Page 1 of 1

Why does your Undergraduate School Matter?

Posted: Tue Jan 15, 2013 6:59 pm
by goingnuclear
It seems like an obvious question at first, but think about it. A highly reputed physics school, say a top 10, implies you had access to (1) a top notch education and (2) plenty of hands on research opportunities. But your physics GRE conveys how successful your education was, and you're required to explicitly list your research in your application anyways. So whatever information the admissions committee wishes to gain from knowing your undergrad school is redundant, as it's all contained throughout the rest of the application.

If some student went to a no-name liberal arts school, got a 990 on the PGRE, and spent a summer studying at SLAC, what does he lack when compared to a student who went to a top-tier physics school (such as MIT or Bob Jones University )? He clearly has a mastery of undergraduate physics and knows how to work in a modern physics research environment.

Re: Why does your Undergraduate School Matter?

Posted: Tue Jan 15, 2013 7:02 pm
by Tom Joad
I think the physics GRE becomes almost unnecessary at a top 10 (maybe just top 5) undergraduate university. For instance, if you've got all A's at MIT I'm sure they would be convinced you know your stuff. As someone who's already at a top tier institution, they would be a safer bet then someone from a no-name school.

Re: Why does your Undergraduate School Matter?

Posted: Tue Jan 15, 2013 11:34 pm
by bfollinprm
goingnuclear wrote:It seems like an obvious question at first, but think about it. A highly reputed physics school, say a top 10, implies you had access to (1) a top notch education and (2) plenty of hands on research opportunities. But your physics GRE conveys how successful your education was, and you're required to explicitly list your research in your application anyways. So whatever information the admissions committee wishes to gain from knowing your undergrad school is redundant, as it's all contained throughout the rest of the application.

If some student went to a no-name liberal arts school, got a 990 on the PGRE, and spent a summer studying at SLAC, what does he lack when compared to a student who went to a top-tier physics school (such as MIT or Bob Jones University )? He clearly has a mastery of undergraduate physics and knows how to work in a modern physics research environment.
In my opinion, there is no difference. The only time knowing the undergrad school matters (in my opinion) is when a school is well known to be an outlier in grade distributions (either higher or lower than average GPAs).

Re: Why does your Undergraduate School Matter?

Posted: Tue Jan 15, 2013 11:53 pm
by Meteorshower
Presumably it's more important for international students then?

Re: Why does your Undergraduate School Matter?

Posted: Wed Jan 16, 2013 12:19 am
by twistor
The answer is that your physics GRE score does not necessarily reflect your aptitude to do physics. It is a highly contrived exam that only serves to narrow the pool of potential applicants. Doing well probably means you're good at solving physics problems. It could also mean you're skilled a memorizing answers (for international applicants) or making educated guesses. A 3 hour long test cannot test your aptitude at deriving results, performing long calculations, or interpreting results. Your grades are a much better indicator of these things and the reputation of your undergraduate school testifies to quality of the courses in which your earned those grades.

Conversely, doing poorly on the physics GRE doesn't necessarily mean you're not good at physics, possibly just not good at tests or had a bad day, whatever. Your grades and undergraduate school are again important for demonstrating your aptitude for physics when your score is lacking.

Re: Why does your Undergraduate School Matter?

Posted: Wed Jan 16, 2013 8:41 am
by admissionprof
twistor wrote:The answer is that your physics GRE score does not necessarily reflect your aptitude to do physics. It is a highly contrived exam that only serves to narrow the pool of potential applicants. Doing well probably means you're good at solving physics problems. It could also mean you're skilled a memorizing answers (for international applicants) or making educated guesses. A 3 hour long test cannot test your aptitude at deriving results, performing long calculations, or interpreting results. Your grades are a much better indicator of these things and the reputation of your undergraduate school testifies to quality of the courses in which your earned those grades.

Conversely, doing poorly on the physics GRE doesn't necessarily mean you're not good at physics, possibly just not good at tests or had a bad day, whatever. Your grades and undergraduate school are again important for demonstrating your aptitude for physics when your score is lacking.

