Log in

No account? Create an account

Previous Entry | Next Entry

grad school attrition

39orangestreet hipped me to an article in the
Chronicle for Higher Education
about the growing rate of attrition of people in PhD programs.
Thinking back on my own experiences (I got an MS and left) and those
of some other students, a primary concern was money. UCLA CS was not
the best funded program, and it was common for students to get jobs in
order to remain in school. This tended to lengthen people's time in
the program considerably; there are people I know of who've been there
for almost 20 years.

Another point raised in the article was how some students become
traumatized in grad school because this is the first time they've ever
failed at something (e.g. failed orals or failed to complete the PhD
as soon as they'd hoped). Having some personal experience with this
(as an undergrad), I would say that this is something that should not
be underestimated; it can be very traumatic to fail when you have
succeeded (in academics) for most of your life, especially if you
worked very hard and didn't understand why you failed.

For example, I was reading some old email about a math PhD student I
knew who failed her orals for the second time and was asked to leave
the program. In the days leading up to the oral, she was very
enthusiastic and seemed very confident that she would pass. She
seemed confident that her committee thought so as well. But she
panicked during the oral and couldn't answer any questions. Her
committee was very unkind (imho), saying that they didn't think she
had any kind of future as a mathematician. Having panicked during an
(in class) exam, and seen other good (prepared) students do similarly,
it seemed odd to me that her committee just wrote her off so
completely. She wound up giving up on math altogether (at least at
the time -- I don't know if she ever went back to it).

There was something else mentioned in the article about how some
professors didn't have a problem with attrition saying that it was
good that "the weak perish" and so forth. I think this is
cruel, but I have heard some professors and grad students
justify this on the grounds that it is difficult to pull a student
through a PhD program, so they need to make very sure that the
student is going to make it through. I have heard of some cases where
a student was failed by their oral committee simply because they
wanted to be sure of the student's commitment.

Anyway, it's highly unlikely given mine and others' grad school
experiences that I'll be pursuing any PhD studies. Perhaps if I
thought I could get something like my friend
Miriam, who has a PhD in
mechanical engineering and designs satellites for NASA, I'd consider
it, but it's not likely.

A side effect of looking at the old email was finding something from
jonsinger, who I used to see at kabuki-west back in the
day when you could get from the peninsula to Berkeley on a weekday in
under an hour. I friended him; hope he enjoys what he sees here if he
has a chance to read it.


( 8 comments — Leave a comment )
Jan. 15th, 2004 06:22 am (UTC)
Hurray! I've just been talking with two different sets of people recently about what had happened to jonsinger, and now I know! Thanks!
Jan. 15th, 2004 06:36 am (UTC)
Her committee was very unkind (imho), saying that they didn't think she had any kind of future as a mathematician.

This is a horribly tricky situation for everyone involved. Because there are plenty of grad students who faculty can "tell" won't succeed as academics. And while it seems and feels truly cruel to tell thiem this bluntly, I know several cases where nobody says anything to the student and then they experience huge issues four years later when they can't find a job -- or a job they like at any rate -- and the faculty knew all along that this would probably happen. It's sort of like the question of when do you make sure a kid knows they won't play baseball professionally? You don't want to tell them this in little league or it would be cruel, but do you really want to wait until they find out the hard way?

Add to this that most academics are not exactly the best when it comes to certain kinds of people-skills, and it is an issue that is painful for everyone. Faculty and students.

I have heard of some cases where a student was failed by their oral committee simply because they wanted to be sure of the student's commitment.

Now this seems unneccisarily cruel. i see the justification, but i don't agree with it at all.
Jan. 15th, 2004 04:15 pm (UTC)
Hmmm ...

How do you determine whether or not a student is likely to become an
academic? Would all, or even most math professors reach the same
assessment of any given student? Is this something you discuss with
your colleagues? (BTW, do you read sci.math? A topic like this came
up recently regarding someone who wanted to get into a graduate
program. Someone posted something to the effect that if a professor
was giving an evaluation of two students, both getting the same
grades, the student who seemed to have more of a "gift"
would get the better recommendation.)

I guess I haven't given you much to go on wrt my email friend's
experiences, but what do you suppose happened? Do you think her
committee just gave up on her? (As an aside, I did some searching for
her on the web; she's still around, but not doing any math. I guess
what struck me most about her experiences is that they effectively
killed off all of her interest in math.)

As for the analogy to baseball, I think that's hard to say. For very
young children, you don't know what their adult physique will be.
When they get a bit older, you can tell some things about their
coordination and so forth, but there still might be problems, such as
eyesight. When you get to the high school level and above, you can
perhaps weed out certain players who don't have the athleticism or
understanding of the game, but there have been players such as Pete
Rose who were able to overcome a lack of athleticism through a very
disciplined training regimen (e.g. he wouldn't go to movie theatres
because he felt they would ruin his eyesight). There are coaches who
give up on players who don't seem to be catching on as well as other
players, but other coaches have brought marginal players up to high
standards. I guess it is a matter of how much the coach is willing to
sacrifice of their time and energy in a player (above and beyond their
normal job duties), or a matter of how much the individual player is
willing to overcome their shortcomings. But ultimately, when you get
out on the field and play, you can tell for sure who is and isn't up
to par, given a sizable number of games. But with professors, getting
the PhD is a hurdle you have to jump before you can become one,
whereas an athlete might have to go a very circuitous route (semipro,
Japan, for example), but once someone signs them to a major league
contract they become major leaguers, regardless of what anyone else in
their past thought of them.

