TestMagic

YoungEconomist · 06-08-2008, 05:59 PM

Is there anyway we could start a thread which has a listing of various schools attrition rates? It seems that this info would be extremely relevant when people are deciding where to attend. I imagine that many people on this board know the attrition rates at a few institutions, and if we could get this information all on one thread, I think many TMers would appreciate it. Hopefully, if we got a good number of schools on there, we could even have a sticky for it, because it seems like just as informative as the Profiles and Results threads. Anyways, it was just a thought, and I'd love to hear feedback on it. If we got a good number of programs, I'd even be willing to put all the information in an excel spreadsheet and post it.

trjohnson · 06-08-2008, 06:46 PM

Attrition rates are only useful to know if you have like 5 or 10 years of cohort data. Taking rates just a year at a time per school isn't a good idea, because a lot of schools have a lot of variance in their attrition rates as years go by. Besides, you should be going to the school with the best program (on the margin) and not basing your decision on something like an attrition rate. I'm of the opinion that if you're that worried about how many people fail out in a given quartile of rankings, you may want to be aiming for a lower quartile of schools.

YoungEconomist · 06-08-2008, 07:07 PM

Quote:

Originally Posted by trjohnson

Attrition rates are only useful to know if you have like 5 or 10 years of cohort data. Taking rates just a year at a time per school isn't a good idea, because a lot of schools have a lot of variance in their attrition rates as years go by. Besides, you should be going to the school with the best program (on the margin) and not basing your decision on something like an attrition rate. I'm of the opinion that if you're that worried about how many people fail out in a given quartile of rankings, you may want to be aiming for a lower quartile of schools.

I tend to disagree with a lot of this.

First, it seems like knowing recent attrition rates is probably better than knowing the last 10 years. I don't care what Wisconsin's attrition rate was 10 years ago, I know it's high in recent years.

Second, how can you say that attrition rates don't matter? If MIT failed 75% of those taking the prelims, but somewhere like Minnesota only failed 10%, are you really telling me you'd still go to MIT. I think attrition is a very reasonable thing to be concerned about, and will likely play a role in the offers I accept. Some schools fail about 50% of those that take the prelims, and I would consider that a pretty risky proposition, and I'd definitely be willing to drop down in the rankings a bit if I could go somewhere where it was only 20%.

GymShorts · 06-08-2008, 07:18 PM

Quote:

Originally Posted by YoungEconomist

Second, how can you say that attrition rates don't matter? If MIT failed 75% of those taking the prelims, but somewhere like Minnesota only failed 10%, are you really telling me you'd still go to MIT. I think attrition is a very reasonable thing to be concerned about, and will likely play a role in the offers I accept. Some schools fail about 50% of those that take the prelims, and I would consider that a pretty risky proposition, and I'd definitely be willing to drop down in the rankings a bit if I could go somewhere where it was only 20%.

I agree. If I was torn between two similar programs, and found out they have greatly different attrition rates, that could be a deal breaker for me. It's valid info.

I do think that attrition from a single year may not be a good indicator, but I don't think you need to go back 5-10 years. I think 3-5 would suffice.

TruDog · 06-08-2008, 07:41 PM

Quote:

Originally Posted by GymShorts

I agree. If I was torn between two similar programs, and found out they have greatly different attrition rates, that could be a deal breaker for me. It's valid info.

I do think that attrition from a single year may not be a good indicator, but I don't think you need to go back 5-10 years. I think 3-5 would suffice.

I agree completely with GymShorts. Programs (or at least prelim committees) can change considerably in a period of a few years, so attrition rates are generally meaningless after about five years. Likewise, looking at just one year can exhibit extreme variance, so looking at 3-5 years is your best bet.

I've said this a couple times before on this forum, but it does seem like Wisconsin's attrition rate is dropping somewhat. There were a few years in the early 2000s during which attrition was almost unreal, but now it's getting to where something closer to 15-20% of all students cannot pass prelims. However, a larger number of students seem to leave the program here before taking prelims (although not in my class). My theory is that Wisconsin takes more chances on students than most other schools, which is great for people like me but also can put people in way over their heads.

I'll just keep studying for prelims so I can help bring down that attrition rate.

asianeconomist · 06-09-2008, 08:59 AM

Hey TruDog, is there a correlation between people who fail the comps and those who were not offered funding ?

