Thursday, April 01, 2010


Ask the Administrator: Limiting Letters of Recommendation

A new, but beleaguered, correspondent writes:

I'm hoping that you might have some advice on limiting letters of recommendation. I teach large undergraduate classes and some smaller graduate classes in the humanities at a 4-year college. Students ask me to write at least 100 letters of recommendation each year, ranging from recommendations to be a dorm assistant, summer camp counselor, study abroad student, and intern, to Ph.D. programs. I am sympathetic to students needing letters: they are required parts of many applications, and I suspect that students really wish they didn't have to ask faculty to write them.

The problem is that I am overwhelmed with requests for letters, often from students I barely know. Would you, or your readers, have any guidance on how to address the many, many requests? I have tried pointing out to a student that I don't know him/her well, only to hear that I know him/her better than any other faculty. Writing letters is essentially volunteer work at my school; they are not included in any kind of year-end assessment of productivity.

Would you or your readers have any suggestions on the problem of recommendation letters for faculty?

Been there. On both sides, actually.

I'm not a fan of letters of recommendation generally; if it were up to me, they'd be consigned to the ashbin of history. The references that I've found helpful, to the extent I've ever found any helpful, have been phone calls. Letters of recommendation can date fairly quickly, they assume a much more highly developed common culture than is usually the case, and they've fallen victim to a terrible epidemic of grade inflation. (I don't know how much of that reflects fear of litigation and how much reflects common courtesy, but there it is.) But for whatever reason, many of the opportunities that students want to pursue require them.

One way to limit the impact of letters is to have relatively standardized content. I don't recommend that, though, since standard letters usually read like standard letters; in many cases, the student would have been better off if you had simply refused. An honest 'no' is easier to handle than a half-assed 'yes.'

Another, and one that I've seen in my travels, is to tell students upfront that you don't write letters for students who didn't get A's, and/or who you haven't had for multiple classes. Some sort of bright-line qualifier based on something the students can understand might help cut down the numbers, though there's always the danger of excluding an especially wonderful student you really didn't want to exclude.

You could always refuse altogether, though I'd argue that you'd be penalizing the innocent.

Some people ask students to write the letters themselves, and they'll sign it if they consider it acceptable. In the age of email, you could ask the student to send you an electronic copy to make editing quick and easy -- just tweak whatever needs tweaking, sign off, and you're done. This could allow reasonable customization while still minimizing your time commitment. Of course, you're taking a chance that a student will write something completely absurd, but you could probably head that off with a relatively short checklist of helpful hints that you could give any student who asks.

I suspect, though, that this question is best crowdsourced, so I'll throw it out there. Wise and worldly readers, have you found a way to handle letters of recommendation without either killing yourself or stranding your students? I'll be fascinated to see what expedients people have developed.

Have a question? Ask the Administrator at deandad (at) gmail (dot) com.

I don't have any great strategies for limiting them to under 100...

I do:

1) Not accept requests unless they're made in-person.

2) In that meeting I tell the student the big themes I'm going to put in the letter.
I didn't know the student well, (s)he earned a B in a class where the average was ...

3) During that meeting I make the student tell me about their other academic work and what other faculty they're requesting letters from.

Sometimes I tell them that X would be a better choice than me and I will only do it if X says no.
Develop a standard form asking them to fill in the particulars of their accomplishments. This lets you write it quickly.

I'd also agree that they need to ask in person, meet with you and have had at least two (maybe three) courses with you.

On your form, make sure they include their grades -- and tell them they need a B+ average in your classes to get a letter from you.
Personally I was pretty insulted when my PhD advisor made me write my own letter. Really, you can't think of anything positive to say about me?

If, as Dean Dad suggests, a standardized letter jumps out as a standardized letter, then I think having a standardized (but still positive) one for students you aren't all that impressed with / don't know all that well, and taking the time to write a customized one for students you really believe in actually helps the letters serve the purpose they're meant to.

Those students who have have really impressed or formed good relationships with their professors will then stand out from the crowd in the pile of applications.

So that's the approach I would take. Have a standard form where you get to fill in some blanks about what class you had them in, how they did, how they seemed to get along with others, and a line about how they came to all the study sessions / always turned in quality work on time / showed a real interest in the material (or all of the above, since they apply to many good students.)

For students you especially want to see succeed, scrap the form letter and tell a story that illustrates why you like them so much.