I agree with Twistor. The physics GRE score is just one datum, and while it does correlate a bit with success in graduate school, I believe the undergraduate school's reputation and grades have a higher correlation. Often, students from schools ranked very poorly are unprepared for the difficulty and competitiveness of graduate school courses. That said, I personally put greater weight on the grades in courses of upper level physics and math (and having taken two courses each of upper level EM and QM, which is usually the case at highly ranked schools and usually not the case at lower ranked, is also important), and letters from supervisors.

Re: Why does your Undergraduate School Matter?

Posted: Wed Jan 16, 2013 10:40 am
by blighter
twistor wrote:The answer is that your physics GRE score does not necessarily reflect your aptitude to do physics. It is a highly contrived exam that only serves to narrow the pool of potential applicants. Doing well probably means you're good at solving physics problems. It could also mean you're skilled a memorizing answers (for international applicants) or making educated guesses. A 3 hour long test cannot test your aptitude at deriving results, performing long calculations, or interpreting results. Your grades are a much better indicator of these things and the reputation of your undergraduate school testifies to quality of the courses in which your earned those grades.

Conversely, doing poorly on the physics GRE doesn't necessarily mean you're not good at physics, possibly just not good at tests or had a bad day, whatever. Your grades and undergraduate school are again important for demonstrating your aptitude for physics when your score is lacking.
I don't understand the implication behind adding the qualifier 'for international applicants'.

Re: Why does your Undergraduate School Matter?

Posted: Wed Jan 16, 2013 10:46 am
by goingnuclear
twistor wrote:The answer is that your physics GRE score does not necessarily reflect your aptitude to do physics. It is a highly contrived exam that only serves to narrow the pool of potential applicants. Doing well probably means you're good at solving physics problems. It could also mean you're skilled a memorizing answers (for international applicants) or making educated guesses. A 3 hour long test cannot test your aptitude at deriving results, performing long calculations, or interpreting results. Your grades are a much better indicator of these things and the reputation of your undergraduate school testifies to quality of the courses in which your earned those grades.

Conversely, doing poorly on the physics GRE doesn't necessarily mean you're not good at physics, possibly just not good at tests or had a bad day, whatever. Your grades and undergraduate school are again important for demonstrating your aptitude for physics when your score is lacking.
I would argue that your GPA is less of an indicator of your aptitude than the PGRE. Your GPA indicates how well you can play the school system rather than how much you learned. If I was good at binging the night before tests and regurgitating it all the next day on the test, only to forget it the following day, would that mean I was a competent student? No, but based on my GPA you would think so. In addition, a lot of schools only ask for your cumulative GPA, which is even less relevant to physics. For example, I have a 3.42 total and a 3.80 in just physics courses, yet I have no doubt that I'll be turned down from many schools for my low overall GPA.

The PGRE tests 4 years of knowledge, and while you can study for it there's really no way to cram the night before. It does test only a certain type of problem solving skills, but I think these are the type of skills that undergraduate aims to equip us with (whether or not it actually reflects what's required of graduate school). I'm not saying the PGRE is a perfect indicator of your aptitude in physics, but when it comes to evaluating your proficiency in physics I do think it's more accurate than your GPA.

Also, what do you mean by being "skilled at memorizing answers" if you're international? Are there ways to cheat on the PGRE (I already took it)?

Re: Why does your Undergraduate School Matter?

Posted: Wed Jan 16, 2013 12:02 pm
by blighter
goingnuclear wrote:
twistor wrote:The answer is that your physics GRE score does not necessarily reflect your aptitude to do physics. It is a highly contrived exam that only serves to narrow the pool of potential applicants. Doing well probably means you're good at solving physics problems. It could also mean you're skilled a memorizing answers (for international applicants) or making educated guesses. A 3 hour long test cannot test your aptitude at deriving results, performing long calculations, or interpreting results. Your grades are a much better indicator of these things and the reputation of your undergraduate school testifies to quality of the courses in which your earned those grades.

