Stop Asking for Letters of Recommendation


Manimekhala would probably give Phra Mahachanok a good letter of recommendation. We can’t say the same about the other poor bastards lost to the sea.

A letter of recommendation is one of the most useless, costly, and biased instruments for selecting academic staff. Why do we still insist on asking for it?

Every time that a call for PhD, postdoc or any other junior position brings a request like

“…interested candidates should arrange for two or three letters of recommendation…,”

it puts in motion a crazy bureaucratic machine involving a lot of people — employers, candidates, referees — who could probably be doing something more useful with their time.

Letters of recommendation are dictionary examples of subjectivity. They condense in a few paragraphs all kinds of prejudices, personal affairs, and emotional biases in such a way that we have to try to read between lines what the referee really wants to tell with that letter.

A referee fancies his pupil and tries to push her up by exaggerating her qualities. Another referee never really got along with an otherwise good candidate, and gives her a cold out-of-the-shelf letter. A third referee has an unconfessed prejudice against women and gives the candidate a plainly negative letter.

(Just to easy my life through the pronouns, I’m supposing a male referee and female employer and candidate.)

Once I was asked to fill out a “recommendation form” for a candidate to a grant in US. It was a pathetic attempt of lending some objectivity to something essentially subjective. The form asked things like to rank the candidate (5% top, 20%, etc.).

I was recommending that candidate, therefore, I wished that her got the grant. I naturally ranked her at the highest level for all topics. I guess many referees did the same for their candidates, although others possibly took zealously the job of trying to rank someone as “20% top” of an unknown universe.

Serious, which kind of information did the selection team expect to get from all those forms?

We insist on asking for letters of recommendation as result of senseless tradition. They costs time, cause embarrassment and render biased information, which would be discarded by any serious scientist during an experiment. How can a scientist dare to use that information to recruit a new assistant?

Let me strengthen my case against the letters of recommendation by highlighting the kind of burden it brings to each actor involved in this pantomime.

  • The Employer: In academy, we usually don’t count on human resources staff. We do the selection ourselves. When I announce a new postdoc position in my group, I usually get something like 30 applications. If I asked “for two or three letters of recommendation”, this would mean I would receive 60 to 90 letters. Being extremely optimist, if it took me only 5 minutes to read, analyze, take notes, and archive each letter, it would cost me 5 to 8 hours to process all of them.
  • The Candidate: If it’s not already enough embarrassing for a young professional to have to knock at two or three doors for asking for letters of recommendation, just multiply this feeling by the dozen times that she will have to do this again and again. Even an excellent top-10 candidate may have to apply over 20 times before getting a job. (Don’t believe it? I did the math in item 3 of this post.)
  • The Referee: I have a handful of co-workers to whom I regularly give letters of recommendation. Some of them, I did already over a dozen times. I don’t really mind although I know that’s a small waste of my time, as I doubt that most of those letters will ever be read. But at least I write the letters myself. I know people who don’t even do that. They ask the candidate to write her own letter and they only sign it afterwards.

I don’t dispute, however, that a letter of recommendation may bring some useful information (which must be taken with a large dose of skepticism). A praise of some mathematical skills, for instance, may tell you something about the adequacy of a certain candidate to a project. But this is a fine-tunning information for the last steps of the selection process.Then, why should we ask for letters of recommendation to all candidates, even those without real chances of success?

Employers may cut a lot the burden on candidates, referees and themselves by asking for letters of recommendation only to short-listed candidates. I do that myself. From the usual 30 applications that I mentioned, I usually short list about 5 candidates who seem to be the best fit for the job. Only those five are asked to send me letters of recommendation.

Although it’s absolutely possible to do this list based exclusively on their CVs, you may argue that a candidate to a postdoc grant or to an assistant professorship may not have enough achievements to stand out of the crowd yet. It’s true, at these early stages, the CVs are very similar, but it’s a big mistake to believe that a letter of recommendation may help to sort the candidates out.

If you find yourself in a situation where all candidates look just the same, don’t try to distinguish between them using poor and biased information. This is exactly the case where rather than letters of recommendation, you need a pair of dice. I explain that in this other post. MB

  • Subscribe in the sidebar to keep informed about new posts.

Categories: Productivity

Tags: , , ,

4 replies

  1. I totally agree with you, specially for a student’s perspective. One of the hardest things to do is to ask for letters of recommendation. Imagine applying for a Phd in the US. Usually we need to apply for 5 places or more, in order to have good chances of being accepted somewhere (if we don’t have direct contacts in these places). Then, we need to ask up to three professors to write these letters, watching the deadline. If one of them miss the date, the application becomes a lot weaker, which usually corresponds to wasting around $100.

  2. As a candidate off course its annoying…I don’t know about employer.

  3. I totally agree with this viewpoint. I think, most of these letters are biased and very often reflects an injudicious report from three or more letters.

A penny for your thoughts...

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: