Weeding-out grad students

  • This has become our largest and most active forum because the physics GRE is just one aspect of getting accepted into a graduate physics program.
  • There are applications, personal statements, letters of recommendation, visiting schools, anxiety of waiting for acceptances, deciding between schools, finding out where others are going, etc.

Post Reply
ComboOrgan
Posts: 16
Joined: Sun Mar 02, 2008 6:39 pm

Weeding-out grad students

Post by ComboOrgan » Tue Mar 10, 2009 3:54 pm

I'm sure a lot of people have heard about Berkely's reputation for accepting more students than they need, then thinning them out through tough classes and hard quals.

Are there other schools that have this reputation? Schools with a high rate of students being forced out?

I heard from someone that Maryland weeds out about 25% of their grad students. Is this true?

admissionprof
Posts: 369
Joined: Sat Feb 02, 2008 7:50 pm

Re: Weeding-out grad students

Post by admissionprof » Tue Mar 10, 2009 4:01 pm

An important factor in considering a graduate school is the attrition rate. What is the number of PhD's per year divided by the number in the entering class per year?

I don't know if Maryland is at 75%, but that wouldn't necessary be lower than most places. Most students who leave a program do so without needing to---some simply don't like research, some find a significant other, some simply feel that spending 5-6 years of one's life getting a PhD isn't worth it. I would expect the national number to be around 75-80%. I've heard, but haven't checked, that Berkeley is much smaller.

Schools with large engineering schools need lots of TAs for the large service courses, and often can't support all of them in research programs. It's easy to check--just ask what the ratio of PhD's to entering students is (and don't forget that there can be huge--and meaningless--single year fluctuations), and ask explicitly why people leave.

User avatar
zxcv
Posts: 402
Joined: Mon Dec 17, 2007 11:08 pm

Re: Weeding-out grad students

Post by zxcv » Tue Mar 10, 2009 4:17 pm

About Berkeley: I have heard that in the not-so-distant past (~5 years) there were strict limits on the number of times one could take the preliminary exams and the department may have had a practice of "weeding out" grad students. This is no longer the case -- I understand that the policies and general practices of the department have changed significantly (even if there are a handful of faculty members who may still think it should be as it used to be!).

This doesn't mean that it's easy to get through or that the department will be holding your hand. You have to take the initiative in finding a research group and making progress, but the serious intent is to get everyone through. I think many schools actually have a far more strenuous set of comprehensive exams and requirements than Berkeley.

There are so many well funded research opportunities at Berkeley in both physics and other departments or labs that I cannot imagine people being admitted as TAs without available spots in research programs. I get the impression that there are more faculty looking for students than students available, although of course this varies widely by field (e.g. condensed matter experiment vs. high energy theory).

I guess my take home message is to suggest that you visit and talk to current students before making assumptions based on hearsay, as in many cases the "word on the street" may not accurately reflect the current conditions.

User avatar
grae313
Posts: 2296
Joined: Tue May 29, 2007 8:46 pm

Re: Weeding-out grad students

Post by grae313 » Wed Mar 11, 2009 11:15 am

zxcv wrote:I guess my take home message is to suggest that you visit and talk to current students before making assumptions based on hearsay, as in many cases the "word on the street" may not accurately reflect the current conditions.
I had heard this rumor about Berkeley as well and when I was at the open house, I was assured by both the faculty and the current graduate students that this was no longer the case. The reputation is left over from many years ago. It was certainly true decades ago, but not any more.

MrKite
Posts: 17
Joined: Sun Feb 01, 2009 8:01 pm

Re: Weeding-out grad students

Post by MrKite » Thu Mar 12, 2009 1:02 pm

My undergrad institution only grants PhDs to 50% of those who start the program - they weed out alot. I just visited a prospective school and they graduate 60% of their incoming PhD prospectives.

From what I have seen, all schools weed out a bit. A 75% graduation rate would be a positive statistic in that respect.

