which grad school...?

Post Reply
cheers
Posts: 8
Joined: Thu Jul 12, 2012 1:34 pm

which grad school...?

Post by cheers » Thu Jul 12, 2012 2:23 pm

Hi there,

I'm an international student and I want to apply for a PhD in CMT the US. Since I'm not really into the US system but I've seen that the applications cost a lot of money (up to $150, incl. sending test scores, for a single int'l application), I'd appreciate very much any advice towards what kind of schools I have a better chance of being admitted to :)

As I already mentioned I'm interested in CMT, in particular phase transitions and exotic states of matter.

OK, let me give you some background info:

1) B.Sc. degrees (double major) in physics (German GPA: 1.49) and mathematics (German GPA 1.48) from a top German university
2) currently in a top M.Sc. program in Germany (expected German GPA ~ 1.2)
3) Upper division courses (physics):
-taken: Electrodynamics, Advanced QM, Advanced Statistical Physics, QFT, QED, Theoretical Hydrodynamics, General Ralativity, Mesoscopic physics
-currently taking: Many-body Theory, Many-body physics with Ultra-cold gases, SUSY, Nonlinear Dynamics
-to be taken: Condensed Matter Field Theory, Conformal Field Theory, Solid State Theory, Renormalization Group & Phase Transitions, Ultra-cold Quantum Gases
4) Upper division courses (mathematics):
-taken: Real Analysis, Lin Alg 1+2, Measure Theory, Topology and Geometry of Surfaces, Differential Geometry, Complex Analysis, ODE's, Functional Analysis, Mathematical QM, Advanced Mathematical QM, Mathematical Statistical Physics
5) test scores:
-general GRE: verbal: 145 (24%), quantitative: 161 (83%), writing: 4.5 (73%)
-physics GRE: 750 (64%)
-TOEFL: still to be taken
6) research experience:
- 2 bachelor theses - in maths and physics (a preprint resulted from the physics thesis in CMT - already on the arXiv, submitted to Phys. Rev. B)
- currently working on my master thesis in CMT
7) teaching experience:
- exercises classes in: Mathematical Methods for Physics, Theoretical Mechanics (twice), QM, Electrodynamics
- lab exercise class in 'physics for chemistry students'
8) physics olympics: 1st, 2nd and 3rd place in a high school physics olympiad (on a national level only)

I know my GRE scores are pretty low for theory but I don't think re-taking the physics test will improve the result significantly (wouldn't waste much time on preparation, cuz I don't have it; I'd rather concentrate on research instead of trying to figure our which is the right lens-maker convention ;) What about the general test - how much of a weight do admission committees usually put on it?

So, what do you think, where do I stand a decent chance? Any advice is much appreciated!

PS: please, don't re-direct me to the thread with the records of previous applicants; I've gone through them but I'm still pretty undecided on where to shoot for.

Cheers!
Last edited by cheers on Fri Jul 13, 2012 1:02 pm, edited 1 time in total.

kangaroo
Posts: 130
Joined: Fri Jan 13, 2012 5:31 am

Re: which grad school...?

Post by kangaroo » Fri Jul 13, 2012 11:04 am

cheers wrote:PS: please, don't re-direct me to the thread with the records of previous applicants; I've gone through them but I'm still pretty undecided on where to shoot for.
Please, that disclaimer is just a cop out, it doesn't make your query anymore legit than the rest of the "please tell me what to do" threads we have around here. The very least you could have done is to demonstrate that you actually read the threads, by coming up with specific examples of what makes you undecided in your choice.
cheers wrote:I know my GRE scores are pretty low for theory but I don't think re-taking the physics test will improve the result significantly (wouldn't waste much time on preparation, cuz I don't have it; I'd rather concentrate on research instead of trying to figure our which is the right lens-maker convention
No one is saying that it's an absolute must, but it is but one (at the very most two) question on the physics GRE. You should review all the practice tests you did and see how many of the "Big 5" questions you got wrong: classical mechanics, EM, QM, stat mech and mathematical methods/graphs. Any physicist worth his/her salt must get all those fundamental questions right, else you're either a non-Hermitian operator (i.e. some mathematician posing as a pseudo-physicist dealing with mathematical non-observables), or just an incompetent physicist. Getting every one of the Big 5 questions right should place you well into the 900 range, at the very least 850. Don't diss an entire test just because a lens equation or particle decay question comes out, it's hardly the majority of the test.