While I understand that a hundred letters a year can, in the aggregate, be very time consuming, for you they are individually just minor inconveniences, whereas for the students, their whole future may be hanging in the balance (at least in their heads). If they are introverted, it may have been hard for them to ask you, and they may not know who else to turn to. So don't just blow them off or send them away, and do make at least some effort to say something nice about them.
If you want to be nice, try a hybrid approach. A standardized letter for the camp/dorm counselor-type position, where Freshman and Sophomores are often seeking them (unreasonable to require multiple courses from the same prof. - they haven't been there that long!), and the personalized letters for the students you really care about. Be honest about which you are giving...a simple, "I haven't gotten to know you that well, so here is my standard form-letter I use. Please fill in the additional details/answer this questionaire, and I'm happy to describe you as a successful (A/B) student" might satisfy a bunch of requests with minimal time.

Form letters for grad. school admissions are better served by answering "no." But then it is reasonable to expect multiple classes with a student, or at least a small seminar...
"Personally I was pretty insulted when my PhD advisor made me write my own letter. Really, you can't think of anything positive to say about me?"

I've heard this from a few people, and I think it's a tremendously wrong-headed response. A good, detailed letter of recommendation will cost someone about 60 to 90 minutes worth of work; filling up one to two pages with detailed and specific arguments is tough, and it's all-too-easy for someone to just forget something that could be a clinching argument in favor of the recommendee.

I notice that in Mary's policy, the professor still gets stuck with the ENTIRE burden of writing these recommendations. Let's say you are writing 100 letters a year, which seems like something well within the bounds of possibility. If only 20 of those are for students you really care about, then that's 20 hours worth of work (BEFORE you have to cope with 15 different and incompatible forms to fill out, either on paper or electronically); if the other 80 "only" take 15 minutes apiece, that's an additional 20 hours. So you've now spent a full working week (once the other time costs are accounted for), for which you are unpaid (although the non-monetary rewards of students' gratitude and success can be pretty great).

Given that Mary writes her comment in the subjunctive, it's pretty clear that she doesn't have to make these kinds of budgeting decisions. It's also incredibly clear that she views these as entitlements. (And, yes, a Ph.D. advisee is in a different situation than an undergrad.) From the wordings she suggests for her proposed letter, it's also quite clear that she doesn't understand the incredible grade-inflation at work; letters that say "she came to all the study sessions" have approximately the same insult-factor as "she was very punctual".

That said, timfc's approach, particularly points 2 and 3, seems too harsh--why not just say no?

Gary King's approach seems to strike me as the right balance:
I normally read out of interest, but I've never commented on this blog. Still, the write-your-own-letter suggestion that has been made kind of forced me to speak up.

Yes, undergrads hate asking you for letters of recommendation as much as you hate writing them. And yes, it's a lot of work, I'm well aware. My dad is a high school history teacher who spends hours every fall writing college recommendations for his students, and he definitely isn't getting compensated for that time!

But we also have absolutely no idea of the proper format for a letter of recommendation. In most cases, we probably gave our college recommendation letter writers self-addressed stamped envelopes and therefore never saw the letter ourselves. Many of the things we're applying for will require a sealed recommendation. We have no idea what we're doing if you ask us to write our own recommendation, and many of us find the very suggestion to be ethically dubious.

So please, if you don't want to write a student a letter of recommendation, say no. Hopefully we'll all figure out that we ought to ask the professors who will say yes, and there ought to be a pattern to those yes answers. In my case this semester, it was the professor with whom I'm taking a second course in a row, whose office hours I visit a couple times a semester, whose classes count for both my majors, with whom I have had in depth discussions on issues unrelated to the classes at hand.

Please, don't tell your students to write their own letter for you to sign. We don't know how to do it, and we are probably even more uncomfortable with that than with asking you in the first place.
Those of us in a position to write letters of recommendation got there because other people were willing to write our letters. We shouldn't forget that.

Some letters take a good bit of time, and they should. But most, for internships and such, don't take that long.

Yes, 100 letters is a LOT. Make sure you save the letters you write to a disk somewhere, in case you need to write for the same student several times over his/her time at school.
When students are asking for more than just a "X was a good student, got an A," I usually ask them for a copy of their resume and, if it's been a while, a copy of a paper they wrote for me. That helps job my memory and gives me a few things to write about. ("Oh, that's right -- she volunteers at the domestic abuse shelter, she was very interested in ethics around women's issues!")

When I've had to request rec letters, I always included a cover letter explaining what I was applying for and why it was important to me, my resume, a copy of the information (school info, job posting, rec form, whatever), and, if applicable, a paper I wrote for that professor, a transcript, or any other supporting information I thought would be helpful.