User avatar
dlenmn
Posts: 578
Joined: Mon Dec 03, 2007 10:19 pm

Re: Weeding-out grad students

Post by dlenmn » Thu Mar 12, 2009 3:21 pm

How much of that is weeding on the part of the department? The attrition rate has to be reasonably high even with no active weeding. For example, I believe that here the official statistic is that 97% of people pass the qual within 4 tries. Other statistics (like prelim pass rate, and passing all the classes) are probably similar, but I'm sure the percent of people getting PhDs is significantly lower.

admissionprof
Posts: 369
Joined: Sat Feb 02, 2008 7:50 pm

Re: Weeding-out grad students

Post by admissionprof » Thu Mar 12, 2009 3:28 pm

dlenmn wrote:How much of that is weeding on the part of the department? The attrition rate has to be reasonably high even with no active weeding. For example, I believe that here the official statistic is that 97% of people pass the qual within 4 tries. Other statistics (like prelim pass rate, and passing all the classes) are probably similar, but I'm sure the percent of people getting PhDs is significantly lower.
Good point. If 95% pass the qual eventually, and 95% pass the prelim (if there is one), and 95% pass their classes, and 95% decide they like doing research and thus they stay on, and 95% do not have a psychological breakdown during the six year period, and 95% do not fall in love with someone who is moving elsewhere, and.....

You can all multiply the probabilities as well as I can.

User avatar
dlenmn
Posts: 578
Joined: Mon Dec 03, 2007 10:19 pm

Re: Weeding-out grad students

Post by dlenmn » Thu Mar 12, 2009 4:54 pm

Agreed. By that math, weeding by the department lowers the PhD rate to ~86%. Beyond that isn't really weeding, and is more personal choice (that's what I was trying to say).

MrKite
Posts: 17
Joined: Sun Feb 01, 2009 8:01 pm

Re: Weeding-out grad students

Post by MrKite » Thu Mar 12, 2009 8:55 pm

I asked them that, and they said that most of that weeding is from the quals. Fail twice and your shown the door, and nearly half are failed.

I visited a different school where they wouldnt give me a %, and I have no idea if the two schools I know are representative.

ComboOrgan
Posts: 16
Joined: Sun Mar 02, 2008 6:39 pm

Re: Weeding-out grad students

Post by ComboOrgan » Thu Mar 12, 2009 9:50 pm

MrKite wrote:I asked them that, and they said that most of that weeding is from the quals. Fail twice and your shown the door, and nearly half are failed.

I visited a different school where they wouldnt give me a %, and I have no idea if the two schools I know are representative.
Is it inappropriate to ask which school this is?

MrKite
Posts: 17
Joined: Sun Feb 01, 2009 8:01 pm

Re: Weeding-out grad students

Post by MrKite » Fri Mar 13, 2009 1:55 am

Ill message you. They are top 50 schools, but not top 10.

I wonder now that I think back... each of the professors who told me this kind of had an odd attitude toward me. They might have not been impressed by me and thus painted a negative light on the attrition rate.

User avatar
twistor
Posts: 1529
Joined: Thu Apr 13, 2006 2:47 pm

Re: Weeding-out grad students

Post by twistor » Fri Mar 13, 2009 1:59 am

dlenmn wrote:How much of that is weeding on the part of the department? The attrition rate has to be reasonably high even with no active weeding. For example, I believe that here the official statistic is that 97% of people pass the qual within 4 tries. Other statistics (like prelim pass rate, and passing all the classes) are probably similar, but I'm sure the percent of people getting PhDs is significantly lower.
Don't most schools only give you two tries? I don't know where you're coming up with that statistic.

