. For example, when they ask, I might note in my email response, "I'd be happy to write you a letter. However, I do want to note that you've generally received B grades in my courses, so my letter would have to reflect your overall performance." On a few occasions, I think I've had the applicant decide to ask someone else for a letter instead–but for the most part, my experience has been that they're usually still happy to have me write them a letter (usually, I think, because their performance in my classes is fairly representative of how they've done in their other classes, as well). In any case, when I do write letters for such students, I try to go about it tactfully but honestly, writing things like, "My experience was that X was a good but not exemplary student. However, their performance in my courses improved significantly over time, and I believe they would succeed and flourish in your program for the following reasons…".

I have no idea whether other people go about things differently, and whether there are readers who think that I'm going about things wrong on (1) and (3)–though I will note that the students that I've written letters for (including average students) have done well at getting into programs. Anyway, what do you all think? How would you answer G's three questions?

Posted in

13 responses to “How to write good recommendation letters?”

  1. anon

    On (1): Maybe I’m wrong about this, but I’ve always thought of my recommendation letter as a chance to explain what I think speaks in favor of their admission, rather than as my summary of all the aspects of the student.
    So I haven’t included “need to improve” areas, but also didn’t think of that move as a deception by omission. And my students seem to get into competitive programs. But I’m interested to see if I’m out of step with the norms here.

  2. grymes

    I have the same practice as anon–I take them to be letters of recommendation, not letters of evaluation. But I’m not confident I’m getting it right: I haven’t served on an admissions committee–and am particularly in the dark about law school admissions, which is what I mainly write letters for–so I will watch this comments section with interest.

  3. anon ancient

    Count me also as someone who writes letters only of recommendation, not evaluation. I take it that my job is to do absolutely everything in my power short of lying to get this student into the best program they possibly can, and thus I take this to be best-served by developing an account of their good attributes as best I can.
    But I also don’t know whether this is effective (it has worked at least a few times) and/or an industry standard and thus will have an eye on these comments to see what others think.

  4. another anon ancient

    I also tend to write letters of recommendation, rather than evaluation. I agree with anon ancient – it’s my job to get my students into the best possible program. I take this seriously because I work at a no-name place, which means that my students are already at a disadvantage relative to those coming from name-brand schools and R1s with famous professors. This strategy has worked.
    On (2), I highlight different characteristics depending on what the letter is for. For law schools, MBA programs, and other professional programs, I tend to include things like ‘leadership’ and ‘communication’ skills (they love that crap), and also explain more clearly why their education in philosophy is directly relevant to what they’re going to study.
    On (3), I’ve never had a grad school letter request from a bad student – I think they know better. I’ve done a few for study abroad, but that’s pretty low stakes – so long as I don’t think they’ll set anything on fire or end up in a foreign prison, I’ll do it.

  5. Marcus Arvan

    Interesting reading all of these comments. I’ve always agreed that it’s my job to get my students into the best possible program. But I’ve always thought that the best way to do this is to offer a candid evaluation of their background, abilities, and promise (including areas where I think they can improve).
    To clarify, I’ve always thought this a good strategy as a letter writer for two related reasons: (1) it could convince the reader that the letter isn’t biased/inflated (a worry that I have heard a lot of people raise concerns about–namely, letters not being very useful because they are all overwhelmingly positive), and (2) it could serve to ensure that my letters are taken more seriously across applicants.
    Here’s what I mean on (2): I’ve always worried that if I write letters that just ignore areas where the applicant could improve, then if that applicant gets into the program and the program ends up having a problem with that feature of the applicant, then future application committees at that program would be more likely to discount my letters with respect to later applicants (viz. “I remember Arvan overselling candidate X in 2019. Why would I trust his letter on applicant Y in 2020?”).
    The way I’ve figured it, I serve applicants and the programs they apply to the best if I’m candid about a candidate’s strengths, while not completely ignoring their weaknesses (though, again, carefully discussing the later so as not to harm the candidate). Judging from the comments thus far, it seems like I’m the only one who does this. But, my more general worries about recommendation letters aside, isn’t this trend a problem (viz. people thinking letters are artificially inflated)?

  6. Chivers Butler

    I have never written a letter in which I mention a student’s weaknesses and am surprised to hear that others do, though Marcus provides a good rationale.
    Whether I write a letter for an average student depends upon what the letter is for. I’m happy to write for scholarship applications and for any program that I honestly believe they have a shot at getting into. Though, like others, I always encourage them to ask those who can write the strongest letters. Sometimes students will ask those that they are most comfortable asking, even though they could get stronger letters from others.

