Grad admissions

Went through the usual problems with admissions this year, and they once again revolved largely around the GRE. We admitted a couple of folks with GREs well below the average, and some with low GPAs as well. Now, we still admitted less than a fifth of our applicants, but there was a bit of bitterness among the faculty about where exactly to draw the line.

Several faculty members want to de-emphasize the use of the GRE in weighing applicants. It’s difficult for me to argue with this logically. There are good reasons to believe that the GRE is worthless; that is, that it does not predict performance in graduate school beyond the first year (pdf). This would be an especially good time to start ignoring the GRE, since it turns out that the test was compromised in Asia last year.

Unfortunately, grades really are not comparable across universities, and admissions essays are often edited or coached. Ideally, we would like to admit students who really are prepared for graduate school, and who are coming here with open eyes. Dorothea and I don’t agree on many grad school related issues, but it’s hard to take issue with this:

I also think admissions departments ought to select for knowledge of the world and awareness of the grad-school milieu. Quit taking people who are there because they don’t know where else to be. Quit taking people who are there because their parents told them to. Quit taking people who believe in what Joseph Duemer rightly calls Ivory Tower myths. Quit taking people who don’t know what a grad-school B means (it means “wise up, laddie/lassie, you’re in trouble”), or what the academic job market is like, or what the real-world job market is like.

As soon as she tells us how to know, I’m there. As it is, we try to make this clear in our first semester, but by then it is often too late.

Dorothea suggests that real-world experience might be a good indicator. Indeed, I have often preferred to work with students who had some time between their undergraduate and grad degrees. In fact, I usually recommend as much to my best undergrads. But this is also not always the best indicator. I’ve worked with some people who have been very successful outside of the academy who still can’t seem to find their legs in grad school.

Which leaves us with this: we don’t know who will do well in graduate school (or after) until they get here. The question becomes: do we admit anyone who is minimally qualified, then fail early and often (for their benefit as much as ours), or do we admit only a tiny “select” few on what are clearly spurious grounds?

This entry was posted in Uncategorized and tagged . Bookmark the permalink. Trackbacks are closed, but you can post a comment.

5 Comments

  1. Jenn
    Posted 5/18/2003 at 3:14 pm | Permalink

    Ugh, just the thought of the GRE throws me into a frenzy! It’s the only requirement that makes me second guess Grad school. I don’t do well on those tests at all :(

  2. Posted 5/18/2003 at 4:29 pm | Permalink

    One could always try asking.

  3. Alex
    Posted 5/18/2003 at 5:01 pm | Permalink

    I suppose, but you seem to imply that either side elevates this beyond a game. For many students, the application is a test and if they get the right answers they get in. This is particularly true of those who are coming for the wrong reasons. So it becomes… what… a Turing test, I guess. In other words, how would we phrase a question in such a way that it gives us a feel for their *real* motivations and preparation, and not get them to give us back what they think we want to hear?

  4. Amber
    Posted 5/30/2003 at 12:15 am | Permalink

    In my opinion, the GREs are not a sound indication of how someone will perform in grad school at all — as you know Alex, my GREs were not the best, but I received a 4.0 this semester and did well last semester also. I think it was George who said that he leans towards quality letters of rec to clue him into whether a student is admissable or not — another idea would be interviewing canidates who are questionable — I think it is a lot harder to give dishonest answers quickly when you are face to face…..

  5. Posted 5/31/2003 at 5:25 pm | Permalink

    Letters of recommendation are always a good thing, and I think we look at everyone’s fairly carefully. They come with two problems though:

    1) It is very rare to find a bad recommendation: there are good ones and great ones, and you must parse out which is which. It’s usually pretty easy to do this (“He has a great personality and tries really hard…”), but sometimes it is not.

    2) And related to #1, the recommendation is only as good as its recommender. This very quickly becomes an aristocracy. Naturally, we will tend to trust those with whom we have worked or former students of the school to give good recommendations. They both have some knowledge of what our grad program is, and presumably have something to lose, in terms of reputation, if they back a rotten apple.

    The problem is that if you don’t know a former grad or someone that someone on the faculty admires, you are out of luck. The other problem is that sometimes these recommendations end up smacking of favoritism. That is, we end up letting someone into the program not because they are good, but because the recommender knows that they have pull here and can get a student in that wouldn’t get in anywhere comparable. This was *not* the case for you, but something along these lines does happen now and again.

    Of course, the answer to this is to look at the package as a whole. Even someone with a great recommendation from a trusted recommender is not going to be admitted with bad GREs and bad grades. In practice, though (to the consternation of several members of faculty) there is no clear policy on this, two out of three is “enough.” Good grades and recs tend to outweigh lower-than-average GREs.

    I do use the GREs as a “low pass filter.” That is, I won’t even consider someone with any score under 400, and am suspect of those with any score under 500. On the other hand, double 800s don’t do anything for me, unless they are backed by good grades or recs.

Post a Reply to Amber

Your email is never published nor shared. Required fields are marked *

*
*

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>