User avatar
coreycwgriffin
Posts: 249
Joined: Fri Nov 07, 2008 1:28 am

Re: Weeding-out grad students

Post by coreycwgriffin » Fri Mar 13, 2009 5:28 pm

twistor wrote:
dlenmn wrote:How much of that is weeding on the part of the department? The attrition rate has to be reasonably high even with no active weeding. For example, I believe that here the official statistic is that 97% of people pass the qual within 4 tries. Other statistics (like prelim pass rate, and passing all the classes) are probably similar, but I'm sure the percent of people getting PhDs is significantly lower.
Don't most schools only give you two tries? I don't know where you're coming up with that statistic.
Two official tries? I know a lot of the schools I've been looking at let you try as many times as you want while you're taking the core courses, and then give you maybe two tries once you're done with the core courses.

evilclaw2321
Posts: 118
Joined: Mon Dec 08, 2008 5:51 pm

Re: Weeding-out grad students

Post by evilclaw2321 » Fri Mar 13, 2009 5:59 pm

My school also only allows two official tries, once in your first year and then again in your second.

shouravv
Posts: 84
Joined: Sat Feb 02, 2008 1:20 pm

Re: Weeding-out grad students

Post by shouravv » Sat Mar 14, 2009 7:05 pm

ComboOrgan wrote:I'm sure a lot of people have heard about Berkely's reputation for accepting more students than they need ...
My understanding is that this a UC-wide problem rather than Berkeley alone. They need tons of TA's, but can't support them all the way to PhD. Given that CA is leading the race to financial fusion, this is only likely to get worse.
admissionprof wrote:An important factor in considering a graduate school is the attrition rate. ...
I posted these links elsewhere earlier, but again: check the info available in these sites, and by compiling last 10-15 years data for the schools you are most interested in (students in, PhDs out, still in program) you can figure out the attrition rate. For a shortcut, you may simply count the number of terminal/exiting Masters and compare that with incoming class sizes. The latter would ignore the fact that some students may simply leave, even without a Masters.

Physics: http://www.aip.org/statistics/trends/ar ... ysrost.htm
Astro: http://www.aip.org/statistics/trends/ar ... rorost.htm

I wish to do such an attrition rate study at least for the Astro schools sometime soon. In the mean time, any takers for Physics?

User avatar
grae313
Posts: 2296
Joined: Tue May 29, 2007 8:46 pm

Re: Weeding-out grad students

Post by grae313 » Sat Mar 14, 2009 8:10 pm

shouravv wrote:
ComboOrgan wrote:My understanding is that this a UC-wide problem rather than Berkeley alone. They need tons of TA's, but can't support them all the way to PhD.
What are your sources? I heard this from friends and professors, but when I visited, I found out that it just wasn't true. If you don't have some very good and current sources for this you shouldn't perpetuate an out-of-date rumor.

shouravv
Posts: 84
Joined: Sat Feb 02, 2008 1:20 pm

Re: Weeding-out grad students

Post by shouravv » Sun Mar 15, 2009 11:48 am

grae313 wrote:What are your sources? I heard this from friends and professors, but when I visited, I found out that it just wasn't true. If you don't have some very good and current sources for this you shouldn't perpetuate an out-of-date rumor.
Apart from the rumors, I basically depended on AIP data. At first I was trolling all the archive files (linked in my previous post) but then I realized that the "2009 Graduate Programs" publication by AIP that was sitting in my junk paper pile had a 5-year running compilation. So I just compared the perceived-as-top schools with each other.

It should be upon individuals to decide, I guess, if they would consider AIP statistics important or not, because some may simply love a school and find a good adviser. The statistics can often give wrong impressions: for example, I did not apply to a few schools last year since they had a very low international to domestic ratio as well as much higher application fees for internationals. But later I heard that one of those schools actually just decided to start admitting more international students in order to diversify. So, that means, my decision was wrong fundamentally despite being right based on the information available to me at that point.