  7. anon ancient

    Good points, Marcus, and definitely some food for thought.

  8. anon

    Marcus, RE: (1). I don’t think mentioning a bad feature of an applicant is the best way of convincing readers that the letter is honest. Instead, the evaluation should just be backed up by enough of a concrete reference to the applicant’s work and accomplishments that it’s clear that your positive claims about the applicant are true, or at least plausible.
    I also think that in an environment where so many letters are so positive, mentioning a bad feature of an applicant could backfire. Rather than making the reader think, “Wow, what an honest letter!” you might make the reader think, “If he feels the need to admit this much, maybe there are even more problems with this applicant!” I guess both reactions are possible, and the question is which reaction is more likely.

  9. Marcus Arvan

    Anon: I get that concern for sure. But for what it’s worth, the students I’ve written letters for have tended to do very, very well about getting into their preferred programs—so at least anecdotally, it doesn’t seem to have hurt their chances. And I guess I still worry about less-than-candid letters backfiring in way (2). Every student has unique weaknesses. I would hate for a student to get into a program, struggle there because of the issue (perhaps in part because the program doesn’t do a great job mentoring students like them), and have people in the program think my recommendation letter was “false advertising.” Anyway, if it seemed like it was hurting the students I recommend, then I would definitely change the practice. But in all honestly, virtually all of the students I’ve written letters for have gotten into their most-preferred program, so ¯_(ツ)_/¯

  10. Evan

    Maybe that’s the price for implementing something without proper thinking before hand. It’s one of those negative unintended consequences. Just try to be as concrete and specific as possible with your description. Vagueness will most likely make the committees more doubtful.
    There are also lots of pros and cons with letters of evaluation. I am very curious to see how that plays out of if were implemented. This could be interesting for research.

  11. By the letter

    I must say, I am more similar to Marcus in my letter writing. I do express worries or weaknesses about candidates. I might call them challenges. In fact, I supported a very good student for a grad program, but I had to say that he was quiet in class (too quiet). The student did not get into one program which he should have – but that could be due to many factors. But he did get in to a decent program with good support, and after the faculty there expressed gratitude for sending him their way. I was concerned that I did not want to set my student up for failure, by disappointing them, and, like Marcus, undermining my own credibility.

  12. I follow Marcus’ procedure (evaluation, not just recommendation), for the same reasons he does. In my case, since I write lots of letters–I teach at a terminal MA program where many of the students wish to go on to PhD programs–maintaining my credibility across multiple applicants is especially important.
    Because I don’t want to hurt the chances of people I’m writing for by doing this, I have a piece of boilerplate I append to the start of each of my letters. It basically says “Praise inflation makes letters less useful. I want my letters to be useful, so I’m open about both the strengths and weaknesses of people I’m writing for. Please keep in mind that I’m willing to mention weaknesses even of people I think very highly of.”
    I have no idea how this comes across to people reading my letters–it very well might seem odd and self-indulgent–but the people I’ve written letters for have, on the whole, seemed to do very well with their Ph.D. applications, so it doesn’t seem like my letters are hurting their chances.

  13. Evan

    If one of the aims of letters of recommendation is to get the students into graduate school, then it seems both the formalistic recommendation and “soft” evaluation can be effective from anecdotal evidences. In terms of on the whole, I don’t know since I don’t have any research on that.
    If you find yourself having more negative than positive things to say about the student, then it’s probably best to not write it and have somebody else who may know the student better write it since it might defeat the purpose of the letter. I think most professors are well-meaning and want what’s best for their students and want them to succeed in graduate school. There will always be tensions between maintaining integrity and honesty and wanting them to get accepted even if they may do poorly.
    But then again, isn’t that what a 4+ year of undergraduate supposed to prepare students for? The next level of education as well? Success isn’t solely individualistic. It’s also environmental.
    As much as I am sympathetic to the worry of potential individualistic failures and weaknesses of students, I won’t brush under the rug the responsibility of prior educational institutions in helping students overcome those weaknesses early on in the first place. Maybe having better preparation for graduate school during undergraduate years can make letters of recommendation so much easier to write.
    It’s something to consider.

Leave a Reply to By the letterCancel reply

Discover more from The Philosophers' Cocoon

Subscribe now to keep reading and get access to the full archive.

Continue reading