TakeruK
Posts: 941
Joined: Mon Jan 02, 2012 3:05 pm

Re: which grad school...?

Post by TakeruK » Fri Jul 13, 2012 11:40 am

cheers wrote:Hi there,
PS: please, don't re-direct me to the thread with the records of previous applicants; I've gone through them but I'm still pretty undecided on where to shoot for.
In addition to what kangaroo said, if someone here was going to sit down with you and spend a lot of time trying to figure out where you should apply, they would probably do this by reading through all the threads with records of previous applicants. It's really the only real "data" we have.

You might not find an exact match but even if you did, things change over the years so there's never a surefire way to know where you will get in. In the end, you will want to apply to many schools that fits your research interests over a large range of ranking/prestige/difficulty/etc. So, I think you should first make a list of schools that you would want to go to, without worrying about whether you can actually get in or not -- just list as many schools as you can find that matches your research interests as well as other factors (location, funding, etc.). This probably won't be a very large list -- when I did it, it was only a little over a dozen.

Then you can check with people who know you better than us random strangers on the Internet (e.g. faculty, post-docs, senior grad students at your school) and see what they think about your chances at these schools. Make sure this list doesn't contain too many schools that might be above or below your level. If your list was top or bottom heavy, add some schools to balance it out. I ended up with 8 schools to apply to in the end, which is on the low end, compared to many people on this forum.

As for scores:
General GRE -- I don't think people care very much. I don't know the new scoring system so I'm not sure what percentiles your scores are. Usually if you meet a cutoff, it's okay; higher scores won't help you. Your writing score looks fine.
Physics GRE -- I would use the profile threads to figure out what kind of places people with your scores can get in.

I don't know how German GPAs work, but if you can show/mention that you are in the top X% then that might help too, especially if you are worried about your GRE score (I don't think it's very bad but maybe low for an international student wanting to do theory, from the profiles on this forum). You seem to have a lot of courses completed and a lot of the PGRE is to ensure the new student is capable to succeeding in graduate courses and quals/comps exams. So your experience and performance in grad level courses, if really good, can help you.

cheers
Posts: 8
Joined: Thu Jul 12, 2012 1:34 pm

Re: which grad school...?

Post by cheers » Fri Jul 13, 2012 1:00 pm

I'm sorry if the PS in my previous post sounds like 'I don't want to do the searching-the-net job myself, can you do it for me?'. I really looked through most of the records but what bothered me most was that I am a foreign international and most international students' records have a really really good PGRE, in contrast to mine.
You should review all the practice tests you did and see how many of the "Big 5" questions you got wrong: classical mechanics, EM, QM, stat mech and mathematical methods/graphs. Any physicist worth his/her salt must get all those fundamental questions right
I think, I can get > 90 of the questions in the exam right, the problem being the time and the traps the authors of the problems put along the way. Now, I know this means I just have to practise and practise, and practise ('practice makes perfect'), but last time I had made it up to the stage that I as able to do the practice tests in less than an hour simply because I knew the problems already. Now, you're going to say: 'Well, buy some practice books and go through these new problems!' - well fine, but you know this takes time, time I could spend doing smth else; and the question I've been asking myself is 'How exactly does this make me a better phycist?!'
Then you can check with people who know you better than us random strangers on the Internet (e.g. faculty, post-docs, senior grad students at your school) and see what they think about your chances at these schools.
I talked to several professors and postdocs at my university and their opinion/advice seems to be divided into re-taking the PGRE to try to boost it up to the 80%-ile, or trying to get another pre-print before december. Of course, the safer option seems to be the PGRE, except I thought I had done my best in preparing for it last time. The argument of those who urged me re-take the test was that there are unofficial pre-screening cut-offs and if I happen not to survive them my application might never be taken into consideration. Any idea how to find out these cut-offs for the individual schools? (it's propably not the average, since every school says they look at the application as a whole; and averages differ for different subfields and for intl's compared to natives, etc.)
I don't know how German GPAs work, but if you can show/mention that you are in the top X% then that might help too
Here's some info on the German GPA system: on average (i.e. depending on the prof.) a 1.0 (best grade) is given for doing >=90% of the exam. Sometimes it happens that no one gets it, sometimes the 90% is reduced, so that someone actually gets a 1.0 - it varies. My grades were gradually improving and in the graduate level courses they became optimal, but I don't know which percentage I fit in.