Anyway, hearsay and statistics should never take precedence over feel-factors, as I have said in other threads (for example, http://www.physicsgre.com/viewtopic.php?f=3&t=2144 and http://www.physicsgre.com/viewtopic.php?f=3&t=1536 ). If it feels right for you during your visits, most probably it indeed is.

shouravv
Posts: 84
Joined: Sat Feb 02, 2008 1:20 pm

Re: Weeding-out grad students

Post by shouravv » Sun Mar 15, 2009 3:03 pm

grae313 wrote:
shouravv wrote:My understanding is that this a UC-wide problem rather than Berkeley alone. They need tons of TA's, but can't support them all the way to PhD.
What are your sources? I heard this from friends and professors, but when I visited, I found out that it just wasn't true. If you don't have some very good and current sources for this you shouldn't perpetuate an out-of-date rumor.
Okay, decided to run the numbers for UC Berkeley based on AIP data, and it suggests that they have a 78% retention rate for Physics and 69% for Astro.

I also looked into UC Santa Cruz Astro, and it seems that they have a 73% retention rate.

These numbers are for most recent graduating classes for which data is publicly available. Similar studies can be done about each schools, I guess; but there is just so much time that can be wasted, even on a weekend ...

UC Berkeley:
..... Physics ....... Astro
Year In ... Out === In ... Out
1994 31 ... 60 === 4 ... 5
1995 39 ... 28 === 5 ... 6
1996 39 ... 24 === 5 ... 8
1997 34 ... 35 === 6 ... 2
1998 46 ... 28 === 7 ... 5
1999 39 ... 34 === 5 ... 3
2000 42 ... 28 === 4 ... 3
2001 40 ... 41 === 4 ... 1
2002 37 ... 23 === 9 ... 3
2003 38 ... 27 === 4 ... 5
2004 37 ... 21 === 9 ... 1
2005 43 ... 40 === 4 ... 5
2006 33 ... 24 === 7 ... 8
2007 31 ... 34 === 4 ... 4
2008 ........... === - ... 1

For Berkeley Physics, for a median Physics PhD length of 7 years, 270 students were accepted between 1994 and 2000, while 210 students got a PhD between 2001 and 2007: that is 78%.

For Berkeley Astronomy, for median Astro PhD length of 6 years, 35 students were accepted between 1997 and 2002, while 24 students got a PhD between 2003 and 2008: that is 69%. (Note: 2001 and 2008 "out" data from dept website. 2001 "in" data is an optimistic (no attrition) guess of "4" since that's the 2007 "out" number.)


UC Santa Cruz Astro:
Year In ... Out
1997 7 ... 6
1998 7 ... 4
1999 4 ... 2
2000 2 ... 6
2001 3 ... 3 [No AIP data]
2002 7 ... 4
2003 7 ... 3
2004 12 .. 4
2005 3 ... 1
2006 7 ... 4
2007 6 ... 3
2008 - ... 7 [No AIP data]

For Santa Cruz Astronomy, for median Astro PhD length of 6 years, 30 students were accepted between 1997 and 2002, while 22 students got a PhD between 2003 and 2008: that is 73%. (Note: 2001 and 2008 "out" data from dept website. 2001 "in" data is an optimistic (no attrition) guess of "3" since that's the 2007 "out" number.)
Last edited by shouravv on Sun Mar 15, 2009 4:37 pm, edited 1 time in total.

shouravv
Posts: 84
Joined: Sat Feb 02, 2008 1:20 pm

Re: Weeding-out grad students

Post by shouravv » Sun Mar 15, 2009 4:18 pm

Here are some schools from other states, for comparison. (Okay, I am indeed procrastinating ...)

