5 ms·

I truly wish you the best of luck, and for an excellent outcome. One thing - you just can't say "why I'm getting a PhD", even from an elite school, the way you

by geebee 13y ago

I truly wish you the best of luck, and for an excellent outcome. One thing - you just can't say "why I'm getting a PhD", even from an elite school, the way you'd say "why I'm getting a law, medical, or business degree". Even if you've been admitted to an elite program, you can really only say "why I'm rolling the dice on the possibility of getting a PhD". At elite programs, there is essentially no involuntary attrition in JD, MD, or MBA degrees. Yale law school has no 1L attrition. Columbia's JD 1L attrition is 0.3%. Attrition rates at elite med school is typically well below 1%. http://taxprof.typepad.com/taxprof_blog/2008/04/law-school-ra-1.html http://taxprof.typepad.com/taxprof_blog/2008/04/law-school-r... (I'll stop with the googling of links here...) Attrition rates for engineering PhD students appear to be about 33%. This (to me, atrocious number) is way better than science or humanities. http://www.insidehighered.com/news/2007/12/07/doctoral http://www.insidehighered.com/news/2007/12/07/doctoral Unlike JD, MD, MBA programs, I don't believe that the numbers are better at elite schools. http://www-inst.eecs.berkeley.edu/~wicse/index.php/papers/lindareport2.pdf http://www-inst.eecs.berkeley.edu/~wicse/index.php/papers/li... I dunno, maybe Berkeley is a risky program, it is a notoriously harsh school. In spite of all I've said, I'm always hesitant to give anyone advice. There are still remarkable careers to be had in academics. Good luck.

aaronbrethorst 13y ago
> Attrition rates for engineering PhD students appear to be about 33%. This (to me, atrocious number) is way better than science or humanities. Perhaps this is because an engineering Ph.D. student is far better positioned to get a high-paying job than their 1L or med school counterparts. The engineering candidate doesn't need to be there to get a job.
- geebee 13y ago
  There are all kinds of reasons. I'd certainly agree that what you pointed out is one of them. That said... then why is the engineering PhD attrition rate lower than in science, which is in turn lower than humanities?
- freyr 13y ago
  Yes, and the med school student will earn substantially more than the PhD student (or the PhD degree holder) after completing med school. So there is significant financial incentive to finish the degree (which has likely put the med school student deep into loan debt already).
- zenbowman 13y ago
  Exactly. 33% isn't bad to be honest. When my advisor quit, I left my PhD without thinking twice, because I got a job at a much higher pay than my PhD stipend, and leaving my PhD didn't present the kind of financial burden that it would have were I in a humanities field. Now I'd like to complete it again, but I'd want to do it part-time, and I have a much better idea of what I'd like to do. As engineers, we are pretty lucky to be in that position, even when things don't quite work out as planned, we still end up pretty well off. So there's not quite the same determination to make a particular course of action work.
KrisJordan 13y ago
I suppose I am part of your Ph.D. 'attrition' statistic. It's less atrocious than you'd think. I enrolled in Brown's Ph.D. CS computer program immediately after undergrad. In retrospect, I was too green to do proper due diligence to find the right fit with advisor and research area. The advisor relationship is extremely important. Within the first year I realized two things: 1) there was no way I'd be happy going deep with this particular research area and advisor for another 5-6 years, 2) I would complete my masters in the first year, fully paid for (with a research assistant "salary" on top) by being on the PhD track. I returned full-time to the company I co-founded while in undergrad after that first year. No two PhD paths are the same, unlike with JD/MBA. Some folks come in and earn a master's along the way. Some come in with a master's. Some don't pass the (widely different depending on school) bars to advance to the Ph.D. candidate stage (I imagine there's a good bit of attrition here). Some don't find an advisor who can fund them. The requirement of contributing something novel to the body of research, no matter how long it takes, is a higher bar than completing coursework and projects. 6,7,8 years is a lot longer than 2-3.
- geebee 13y ago
  There's room for disagreement about whether this is atrocious. But at its core, do you think that the 33% attrition rate is a problem?
keypusher 13y ago
I suspect Berkeley is on the high end of the spectrum as far as attrition goes. I went there for undergrad, and the pressure and difficulty was extreme in areas like EECS. For perspective, after graduating I moved to the East Coast and worked in a research lab at MIT and took courses at Harvard, both of which felt quite laid back after Berkeley. It was the only school experience I had where I saw smart people work crazy hard and still receive C's and D's in some classes. The PhD candidate I lived with decided to take a year long internship with Intel (as a vacation) for mental health reasons.

by geebee 13y ago

aaronbrethorst 13y ago

> Attrition rates for engineering PhD students appear to be about 33%. This (to me, atrocious number) is way better than science or humanities. Perhaps this is because an engineering Ph.D. student is far better positioned to get a high-paying job than their 1L or med school counterparts. The engineering candidate doesn't need to be there to get a job.

geebee 13y ago
There are all kinds of reasons. I'd certainly agree that what you pointed out is one of them. That said... then why is the engineering PhD attrition rate lower than in science, which is in turn lower than humanities?
freyr 13y ago
Yes, and the med school student will earn substantially more than the PhD student (or the PhD degree holder) after completing med school. So there is significant financial incentive to finish the degree (which has likely put the med school student deep into loan debt already).
zenbowman 13y ago
Exactly. 33% isn't bad to be honest. When my advisor quit, I left my PhD without thinking twice, because I got a job at a much higher pay than my PhD stipend, and leaving my PhD didn't present the kind of financial burden that it would have were I in a humanities field. Now I'd like to complete it again, but I'd want to do it part-time, and I have a much better idea of what I'd like to do. As engineers, we are pretty lucky to be in that position, even when things don't quite work out as planned, we still end up pretty well off. So there's not quite the same determination to make a particular course of action work.

KrisJordan 13y ago

I suppose I am part of your Ph.D. 'attrition' statistic. It's less atrocious than you'd think. I enrolled in Brown's Ph.D. CS computer program immediately after undergrad. In retrospect, I was too green to do proper due diligence to find the right fit with advisor and research area. The advisor relationship is extremely important. Within the first year I realized two things: 1) there was no way I'd be happy going deep with this particular research area and advisor for another 5-6 years, 2) I would complete my masters in the first year, fully paid for (with a research assistant "salary" on top) by being on the PhD track. I returned full-time to the company I co-founded while in undergrad after that first year. No two PhD paths are the same, unlike with JD/MBA. Some folks come in and earn a master's along the way. Some come in with a master's. Some don't pass the (widely different depending on school) bars to advance to the Ph.D. candidate stage (I imagine there's a good bit of attrition here). Some don't find an advisor who can fund them. The requirement of contributing something novel to the body of research, no matter how long it takes, is a higher bar than completing coursework and projects. 6,7,8 years is a lot longer than 2-3.

geebee 13y ago
There's room for disagreement about whether this is atrocious. But at its core, do you think that the 33% attrition rate is a problem?

keypusher 13y ago

I suspect Berkeley is on the high end of the spectrum as far as attrition goes. I went there for undergrad, and the pressure and difficulty was extreme in areas like EECS. For perspective, after graduating I moved to the East Coast and worked in a research lab at MIT and took courses at Harvard, both of which felt quite laid back after Berkeley. It was the only school experience I had where I saw smart people work crazy hard and still receive C's and D's in some classes. The PhD candidate I lived with decided to take a year long internship with Intel (as a vacation) for mental health reasons.