TruDog · 06-09-2008, 10:29 AM

Quote:

Originally Posted by asianeconomist

Hey TruDog, is there a correlation between people who fail the comps and those who were not offered funding ?

It seems like the few students on fellowship tend to do better on prelims, but these are the students who are expected to be the best in the department. Otherwise, those who get TA or research positions really don't seem to do any better on prelims than those not offered funding at all.

asquare · 06-09-2008, 01:31 PM

When thinking about attrition, it's important to consider attrition at different stages of the PhD program separately. I don't mean to suggest that we can determine who is "to blame" for various types of attrition, but I do think departures at different stages convey different information about the departments.

Some people leave during the first year of the PhD, before the first attempt at prelims. IMO, that sort of attrition is entirely personal, either because of some personal issue that came up during the year, or because the student didn't have realistic expectations about the PhD program, or because they just changed their minds about what they want to do. I don't think this type of attrition has much to do with the department at all, and therefore isn't particularly informative for the prospective student choosing one school over another.

Other students leave after failing the prelims once, but before being asked to leave. Attrition at this stage is harder to parse out. Students who leave at this stage aren't taking advantage of all of the opportunities afforded to them, but they may feel squeezed or pressured out, or inadequately supported. This is especially true if they are denied funding in conjunction with failing exams.

Attrition as a result of failing prelims and being asked to leave is what I think most people have in mind when they ask about attrition. (I'd lump people being asked to leave because of extremely low grades into this category, also, though that's much less common.) There certainly is information about the program in the ultimate pass rate for the prelim exams. However, people who have posted earlier are correct in suggesting that 1) a single year's pass rate may be an outlier and doesn't provide much information and 2) there can be substantial medium-run trends and other policy changes within programs are correct.

The last category of attrition I would worry about is attrition among students who have reached candidacy. These students have completed coursework and exams, but for some reason don't succeed in receiving their PhDs. This can (though doesn't necessarily) suggest that the department doesn't provide adequate support throughout the research process, that advisors are in short supply, that funding for upper year students is scarce, or that there is a poor environment in the department. Of course, some students don't finish for reasons entirely of their own -- they decide that while they were good at course work, they don't like research, or "life" happens and they need to relocate, earn money, or devote time to their families.

I'd encourage you to ask departments about attrition at various stages of the program. Ask about what the department views as its role in reducing each type of attrition (or if the department even thinks it needs to be reduced), and how the department supports students in each stage of the program. Consider your own strengths and weaknesses -- a student with a strong math background and excellent test taking skills, but no research experience or who knows him/herself to be undisciplined, might worry more about a program that has a low completion rate among students who achieve candidacy than about a program with a low prelim pass rate. The opposite might be true for a student who is worried about the course work, but has substantial work and research experience and is more confident about that stage of the program.

In general, you want to think beyond the first year of the program. The prelims seem like a huge deal until they are over, and then they are completely irrelevant...

(PS. Good luck to those of you studying now!)

YoungEconomist · 06-09-2008, 05:08 PM

Quote:

Originally Posted by asquare

When thinking about attrition, it's important to consider attrition at different stages of the PhD program separately. I don't mean to suggest that we can determine who is "to blame" for various types of attrition, but I do think departures at different stages convey different information about the departments.

I totally agree. I'm most worried about the % of people who do not pass prelims and are asked/told to leave. IMHO, I believe this is the most telling. Those that drop out during the first year are probably unprepared, don't like graduate econ, personal issues, life happens, etc. However, it's disheartening to know that various bright, hard working students, that work hard during the first year, and learn a lot will be asked to leave because of the prelims.

asianeconomist · 06-10-2008, 01:44 PM

Quote:

Originally Posted by asquare

The last category of attrition I would worry about is attrition among students who have reached candidacy. These students have completed coursework and exams, but for some reason don't succeed in receiving their PhDs. This can (though doesn't necessarily) suggest that the department doesn't provide adequate support throughout the research process, that advisors are in short supply, that funding for upper year students is scarce, or that there is a poor environment in the department. Of course, some students don't finish for reasons entirely of their own -- they decide that while they were good at course work, they don't like research, or "life" happens and they need to relocate, earn money, or devote time to their families.

Thanks a lot for the useful post.

I would have thought that this type of attrition would be extremely rare, or is it not ?