Harvard Astro:
Year In ... Out
1995 5 ... 4
1996 6 ... 7
1997 5 ... 5
1998 5 ... 5
1999 6 ... 1
2000 4 ... 10
2001 11... 6
2002 10... 6
2003 7 ... 6
2004 9 ... 6
2005 8 ... 1
2006 10... 4
2007 4 ... 5
2008 - ... 6

1997-2002 Incoming: 41
2003-2008 Graduating: 28
Retention Rate: 68%


Princeton Astro:
Year In ... Out
1995 5 ... 5
1996 5 ... 5
1997 4 ... 3
1998 5 ... 3
1999 2 ... 2
2000 3 ... 4
2001 5 ... 5
2002 3 ... 2
2003 3 ... 3
2004 6 ... 5
2005 5 ... 2
2006 5 ... 2
2007 3 ... 4
2008 - ... 3

1997-2002 Incoming: 22
2003-2008 Graduating: 19
Retention Rate: 86%


Ohio State Astro:
Year In ... Out
1995 5 ... 3
1996 2 ... 0
1997 3 ... 7
1998 6 ... 2
1999 5 ... 1
2000 5 ... 2
2001 3 ... 3
2002 6 ... 4
2003 2 ... 3
2004 5 ... 2
2005 6 ... 5
2006 3 ... 6
2007 4 ... 4
2008 - ... 4

1997-2002 Incoming: 28
2003-2008 Graduating: 24
Retention Rate: 86%


Arizona Astro:
Year In ... Out
1994 6 ... 4
1995 2 ... 5
1996 8 ... 6
1997 8 ... 8
1998 5 ... 7
1999 6 ... 3
2000 7 ... 1
2001 11... 2
2002 4 ... 5
2003 12 ... 3
2004 5 ... 10
2005 6 ... 2
2006 8 ... 3
2007 7 ... 11

1996-2001 Incoming: 45
2002-2007 Graduating: 34
Retention Rate: 76%


Columbia Astro:
Year In ... Out
1995 3 ... 4
1996 4 ... 3
1997 1 ... 1
1998 3 ... 2
1999 4 ... 3
2000 3 ... 3
2001 8 ... 0
2002 3 ... 1
2003 1 ... 1
2004 7 ... 3
2005 6 ... 2
2006 4 ... 4
2007 7 ... 3
2008 - ... 3

1997-2002 Incoming: 22
2003-2008 Graduating: 16
Retention Rate: 73%


Cornell Astro:
Year In ... Out
1995 3 ... 6
1996 3 ... 4
1997 1 ... 6
1998 7 ... 3
1999 5 ... 1
2000 6 ... 3
2001 3 ... 3
2002 6 ... 3
2003 6 ... 2
2004 1 ... 3
2005 7 ... 7
2006 7 ... 3
2007 10...10
2008 - ... 5

1997-2002 Incoming: 28
2003-2008 Graduating: 30
Retention Rate: 107%
Note: Okay, this is an imperfect method, but you get the idea ...


=== Repeat ===

Berkeley Astro:
Year In ... Out
1995 5 ... 6
1996 5 ... 8
1997 6 ... 2
1998 7 ... 5
1999 5 ... 3
2000 4 ... 3
2001 4 ... 1
2002 9 ... 3
2003 4 ... 5
2004 9 ... 1
2005 4 ... 5
2006 7 ... 8
2007 4 ... 4
2008 - ... 1
1997-2002 Incoming: 35
2003-2008 Graduating: 24
Retention Rate: 69%


Santa Cruz Astro:
Year In ... Out
1995 7 ... 4
1996 2 ... 9
1997 7 ... 6
1998 7 ... 4
1999 4 ... 2
2000 2 ... 6
2001 3 ... 3
2002 7 ... 4
2003 7 ... 3
2004 12 .. 4
2005 3 ... 1
2006 7 ... 4
2007 6 ... 3
2008 - ... 7
1997-2002 Incoming: 30
2003-2008 Graduating: 22
Retention Rate: 73%
Last edited by shouravv on Thu Mar 19, 2009 6:14 pm, edited 3 times in total.

excel
Posts: 257
Joined: Tue Feb 19, 2008 2:33 am

Re: Weeding-out grad students

Post by excel » Sun Mar 15, 2009 6:29 pm

Cornell rocks! :mrgreen:

User avatar
quizivex
Posts: 1031
Joined: Tue Jan 09, 2007 6:13 am

Re: Weeding-out grad students

Post by quizivex » Sun Mar 15, 2009 7:09 pm

I praise shouravv for the effort here, but I'm skeptical of this method, even as an estimate. What he has done, is suppose that all students take the median amount of time, 6 years, to graduate. Then all students entering from 1997-2002, and only those students, will be graduating in 2003 to 2008, unless they drop out. Thus, it's easy to calculate the retention rate for that period.