I'll try to edit my previous post to update it with the percentages on the general GRE :)

Cheers!

bfollinprm
Posts: 1203
Joined: Sat Nov 07, 2009 11:44 am

Re: which grad school...?

Post by bfollinprm » Fri Jul 13, 2012 6:09 pm

If you are shooting for a top school your only chance any way will be someone at the school really wanting you to work with them*; if you have a professor asking for a student to be admitted that student is rarely denied admission. If you really want to know your chances at a particular school, there's nothing stopping you from contacting professors at the institution, enquiring about their research, mentioning that you're interested in working with them, and asking if they see you as a successful applicant at the school. If you're looking at mid-range (~20-50 on the NRC or US News rankings) schools, your PGRE will be a negative, but won't disqualify you. You do have 1-2 publications as a theorist, so that's a plus. If you've had to take a comprehensive exam during your masters program, or even to finish your bachelors, you can ask a letter writer to comment on your performance on that exam. Generally the PGRE is to make sure someone actually understands physics and isn't a poser: if you can prove in another way you aren't, then people should care about your PGRE score less. Yes, some schools (the bigger ones) do have hard PGRE cutoffs, and maybe at a few you'll be in danger of missing them (Stanford, probably, and maybe U Chicago). But there's other places to do quality CMT (there are a ton of CM Theorists out there).

*A perfect score on the PGRE for an international student just makes sure they don't throw your application into a pile to consider after they've considered all reasonable domestic applicant at places like MIT, CalTech, and Harvard (which, considering the number of applications they get, might happen with a 990). The larger the school the less this is true, and in general private schools are a little bit more reasonable about this than the public schools (they accept more international students). The better way to get your application off of this pile is to get someone to personally move it to the top of the other one.
kangaroo wrote: No one is saying that it's an absolute must, but it is but one (at the very most two) question[s] on the physics GRE. You should review all the practice tests you did and see how many of the "Big 5" questions you got wrong: classical mechanics, EM, QM, stat mech and mathematical methods/graphs. Any physicist worth his/her salt must get all those fundamental questions right, else you're either a non-Hermitian operator (i.e. some mathematician posing as a pseudo-physicist dealing with mathematical non-observables), or just an incompetent physicist. Getting every one of the Big 5 questions right should place you well into the 900 range, at the very least 850. Don't diss an entire test just because a lens equation or particle decay question comes out, it's hardly the majority of the test.
Honestly, man, anyone who thinks a physicist earns their salt by getting questions right on the PGRE is in for a rude awakening when your comfy answers-in-a-solutions-manual coursework gives way to research. I know tons of professors at the top of their fields who in no way aced that exam.* It's absolutely right that studying for the PGRE does not make you a better physicist, though as long as there are people out there who do devote a summer to learning how to do freshman physics in a minute or less there will always be pressure to waste time preparing for the exam. It might be useful to retake the exam; but if it's going to take more than 100 hours to increase your score significantly I wouldn't bother. Work on getting another publication, aggressively contact professors, and find palatable contingency plans in case you don't get into your dream school.

*To be fair I also know some who think it's very important. I find they and I would be poor research partners.



Post Reply