Conversely, doing poorly on the physics GRE doesn't necessarily mean you're not good at physics, possibly just not good at tests or had a bad day, whatever. Your grades and undergraduate school are again important for demonstrating your aptitude for physics when your score is lacking.
I would argue that your GPA is less of an indicator of your aptitude than the PGRE. Your GPA indicates how well you can play the school system rather than how much you learned. If I was good at binging the night before tests and regurgitating it all the next day on the test, only to forget it the following day, would that mean I was a competent student? No, but based on my GPA you would think so. In addition, a lot of schools only ask for your cumulative GPA, which is even less relevant to physics. For example, I have a 3.42 total and a 3.80 in just physics courses, yet I have no doubt that I'll be turned down from many schools for my low overall GPA.

The PGRE tests 4 years of knowledge, and while you can study for it there's really no way to cram the night before. It does test only a certain type of problem solving skills, but I think these are the type of skills that undergraduate aims to equip us with (whether or not it actually reflects what's required of graduate school). I'm not saying the PGRE is a perfect indicator of your aptitude in physics, but when it comes to evaluating your proficiency in physics I do think it's more accurate than your GPA.

Also, what do you mean by being "skilled at memorizing answers" if you're international? Are there ways to cheat on the PGRE (I already took it)?
Yes, exactly. I don't know about other schools, but at my school the exams precisely require you to memorise and regurgitate the solutions. There's no way you can even pass the course without doing that. You aren't given enough time to actually think and plan a solution. Maybe that's why I didn't find time too constraining in the PGRE like everyone else seems to.

Re: Why does your Undergraduate School Matter?

Posted: Wed Jan 16, 2013 1:10 pm
by twistor
goingnuclear wrote:
twistor wrote:The answer is that your physics GRE score does not necessarily reflect your aptitude to do physics. It is a highly contrived exam that only serves to narrow the pool of potential applicants. Doing well probably means you're good at solving physics problems. It could also mean you're skilled a memorizing answers (for international applicants) or making educated guesses. A 3 hour long test cannot test your aptitude at deriving results, performing long calculations, or interpreting results. Your grades are a much better indicator of these things and the reputation of your undergraduate school testifies to quality of the courses in which your earned those grades.

Conversely, doing poorly on the physics GRE doesn't necessarily mean you're not good at physics, possibly just not good at tests or had a bad day, whatever. Your grades and undergraduate school are again important for demonstrating your aptitude for physics when your score is lacking.
I would argue that your GPA is less of an indicator of your aptitude than the PGRE. Your GPA indicates how well you can play the school system rather than how much you learned. If I was good at binging the night before tests and regurgitating it all the next day on the test, only to forget it the following day, would that mean I was a competent student? No, but based on my GPA you would think so. In addition, a lot of schools only ask for your cumulative GPA, which is even less relevant to physics. For example, I have a 3.42 total and a 3.80 in just physics courses, yet I have no doubt that I'll be turned down from many schools for my low overall GPA.

The PGRE tests 4 years of knowledge, and while you can study for it there's really no way to cram the night before. It does test only a certain type of problem solving skills, but I think these are the type of skills that undergraduate aims to equip us with (whether or not it actually reflects what's required of graduate school). I'm not saying the PGRE is a perfect indicator of your aptitude in physics, but when it comes to evaluating your proficiency in physics I do think it's more accurate than your GPA.

Also, what do you mean by being "skilled at memorizing answers" if you're international? Are there ways to cheat on the PGRE (I already took it)?
The PGRE does not test 4 years of knowledge. I've already stated why this is true. No 3 hour long exam can test 4 years worth of knowledge. As far as evaluating your aptitude to do physics, that's not even close to being true. Graduate school is not simply a continuation of taking classes. You need to do real research, learn a lot of new skills that aren't taught in any class, and hopefully make a real contribution to your field. You can be very good at that despite your inability to recall offhand some arcane formula.
Also, what do you mean by being "skilled at memorizing answers" if you're international? Are there ways to cheat on the PGRE (I already took it)?
International applicants often have access to books which list the answers to the tests. They're circulated in foreign languages. One Chinese student told me personally that he knew people in China who would get the books and not only memorize the answers, but also what page number in the book that answer came from.