But as we know, most PhDs won't take exactly 6 years. Many will take 4 or 5, or more than 6. Therefore, we get 4 errors in this calculation.

1) Students entering before '97 who took a longer time will be added improperly to the "out" column.

2) Students entering early in the '97-'02 period who graduated early will improperly be excluded from the "out" column.

3) Students entering late in the '97-'02 period who will graduate but are taking too long, will improperly be excluded from the "out" column.

4) Students entering after '02 who graduated early will be improperly added to the "out" column.

That's a lot of errors, and we can't assume they'll cancel out, especially considering the rather small sample sizes involved. Also, the 6-year median is usually rounded up or down, and this throws off what happens at the border years. And the large variation in incoming classes from year to year,esp in smaller astro programs, (PPPL for instance had 8 last year and 4 this year), I'd be careful using this method.

shouravv
Posts: 84
Joined: Sat Feb 02, 2008 1:20 pm

Re: Weeding-out grad students

Post by shouravv » Sun Mar 15, 2009 9:02 pm

quizivex wrote:I'm skeptical of this method ... I'd be careful using this method.
Fully agree with quizivex. It's not even reasonable to assume that quizivex's points 1 and 4 balances points 2 and 3. The only way of doing it absolutely right is knowing exactly who came in and graduated when in each program. Unfortunately, that much data is not available, so I went for the next best thing.

shouravv
Posts: 84
Joined: Sat Feb 02, 2008 1:20 pm

Re: Weeding-out grad students

Post by shouravv » Mon Mar 16, 2009 12:08 pm

quizivex wrote:That's a lot of errors, and we can't assume they'll cancel out, especially considering the rather small sample sizes involved.
BTW, I think this "method" works better for Physics programs since there the sample size not at all small. I was only responding to grae313's concern that may be I was only depending on rumors when talking about UC schools. Since for Berkeley physics it is clear that they are essentially losing 1 out of 4 students consistently (210 graduation from 270 recruits, small number statistics error unlikely), may be the rumors are not as unsubstantial as grae313 was told when she visited.

Or, may be (and I honestly wish so) someone can give us more specific data about the UC schools to prove that they are not weeding out graduate students, and then we'll all be happy.

User avatar
zxcv
Posts: 402
Joined: Mon Dec 17, 2007 11:08 pm

Re: Weeding-out grad students

Post by zxcv » Mon Mar 16, 2009 7:29 pm

Berkeley hires a large number of TAs for introductory courses from outside the physics department, and those are the folks losing their jobs as the California budget gets worse. The department still has more demand for physics TAs than graduate students, by a fairly large margin. If grad students were only being hired to teach, then we would see reduced admittance numbers coinciding with the budget crisis. That's not happening.

Also, all things considered, a 78% retention rate as you calculate is not bad, placing Berkeley well within the norm. Clearly students leave PhD programs for lots of reasons and it's hard to pin down the blame exactly, but clearly rates approaching 100% are impossible for even the most friendly institutions to obtain for reasons entirely outside their control.

anerac
Posts: 17
Joined: Sun Oct 12, 2008 3:18 pm

Re: Weeding-out grad students

Post by anerac » Mon Mar 16, 2009 8:18 pm

This forum had me worried about $$ and qual-weeding so I asked UCSC during my visit. They said they do not actively try to weed students out; if you're a good student and researcher they will find a way to keep you -- and I quote, "The rules may be bent." Also asked about the rumor that you're only funded as a TA til your masters... unequivocally FALSE. They wanted to know where I heard this so they could respond to the rumors.