I do this whether asking in person or asking long-distance, and virtually everyone has been grateful for the supporting material that makes their job easier -- and I think usually impressed that I had everything pre-assembled. (Well, I never descended on a professor with materials in full array; I usually asked after class and arranged to come by during office hours to discuss it if he was amenable; I'd feel bad putting someone on the spot with a ton of information!) When asking long-distance I usually do a sort-of "cover letter" e-mail explaining what I'm applying for and why I think Professor X would be a good recommender, and if they consent I send the supporting info.

But yeah, asking for their resumes helps a lot.
"We have no idea what we're doing if you ask us to write our own recommendation, and many of us find the very suggestion to be ethically dubious."

Yeah, I have to agree with that one. With all of the talk about academic dishonesty/misconduct, how is a student written/faculty submitted LOR any different than a web-copied, student submitted paper?

If the committee evaluating letters of recommendation really wanted to hear the student's words, presumably they'd ask for an essay. Besides, can't a committee *tell* the difference between a letter written by a faculty member who has been doing this for years, and a student who has no clue how to write one?

(As for the argument "well I'm busy" yeah, students are busy too, and somehow that just doesn't justify copying and pasting from the internet.)
I have to admit that the thought of "ghost written" letters of recommendation are a bit concerning to me. It reminds me of all the trouble doctors have gotten themselves into signing their name onto medical studies ghostwritten by pharmaceutical companies. In the eyes of the general public it isn't a question of the accuracy of the information but rather one of bias and a hidden agenda.

Writing letters of recommendation for students can be time consuming, and yes, it can seem quite onerous when one isn't very familiar with a student. I've had to write plenty of them myself, and I am often left feeling like I'm providing the student with a sub-par, form letter. And yes, if I don't know a student very well, I will try to talk to them to see if there is someone else better suited to write the letter for them, but if there truly isn't, then I'll do my best.

I do believe that it's reasonable for a faculty member, or any university representative for that matter, to set a limit on how many letters of recommendation he/she writes. Letters of recommendation are not a right that students can demand, and it's very reasonable for one to decline a request for whatever reason. With that said, remember that there are plenty of students who don't develop deep and meaningful relationships with any of their faculty or university support staff during their education, yet they still need our help. If nothing else, the request from an unfamiliar student can provide an opportunity to talk to him/her about the value of networking and building deeper and richer relationships, as these are important skills that will serve the student throughout his/her personal and professional life.
I really like the:

A or B+ cutoff. You'd be surprised at the weird expectations students have.

Form letter for camp counselor. They're obviously not looking for specifics, just someone to say the kids aren't obvious loonies.

College training in what a rec letter looks like. Most of us who need them haven't ever gotten one.
Let's clear something up. If you're asking for someone to draft a letter of recommendation, then the expectation is NOT that the draft will be signed as-is, but rather that the draft is just that: a draft. I don't see how this is any more "ethically dubious" than the idea that one should have a packet of supporting materials or (as Gary King, above, suggests) a "sliced-bread memo".

It's certainly in no way like the submission of plagiarized work, inasmuch as the substance of the recommendation, and not its originality, is at issue. (Is it any more dishonest for a president or congressman to sign a letter that he didn't write himself?)

(I won't address the high school point, because that's a very different kettle of fish, and because, frankly, the opportunity cost to a high school teacher is different from that of a practicing scholar.)

What's amazing to me is the tenor of the responses to this idea--people are "insulted" or question the ethical integrity of the professor in question. Assuming the professor maintains final editorial control over the document (which she should!), then this is nothing more than a good way of managing both parties' time.

I think we should distinguish between three kinds of "draft your own" letters of recommendations:

1) When the requestor/requestee are in fact very familiar with each other,

2) When they are not, and

3) When the requestor and requestee relationship is purely marginal--in the sense that without the first draft, the requestor wouldn't finish it.

I think that most of the time we are talking about #1, because otherwise the answer should probably be "no" (always in the case of #2, and usually in the case of #3).

The substance of this thread, however, suggests a novel way to bring down the number of letters below 100: only agree to write 99, even if that means saying "no" a lot. And, yes, most of us got where we are because we had someone write letters for us--but that doesn't imply that we have an obligation to write on behalf of B or C students.

I do notice that "Undergrad's" suggestions simply mean that the professors who "say yes" now will quickly stop doing so. And I don't understand why undergrads "hate asking as much as [we] hate writing] them; look, it's part of the game, and we all understand this. The least you can do is write a thank-you card. The most you can do is drop us a line from the law school/grad school/fellowship/study-abroad semester to let us know how things are going.
For academic recommendations, I state upfront my criteria for agreeing to write a letter: (a) you had to have at least an A- in my class; AND (b) the class should have had a research paper for which you had at least an A-; IF (b) is not met, you can substitute another non-research-paper class in which you got at least an A-. This means I generally don't write a whole lot of rec letters.

For work-related recommendations, you had to have worked for/with me on a project to even ask, and I may not agree to write a letter (you would probably know already whether I would be likely to agree).

Love this line: "...frankly, the opportunity cost to a high school teacher is different from that of a practicing scholar"

This reminded me so much of a very specific type of professor with which I've had the pleasure to study.

The kind who realizes that being a research professor is different from being a high school teacher?

I'm not claiming that one is better than the other. I'm saying that one has different demands than the other. These are categorically different statements.

I mostly hate asking because I work under the assumption that the professor has the right to say no, and I'm nervous! Asking a professor to write a recommendation means I'm asking a person whom I highly respect to confirm that they like me and my work enough to want to help me towards my future, and that's a nerve racking experience every time, even when I'm pretty sure the professor will say yes. But if we work under the assumption that the professor should feel free to say no to some students, I always go in assuming there's a chance my first choice of letter writer will turn me down. While it hasn't happened yet, I'm still a neurotic young undergrad, which is a place we've all been in at some point.

I will say- if students don't send a proper thank you note for your letters, they don't deserve a second letter! Has my generation really reached a point where they don't know that every letter of recommendation should be followed up with a thank you note at a minimum? My letter writers this semester (for a combination of internship applications & internship funding applications) all received hand-written thank you notes as well as home baked cookies.
(I won't address the high school point, because that's a very different kettle of fish, and because, frankly, the opportunity cost to a high school teacher is different from that of a practicing scholar.)

So true. I work 10 months a year, 50-60 hours a week. Summers are 'off' — 10 weeks to relax, plan for the next year, write new curriculum documents for my employer, and study (at my own expense) to stay somewhat current in my field. (Figure 10-20 hours a week work, on average.)

Writing letters of recommendation comes out of my personal time. Or my sleep, some weeks.

I don't know how this compares with a "practicing scholar".
I serve on a scholarship board so I see thousands of letters of reference a year. The few dozen that I write each year don't feel like that much of a burden but I do require students meet me to discuss their reasons for requesting the reference and their self-perception, as well as getting some idea of what I can write on their behalf.

I have been honest when students with "C" averages and C work in my classes approached me for references that I can only speak to their accomplishments in truthful terms. Some still find that suitable (mostly if they need references that are for summer jobs, I suppose.)

But the number of letters of reference that I've seen torpedo promising applications? Sadly, that's pretty high. If you're too busy to write a letter that says something individual and positive about the candidate (besides the killer terms of punctual and reliable or industrious!), say no to the request rather than have them work hard on the application that your faint praise undermines.
I think that part of what helps with requests for letters is, as others have noted, having very clear instructions about a) level of performance/acquaintance with the professor, b) necessary supporting materials that will allow the professor to write a strong letter, and c) how and when the request should be made. If you outline what you need in order to even consider writing a letter up front, that cuts down on the number of requests: students will go to another faculty member who *doesn't* give those clear guidelines because it seems easier.

I have something on my web page that outlines the proper way to ask for the letter, the materials that I need in order to write a strong letter, as well as the timeline for when students should ask for letters. What this means is that students who ask me have a totally transparent document from which to work as they consider whether I'm the appropriate person to recommend them, and it also teaches them how to approach *other* faculty for letters of recommendation. (Basically, my format is very similar to the one that my honors college in undergrad gave me, and especially if one is teaching a lot of first generation college students, they really don't know the protocol for asking for such things unless somebody gives them specific directions.)

I have never asked a student to draft a letter of recommendation for me, and personally I have trouble with that, especially if the recommendation that I'm submitting is supposed to be confidential (as in for grad school or other post-grad degrees like law and MBA). On the other hand, I don't think it's asking too much to expect that students submit a paper that they wrote for me, a description of the thing they're applying for and their reasons for applying, and a brief cv.

I have templates for three different kinds of LORs for students - a grad school template, a scholarship/other (study abroad, internship, dorm, whatever) template, and a professional school (law school primarily) template. Once the students provide me with supporting materials, I personalize those templates with their information, and with my own reflections on the student. I change adjectives, I change the specific information about the work that the student did for me, but the structure is all there. And if I'm writing multiple recommendations for the same student? Writing the initial letter takes 1-2 hours, but subsequent versions take about 5-10 minutes a piece. I do not start from scratch with each and every letter.

But to conclude, the best advice I have about limiting the requests is just to say that you will not write recommendations for students who give you less than 6 weeks notice. Yes, you'll still get some who beg you for a last-minute recommendation, but that number will be fewer, and if it's a student you don't know well or that you couldn't write a strong recommendation for, you don't need to make that your reason for rejecting the request: you can just point to your policy about time of request, and they will accept it without trying to argue about how you know them better than anybody else or how they really did the best work that they've ever done for you.