Getting back to my email friend, you said the student might not find a
job, or might find one they don't like. Perhaps if she had her heart
set on becoming a full professor at a top university, her committee
thought that wasn't going to happen. I don't know for sure what she
wanted to do after she got her PhD; perhaps her committee didn't know
either and that guided their decision.

BTW, I see this topic has popped up in academics_anon. I
may post some thoughts there if I have time.

Jan. 15th, 2004 11:15 am (UTC)
My instinct (having taken the BS and run, for all practical purposes) is that the same type of filter is in place at all levels of education; the openings just get finer as you go up. Some people will never make it to college; some will not make it to grad school; some will not make it out with the PhD. The people in the last group are the ones lucky enough not to have encountered significant failure in an earlier stage, but everyone will fail eventually if you keep pushing.

Just as some people fail to acquire the PhD, others fail to acquire the BS. Their frustration is just as real; it's got to suck to spend five or six years in college before learning that you just aren't going to get the degree. In all cases, what everyone would really like is perfect predictive power during the application process, but that will never happen. Can we get better? I suspect so, and maybe that's where effort needs to be spent.

However, it also sounds like there is an additional factor for the PhD that doesn't apply earlier: the advisor and/or committee. Some of the practices you describe sound pretty cruel and heartless to me, and if I were the student on the receiving end I'd be pretty pissed too. Particularly if my committee lied to me about failing an exam just to see what I'd do! I have trouble seeing a justification for that.
Jan. 16th, 2004 03:46 pm (UTC)
"Can we get better [at predictive power during the application
process]? I suspect so, and maybe that's where effort needs to be

How do you think the application process can be improved?

Jan. 17th, 2004 06:50 am (UTC)
(I'm an occasional reader of academics_anon, and that's how I ended up here.)

My instinct (having taken the BS and run, for all practical purposes) is that the same type of filter is in place at all levels of education; the openings just get finer as you go up.

I don't believe this is true. If I did, I would be much less concerned about dropout rates. There's a number of things you have to learn in graduate school than don't have classes taught on them. You have to learn how to do research, how to give talks, how to find research questions or problems, and so on. The way you're supposed to learn these things, ideally, is through your advisor. But if your advisor sucks (I know some very good researchers who aren't so hot as advisors), or hates you, or is just busy, you don't learn these things.

If your trampling along, trying to blindly get somewhere with no guidance, starting each day with the "I need to make progress" without knowing what progress is, it's really not a fun trip.

Contrast that with having a supportive advisor, who works with you and thereby shows you how work is done, or who tells you to read these papers because, and suggests that X might be an interesting problem, or suggests you talk to Y about Z, and talks with you himself. Someone who is willing to tell you *why* your talks suck and help you fix them, for example.
(Sometimes, the person who does the things isn't an advisor. Maybe it's a more senior graduate student, a postdoc, or a combination of people.)

The person who has a real supportive advisor is going to do better than the blind stumbler, even if the blind stumbler is "smarter" or "better suited to research". I've done both. It makes a world of difference, and I'm the same person in both cases.

Departments who take on students they can't properly care for are, IMO, irresponsible.
Jan. 17th, 2004 04:41 pm (UTC)
I agree, and would add other things to this list, such as selecting a
committee, refereeing papers, and writing a grant proposal.
Jan. 26th, 2004 01:54 am (UTC)
The advisor can make or break a student, yes, and some programs don't pay enough attention to that and the students get screwed. (An incoming student is not in a good position to choose an advisor; he hasn't been around and doesn't know the faculty.)

Some grad-student skills, like research and structuring a vague problem, resemble tasks that some undergrads encounter in smaller doses (group projects, internships, etc). Two of my most valuable undergrad experiences were (1) a software-engineering course where groups had to take projects (that they chose) from requirements through to acceptance and (2) an internship where the project lead din't really know what he wanted from me, so I had to figure a lot of it out on my own. Would I make a good grad student? Probably not. Would I be worse off if I hadn't had those epxeriences? I'm guessing yes.

So I guess we should be pushing harder on getting relevant experience into undergrad programs, so grad school isn't completely alien. After all, even if you never go to grad school, you are still going to face challenges of vagueness, insufficient support, and so on in your career -- you may as well get some guidance on that before you enter the real world.

Can anyone effect this change on a large scale? No clue; I'm not an academic, but large-scale change sounds hard so I'm pesimistic. Maybe grad schools need to place more of an emphasis on that kind of experience (to force undergrad programs to provide, if they want their students to go on). Or maybe we should encourage people to go work as staff on academic projects, or for small chaotic companies, or something like that, for a year or two before entering grad school. I'll admit I'm grasping at straws here, and I realize that no one can just wave a wand and make change happen. So I'm just thinking out loud here.
( 8 comments — Leave a comment )