Re: Why does your Undergraduate School Matter?

Posted: Wed Jan 16, 2013 2:51 pm
by goingnuclear
I'm not saying the PGRE can cover EVERYTHING in 4 years, as I agree with you that there's no way a 3 hour test can cover that much. But it tests you on all the essential classes which you take over 4 years, from baby classical mechanics up through quantum and electrodynamics.

Imagine if, instead of taking one physics GRE like we do now, there were different sub-subject PGREs offered after every semester. So, for example, you would take a semester in classical mechanics, and then immediately after take a standardized 100 question test on just classical mechanics. You would take a different 100 question test for electrodynamics, stat mech, quantum, etc. This method of testing, although annoying and time consuming, would be able to adequately cover everything learned in physics courses (in my opinion). The flaw in this, though, is that it is no different from a final exam - students would have the ability to cram for a couple nights before, regurgitate for the test, and then forget about it forever. By offering a single test near the end of college, after all is said and done, it separates the people who truly retained and understood the physics from the ones who just knew how to do the right things to get good grades.

From looking at the numbers from the past few years, the PGRE/GPA data has shown this method of "learning" in a large number of applicants. Students with 3.8+ total GPAs, and then sub-50% PGRE scores. I know some people think the PGRE isn't a true measure of knowledge, blah blah blah, but if you aren't hitting the 50% mark on the PGRE then a 3.8+ GPA is clearly not an accurate representation of your physics aptitude. It just means you knew what to do to get good grades.

Again, not saying the PGRE is a perfect representation of who you are as a student, and like a lot of data it should be taken with a grain of salt. But I do think it's a better depiction of your physics aptitude than your GPA, simply because it's harder to "regurgitate and forget" on the PGRE.

That's both interesting and sad to hear about cheating internationally. That must make it really frustrating for an honest international applicant, as it makes it unfairly competitive for them.

Re: Why does your Undergraduate School Matter?

Posted: Wed Jan 16, 2013 3:08 pm
by bfollinprm
I'd believe a statement like "The PGRE traces inaptitude in physics," but I wouldn't think that the PGRE (at least the one I took) really tested my ability to do physics, or even really got a solid handle on my physics knowledge. Top scores are probably correlated with hard workers, though, so I'd think there's still discerning power in the higher end of the spectrum useful to admissions committees. Definitely can't replace transcripts and recommendation letters though, since these probably trace physics aptitude much better.

I also disagree that someone with aptitude in physics should score > 50%ile in all cases. I know it was true for me (I got >50%ile the first time I took a practice test), but I'm also a good multiple-choice test taker. It's definitely a test that can psych you out.

Re: Why does your Undergraduate School Matter?

Posted: Wed Jan 16, 2013 3:42 pm
by twistor
I know some people think the PGRE isn't a true measure of knowledge, blah blah blah, but if you aren't hitting the 50% mark on the PGRE then a 3.8+ GPA is clearly not an accurate representation of your physics aptitude.
I know some people think GRADES AREN'T a true measure of knowledge, blah blah blah, but if you aren't hitting the 50% mark IN YOUR COURSES then a 990 ON THE PGRE is clearly not an accurate representation of your physics aptitude.

The bottom line is that you have several people who have already been through the process, been through graduate school (or nearly so), and at least one professor on an admissions committee telling you that a person's score on the PGRE is not a good measure of an applicant by itself. Perhaps you are embarrassed by the school you went to or you want to inflate your ego about how well you did. Or maybe you don't think it's fair that someone from a better school has a marginally better shot at getting in and want to level the playing field. In any case, the school matters for all the reasons listed above. You can choose to accept it or ignore it but in any case you cannot change the admissions processes. In fact, once you are accepted somewhere you will likely forget all about it and move on with your life.