Oh, and one last thing, 100 letters? That's a LOT. If I took each letter individually that I wrote this year, my number is somewhere around 30 (love that for law school apps it's one letter that is entered into a database!). 100 is just too many.
Yikes! 100 letters is a lot!

Honestly, though, I write a fair number of letters myself--although nowhere near 100! And of course it makes a dent in my ability to do other stuff that's part of my job. But the recommendations are part of my job, too; they're not some outside thing intruding on it. Yes, of course there are downsides--it's never fun to try to write a thoughtful a letter at 2am because the student got into a bind and needs the letter TOMORROW--but that's not a good situation for anyone, least of all the student who is trying, also at the last minute, to prepare an application that is surely more complicated than a single letter. And on the rare occasion, absolutely, I just can't say yes because the only honest letter wouldn't be a strong recommendation. But limiting letters of recommendation for the sake of limiting them doesn't sit well with me. For me, at least, it's part of the job--just like lecturing or working with students on their papers or helping students plan classes for next semester or sitting on committees or picking new textbooks, writing recommendations is part of the job. Often--like when you get to sit down and remember that awesome student from a few semesters ago, or when you discover just how fabulous your current student is through listing his or her achievements in writing--it's even a really fun part.

Then again, perhaps it becomes much less fun once you get to 100 each semester.
For undergraduates, I cheerily write letters for internships, internal awards, and other things where it's clear that being an A or B student and having enthusiasm are crucial. Those letters take little time, and once you've written one, it's easy to revise, changing what's necessary, in good conscience.

When undergrads apply to grad school, or grad students apply for postdocs or jobs, I'm choosier: I ask myself, is this someone for whom I'd be willing to put my reputation on the line? If not, I tell them that they should find another recommender who can be more positive. If the answer is yes, I take the time to write a letter that emphasizes the student's best qualities. Once I write it, it's on my hard disk and I can easily produce updated versions when necessary.

Most students thank me for recommendations. Some don't, but then, my undergrad professors probably wrote many recommendations without being thanked. I feel that I should pay it forward: that is, write recommendations for students who seem to be where I was at their age, and maybe for others too when they show promise.
I was recently asked to write a letter for a student who has been in two of my classes and was an average student. I wrote a letter of recommendation for him. Afterwards, I became aware of some issue that I did not know about. I would still have written the letter but it would have been "toned down" a bit.

Going forward, I will advise students in the beginning of the semester that I will not write letters of recommendation for students currently in my claass. This doesn't "sit well" with me in terms of being objective in the grading process. No will I write letters for students having a pattern of poor attendance. I will also require that requests be sent by email and I wll then let the student know if I wil be writing a letter for them. For the most part, I can say I have been glad to write letters for studeents.
As a career counselor, I constantly wrestle with this issue, as I am involved in many internship selection committees. Our faculty complain about the number of recs they are required to write, but when they serve as reviewers, they greatly value the insight of their peers (most of the letters I've seen are very detailed and thoughtful). And therefore, when an application is submitted without a letter of recommendation, they blame the student, even though the faculty submit letters directly to the career center, and the students have no way of confirming they've arrived. Many students do follow up with us, but they have no recourse if the professor hasn't submitted the letter. They are (understandably) reluctant to nudge a professor either before or after the deadline. So I entreat you: whether you use a template, ask the students to write their own letters, require an in-person meeting- whatever - PLEASE SUBMIT LETTERS ON TIME.
Two factors make me hate writing most LORs. First, LOR-creep, the constant increase in the number of places that require students to obtain LORs. Second, and somewhat relatedly, is the declining quality of students who now need LORs. The declining quality of students seeking LORs is caused by at least three things: 1)LOR-creep, 2)grade inflation, and, 3)education expansion. Grade inflation has made many more students think they should go on to more schooling, and educational expansion has made many more students believe they have to go on to more schooling. The number of students who should be in grad school has not increased, but now they are a smaller proportion of the number of people requesting LORs even for grad school. All this has greatly reduced the "pay it forward" sense of the task most of the time.
I have read your blog its very attractive and impressive. I like it your blog.

Java Training in Chennai Core Java Training in Chennai Core Java Training in Chennai
I have read your blog its very attractive and impressive. I like it your blog.

Java Online Training Java Online Training JavaEE Training in Chennai Java EE Training in Chennai
Post a Comment

<< Home

This page is powered by Blogger. Isn't yours?