06-08-2008, 05:59 PM	#1 (permalink)
YoungEconomist TestMagic Guru Join Date: Dec 2006 Posts: 1,198	Attrition Rates Is there anyway we could start a thread which has a listing of various schools attrition rates? It seems that this info would be extremely relevant when people are deciding where to attend. I imagine that many people on this board know the attrition rates at a few institutions, and if we could get this information all on one thread, I think many TMers would appreciate it. Hopefully, if we got a good number of schools on there, we could even have a sticky for it, because it seems like just as informative as the Profiles and Results threads. Anyways, it was just a thought, and I'd love to hear feedback on it. If we got a good number of programs, I'd even be willing to put all the information in an excel spreadsheet and post it.

Thread Tools
Show Printable Version Email this Page
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

06-08-2008, 06:46 PM	#2 (permalink)
trjohnson Eager! Join Date: Apr 2007 Posts: 62	Attrition rates are only useful to know if you have like 5 or 10 years of cohort data. Taking rates just a year at a time per school isn't a good idea, because a lot of schools have a lot of variance in their attrition rates as years go by. Besides, you should be going to the school with the best program (on the margin) and not basing your decision on something like an attrition rate. I'm of the opinion that if you're that worried about how many people fail out in a given quartile of rankings, you may want to be aiming for a lower quartile of schools.

06-09-2008, 08:59 AM	#6 (permalink)
asianeconomist Loving the game Join Date: Jul 2007 Location: Where the grass is greener Posts: 884	Hey TruDog, is there a correlation between people who fail the comps and those who were not offered funding ?

06-09-2008, 01:31 PM	#8 (permalink)
asquare TestMagic Guru Moderator Join Date: Mar 2006 Location: Ann Arbor Posts: 1,311	When thinking about attrition, it's important to consider attrition at different stages of the PhD program separately. I don't mean to suggest that we can determine who is "to blame" for various types of attrition, but I do think departures at different stages convey different information about the departments. Some people leave during the first year of the PhD, before the first attempt at prelims. IMO, that sort of attrition is entirely personal, either because of some personal issue that came up during the year, or because the student didn't have realistic expectations about the PhD program, or because they just changed their minds about what they want to do. I don't think this type of attrition has much to do with the department at all, and therefore isn't particularly informative for the prospective student choosing one school over another. Other students leave after failing the prelims once, but before being asked to leave. Attrition at this stage is harder to parse out. Students who leave at this stage aren't taking advantage of all of the opportunities afforded to them, but they may feel squeezed or pressured out, or inadequately supported. This is especially true if they are denied funding in conjunction with failing exams. Attrition as a result of failing prelims and being asked to leave is what I think most people have in mind when they ask about attrition. (I'd lump people being asked to leave because of extremely low grades into this category, also, though that's much less common.) There certainly is information about the program in the ultimate pass rate for the prelim exams. However, people who have posted earlier are correct in suggesting that 1) a single year's pass rate may be an outlier and doesn't provide much information and 2) there can be substantial medium-run trends and other policy changes within programs are correct. The last category of attrition I would worry about is attrition among students who have reached candidacy. These students have completed coursework and exams, but for some reason don't succeed in receiving their PhDs. This can (though doesn't necessarily) suggest that the department doesn't provide adequate support throughout the research process, that advisors are in short supply, that funding for upper year students is scarce, or that there is a poor environment in the department. Of course, some students don't finish for reasons entirely of their own -- they decide that while they were good at course work, they don't like research, or "life" happens and they need to relocate, earn money, or devote time to their families. I'd encourage you to ask departments about attrition at various stages of the program. Ask about what the department views as its role in reducing each type of attrition (or if the department even thinks it needs to be reduced), and how the department supports students in each stage of the program. Consider your own strengths and weaknesses -- a student with a strong math background and excellent test taking skills, but no research experience or who knows him/herself to be undisciplined, might worry more about a program that has a low completion rate among students who achieve candidacy than about a program with a low prelim pass rate. The opposite might be true for a student who is worried about the course work, but has substantial work and research experience and is more confident about that stage of the program. In general, you want to think beyond the first year of the program. The prelims seem like a huge deal until they are over, and then they are completely irrelevant... (PS. Good luck to those of you studying now!)

TestMagic

Free GMAT, GRE, SAT Test, TOEFL practice tests