The field you go into will likely be so specialized that in a couple years you will have forgotten just how much charge a capacitor can hold or or whether some odd commutator had a factor of i or -i.

Re: Why does your Undergraduate School Matter?

Posted: Wed Jan 16, 2013 5:01 pm
by blighter
twistor wrote:
I know some people think the PGRE isn't a true measure of knowledge, blah blah blah, but if you aren't hitting the 50% mark on the PGRE then a 3.8+ GPA is clearly not an accurate representation of your physics aptitude.
I know some people think GRADES AREN'T a true measure of knowledge, blah blah blah, but if you aren't hitting the 50% mark IN YOUR COURSES then a 990 ON THE PGRE is clearly not an accurate representation of your physics aptitude.
Again my argument is your GPA is correlated to the university you attended. The grades are hugely inflated in some and quite the opposite in some others. At my school, no matter how well everyone does on a course, the professors somehow think that they have to give A's only to the top 10%, B's to the top 10-30%, C to 30-60%, D to 60-85 and F to the bottom 15%. So around 15-20% of the class invariably flunks the course. Moreover the number of courses we take in a term is a lot more than the US. PGRE hardly required any preparation after all this. I just had to look up some formulae in optics — which I ended up forgetting when I needed them anyway — a couple of days before the test. I'm not even at the top of my class.

Re: Why does your Undergraduate School Matter?

Posted: Wed Jan 16, 2013 5:08 pm
by blighter
twistor wrote:whether some odd commutator had a factor of i or -i.
I hardly think PGRE asks such questions. Even if it does, those can easily be skipped to work on the 'better' problems. If better grades indicate better mathematical and derivation skills, it shouldn't take you more than a few seconds to actually derive that. I derive the formulas invariably. Heck, I can't even remember Maxwell's/Schrodinger equations. If something requires me to remember them, I'm stuck.

I do agree that PGRE doesn't truly test how good you are at physics. My point is GPA is no better at that.

Re: Why does your Undergraduate School Matter?

Posted: Wed Jan 16, 2013 5:11 pm
by TakeruK
goingnuclear wrote: Imagine if, instead of taking one physics GRE like we do now, there were different sub-subject PGREs offered after every semester. So, for example, you would take a semester in classical mechanics, and then immediately after take a standardized 100 question test on just classical mechanics. You would take a different 100 question test for electrodynamics, stat mech, quantum, etc. This method of testing, although annoying and time consuming, would be able to adequately cover everything learned in physics courses (in my opinion). The flaw in this, though, is that it is no different from a final exam - students would have the ability to cram for a couple nights before, regurgitate for the test, and then forget about it forever. By offering a single test near the end of college, after all is said and done, it separates the people who truly retained and understood the physics from the ones who just knew how to do the right things to get good grades.
This would be a complete nightmare for me!! Despite being an undergrad Astronomy student, I don't think my "physics aptitude" is "that bad", but if my courses had exams like this, I would probably do very poorly. I do not think standardized testing is a measure of anything other than ability to take standardized tests. Also, this method of testing is VERY DIFFERENT, in my opinion, than a final exam. In my undergrad, at least at the senior courses, final exams are usually 2.5 to 3 hours long and contain about 3 or 4 large, problem-based questions (sometimes they are broken up into steps, like prove some relation, then use it to solve something etc.). We usually get a 1 page cheat-sheet that we're allowed to bring into the test. The question is usually something we have never seen before in class, but it might be a reworked homework problem or an extended example of something we did see. So, the emphasis is NOT on remembering what we learned during the term, but actually learning the skills and techniques in order to solve problems related to what we've already seen. In addition, at the senior level in my schools, the grade weighting is something like 30-40% homework, ~25% midterm, and the rest is a final exam or a final project. Test taking is not as big of as big of a factor in your grades!

I think exams like this are much more useful for future physics education and research. In grad school, during quals, or other oral defenses/exams, we will be asked to solve problems on the board, for example. In a multiple choice standardised test, you either get it right or you get it wrong. If you made a small error near the end, but knew the general idea on how to tackle the problem, you still get the same mark as someone who didn't know what to do at all and just guessed. This is not a way to measure knowledge, in my opinion.

