Comments on: A Recommendation Letter Word Cloud

By: devicerandom

devicerandom — Sat, 24 Jan 2009 10:53:40 +0000

Well, start from the non-graceful ones. Since I am a fresh molecular biology/biophysics postdoc, I’d like to know them.

By: Graduate

Graduate — Sat, 24 Jan 2009 06:21:28 +0000

Good question, devicerandom. I wish I had an answer for you (I can think of many exit strategies, none of them “graceful”). Maybe those on this blog who are encouraging grad school applicants should try to answer your question.

By: devicerandom

devicerandom — Fri, 23 Jan 2009 21:36:34 +0000

Or, maybe you’ll face the dead end after one or two or three postdoc positions.

In this case, what are graceful exit strategies?

By: Julianne

Julianne — Thu, 22 Jan 2009 18:42:45 +0000

Thus proving that British letters are indeed different than those in the US.

By: Peter Coles

Peter Coles — Thu, 22 Jan 2009 18:19:04 +0000

The best description I ever read in a reference letter was “gormless troglodyte”.

By: Julianne

Julianne — Thu, 22 Jan 2009 03:35:58 +0000

Kyle G — All of the students I wrote for were successful in applying to various programs/fellowships (i.e. the letters were letters I have written, not received). I gracefully bow out of writing letters for students who I can’t back fully. Not every student got into every program they applied for, of course, but all found places they were happy with. So, the word cloud is indeed for successful applicants. And also, I don’t think that “research” and “work” would actually show up in all applications, since a significant fraction of letters are from people who only supervised a student in coursework. Those letters tend to be short, since there’s usually not much to talk about (grade in the class, performance on problem sets vs labs vs tests, interactivity in class, and that’s about it). However, the strongest letters will always contain involved discussions of research, since in that case the student performed a significant enough bit of work that there’s something to talk about!

By: Kyle G

Kyle G — Thu, 22 Jan 2009 00:28:35 +0000

I don’t think it’s very useful to see wordclouds of all recommendations. I mean, far fewer than half of all applicants are accepted into graduate positions for the places where you have access to the letters, right? That means unsuccessful letters are weighted more heavily than successful letters in your model.

Let’s see what successful letters have in them! Or, better yet, let’s see what successful letters have in them once you trim out the top and bottom 5% of words or something. Who cares about seeing a word like “research” or “work” in the cloud? Wouldn’t that show up frequently in almost all applications, successful or not?

Let’s see what adjectives show up, too (of course, getting rid of numbers (which are adjectives) and other things like that)!

By: ts

ts — Wed, 21 Jan 2009 22:31:30 +0000

If you are smart enough to seriously consider a career in academic science, it shouldn’t take that long to figure out what Graduate cited above is true, on average. Assuming the number of positions stay constant, over his or her lifetime a faculty member needs only one successful advisee to replace that position when he or she retires. Anything more will produce an overabundance of “postdocs,” and many professors advise more than one grad student at once. Therefore it is a bad sign when there is an astonishingly high number of people hired at the postdoc level in your field of choice.

By: Graduate

Graduate — Wed, 21 Jan 2009 21:46:52 +0000

Alexander: Of course most people go to graduate school aspiring towards a distinguished academic career, but you’re more likely to complete your training for an academic career and then find yourself at a dead end (unless your definition of an academic career includes adjunct community college faculty). Or, maybe you’ll face the dead end after one or two or three postdoc positions. To paraphrase the advice once given to me by a professor at a major research university: you are unlikely to become a professor at a major research university unless you are “the best student that your professor ever encounters in his or her career.” Do you think you might be that best student? If so, go for it, and good luck to you.

By: GTQ

GTQ — Wed, 21 Jan 2009 05:26:55 +0000

Glad to see some of that charming arrogant humor back on this blog. The whole reason I went into theoretical physics is to become smart enough to justify my arrogance which comes so naturally
to me. Isn’t that the whole point of learning all this stuff about our universe? Now when I sip chai with snooty people, I can bullshit with authority.