Berkeley may be different since it's a larger dept.

shouravv
Posts: 84
Joined: Sat Feb 02, 2008 1:20 pm

Re: Weeding-out grad students

Post by shouravv » Tue Mar 17, 2009 8:48 am

anerac wrote:Also asked about the rumor that you're only funded as a TA til your masters... unequivocally FALSE.
It is indeed false. I hope my posts did not give you this wrong idea. It is not unusual for post-qual grad students to be supported as TA's, but they increasingly tend to become RA's/fellows or at least they prefer to be. This is certainly a school dependent factor and cannot be generalized. For an example: http://www.physicsgre.com/viewtopic.php?f=14&t=2204

The only thing that can be said generally is that RA's/fellows have more time for research than TA's. The departments also have less of a moral justification to demand both research and teaching services from post-qual students who are no longer taking grad-classes while only paying them for teaching. However, whether grad students are primarily students or employees is a whole different convulated debate, so lets stay out of that.

bronco199
Posts: 19
Joined: Wed Mar 26, 2008 10:44 am

Re: Weeding-out grad students

Post by bronco199 » Wed Mar 18, 2009 3:55 pm

Anyone know about weeding out at MIT? I know their exams are beastly - but is the idea to get everyone through. Anyone heard anything or have any perspective?

cato88
Posts: 420
Joined: Wed Sep 03, 2008 12:46 am

Re: Weeding-out grad students

Post by cato88 » Wed Mar 18, 2009 5:02 pm

A whole lot of people pass the exams and I think they try their best to keep students.

User avatar
gliese876d
Posts: 57
Joined: Sat Oct 11, 2008 12:51 pm

Re: Weeding-out grad students

Post by gliese876d » Wed Mar 18, 2009 6:58 pm

Maybe I missed it in the thread somewhere but is there a link to the AIP data so I could compare some other schools not mentioned thus far...

shouravv
Posts: 84
Joined: Sat Feb 02, 2008 1:20 pm

Re: Weeding-out grad students

Post by shouravv » Thu Mar 19, 2009 9:15 am

gliese876d wrote:Maybe I missed it in the thread somewhere but is there a link to the AIP data so I could compare some other schools not mentioned thus far...
Physics: http://www.aip.org/statistics/trends/archives/physrost.htm
Astro: http://www.aip.org/statistics/trends/archives/astrorost.htm

astrofan
Posts: 140
Joined: Mon Feb 11, 2008 12:55 am

Re: Weeding-out grad students

Post by astrofan » Thu Mar 19, 2009 11:49 am

Does anyone have any idea of how to calculate a good retention rate from the numbers given. I just can't think of a good method of doing so.

I wish they just gave a survey of how many students leave a program without a PhD.

User avatar
grae313
Posts: 2296
Joined: Tue May 29, 2007 8:46 pm

Re: Weeding-out grad students

Post by grae313 » Thu Mar 19, 2009 8:38 pm

gliese876d wrote:Maybe I missed it in the thread somewhere but is there a link to the AIP data so I could compare some other schools not mentioned thus far...
http://www.gradschoolshopper.com

User avatar
dlenmn
Posts: 578
Joined: Mon Dec 03, 2007 10:19 pm

Re: Weeding-out grad students

Post by dlenmn » Mon Mar 23, 2009 8:54 pm

twistor wrote:Don't most schools only give you two tries? I don't know where you're coming up with that statistic.
I don't know what most other schools do. My statistic (which as I said, is for my program) comes from the department, "In the past five years, 97% of Physics graduate students have passed the Qualifying Examination." (Source). They don't say exactly how it's calculated. (Is that including people who drop out before failing out? What about people who pass at the masters (but not PhD) level and leave with that? Etc.) I doubt that calculating it differently would lead to a drastically different answer.



Post Reply