In our research life, we will be able to access the literature to find equations etc. The goal of a course isn't to learn/memorize all the steps to applying Method A to specific Problem Y. Instead, we should learn that Method X is a good tool to use when faced with situations like Problem Y as well as when it's a good idea to use Method A or should we be using Method B or C instead. In addition, in "real life" science, we will rarely encounter specific Problem Y. Instead, we might have a problem that we can say, "oh this looks a little like Problem Y", let's see if we can use Method A on it, etc.

By having a test like the PGRE at the end of the degree, the students with the highest score tends to be the intersection of the set of students who were able to retain all of the information they learned in their coursework AND the set of students who are good at taking standardised tests.
From looking at the numbers from the past few years, the PGRE/GPA data has shown this method of "learning" in a large number of applicants. Students with 3.8+ total GPAs, and then sub-50% PGRE scores. I know some people think the PGRE isn't a true measure of knowledge, blah blah blah, but if you aren't hitting the 50% mark on the PGRE then a 3.8+ GPA is clearly not an accurate representation of your physics aptitude. It just means you knew what to do to get good grades.

Again, not saying the PGRE is a perfect representation of who you are as a student, and like a lot of data it should be taken with a grain of salt. But I do think it's a better depiction of your physics aptitude than your GPA, simply because it's harder to "regurgitate and forget" on the PGRE.
I agree that your GPA is affected by whether or not you knew what to do to get good grades. But as I said above, the PGRE is also a measure of knowing what to do to score well on standardised tests. The real question is what set of skills do you want your students to have? In my opinion, knowing what to do to get good grades means knowing what material is important in a course and what are the key ideas that the prof expect you to know (so that it will appear on the test) as well as the other skills I already mentioned above about my experience with courses. Knowing how to do well on the PGRE means memorizing the database of questions and practising standardised testing. I would argue that "knowing how to do well in courses" is a much more useful skill for a scientist than "knowing how to do well in the PGRE" because if courses are taught well, they can teach students the skills to do research but the PGRE cannot teach anything more than memorization and testing ability. (Sure, there are some computational tricks that are useful like order of magnitude estimation and getting a feeling for what is "right" but these strategies are less useful in modern PGREs and also I think PGRE scores are weighted too heavily to just evaluate this skill.)

Re: Why does your Undergraduate School Matter?

Posted: Wed Jan 16, 2013 5:34 pm
by blighter
TakeruK wrote:This would be a complete nightmare for me!! Despite being an undergrad Astronomy student, I don't think my "physics aptitude" is "that bad", but if my courses had exams like this, I would probably do very poorly. I do not think standardized testing is a measure of anything other than ability to take standardized tests. Also, this method of testing is VERY DIFFERENT, in my opinion, than a final exam. In my undergrad, at least at the senior courses, final exams are usually 2.5 to 3 hours long and contain about 3 or 4 large, problem-based questions (sometimes they are broken up into steps, like prove some relation, then use it to solve something etc.). We usually get a 1 page cheat-sheet that we're allowed to bring into the test. The question is usually something we have never seen before in class, but it might be a reworked homework problem or an extended example of something we did see. So, the emphasis is NOT on remembering what we learned during the term, but actually learning the skills and techniques in order to solve problems related to what we've already seen. In addition, at the senior level in my schools, the grade weighting is something like 30-40% homework, ~25% midterm, and the rest is a final exam or a final project. Test taking is not as big of as big of a factor in your grades!
Wow! If that's how people are tested in US/Canada, then GPA is indeed a good indicator. My opinions about the grades are biased to the Indian context, more specifically to my college — because frankly the system and the misguided sense of rigour is kinda specific to my college.

I do concede that there have been people who conformed to the system, had an almost perfect GPA and went on to do great physics as well. But they're just the anomalies. Expecting everyone to be like them is just ridiculous. The average types like me are totally f*cked.

Re: Why does your Undergraduate School Matter?

Posted: Wed Jan 16, 2013 6:27 pm
by goingnuclear
twistor wrote:The bottom line is that you have several people who have already been through the process, been through graduate school (or nearly so), and at least one professor on an admissions committee telling you that a person's score on the PGRE is not a good measure of an applicant by itself. Perhaps you are embarrassed by the school you went to or you want to inflate your ego about how well you did. Or maybe you don't think it's fair that someone from a better school has a marginally better shot at getting in and want to level the playing field. In any case, the school matters for all the reasons listed above. You can choose to accept it or ignore it but in any case you cannot change the admissions processes. In fact, once you are accepted somewhere you will likely forget all about it and move on with your life.
...Whoa. o_O Lets take a step back. Clearly my opinion of the schooling system personally offended you, or something. I am not trying to "inflate my ego" or "embarrassed by my school," nor am I saying anything is unfair, so quit being a dick and insulting me simply because I was trying to spur a discussion. You use phrases like "bottom line" and "accept it or ignore it," acting like there is some absolute truth to this discussion and I'm clearly wrong for not agreeing with you. Believe it or not, the issue of GPA vs Aptitude is an extremely subjective matter, and just because you disagree with me does NOT mean I'm wrong and therefore should be subjected to ridicule. So get off your damn high horse and try to talk without being condescending.

But yeah, originally when I was talking about GPA, I think I was referring to cumulative rather than major GPA. After all, most applications (from what I can tell) ask for your cumulative but don't always explicitly ask for your physics gpa. As people have mentioned, I do think indepth tests which require a decent amount of time per problem are a better indicator of your physics prowess than a multiple choice test, since the PGRE doesn't really allow much depth in each problem. However, I'm a little torn with the issue of cheat sheats. I remember a kid in my quantum class, who wrote out every possible solution to the Schrodinger equation for pretty much every 1-D potential you could think of, including wave functions, energy levels, expectation values, etc. I would be more in favor of the professor writing the basic equations on the board (such as Schrodinger's, Heisenberg's, etc), but that's about it. And I definitely don't agree with homework being a majority of your grade, as students can easily use the internet to find all the answers they need.

Blighter sums up my thoughts on the matter - PGRE isn't a perfect indicator of physics aptitude, but your GPA isn't any better. And in my opinion, it's a little worse.

I just realized how far the discussion deviated from the original topic of the thread, hah.

Re: Why does your Undergraduate School Matter?

Posted: Wed Jan 16, 2013 8:24 pm
by TakeruK
goingnuclear wrote: But yeah, originally when I was talking about GPA, I think I was referring to cumulative rather than major GPA. After all, most applications (from what I can tell) ask for your cumulative but don't always explicitly ask for your physics gpa.
I think it's incorrect to assume that since they do not ask for you to write your Physics GPA in a box on the form that they do not consider it. The user admissionprof above said that he/she gives more weight to certain senior courses. It's not that hard to get a major GPA from a transcript. I know that some schools do not even ask for you to enter a cumulative GPA into a box because they rather compute it themselves.
However, I'm a little torn with the issue of cheat sheats. I remember a kid in my quantum class, who wrote out every possible solution to the Schrodinger equation for pretty much every 1-D potential you could think of, including wave functions, energy levels, expectation values, etc. I would be more in favor of the professor writing the basic equations on the board (such as Schrodinger's, Heisenberg's, etc), but that's about it.
I think that's a great use of cheat sheets and almost expected by the profs. For example, Chapter 2 of Griffiths is a wonderful summary of all the 1D potentials and their solutions. I always put that information in all of my subsequent QM exam cheat sheets. I don't think I have ever had a QM exam that is something like "Solve the Schroedinger Equation for a particle in a square well". Usually, the prof invents some clever combination of two potentials or whatever (it's been awhile since QM for me) so that you have to figure out which of your cheat sheet contents to combine together.

Many other profs also choose to not allow cheat sheets and give out equations instead (although it's only good if they actually tell you which equations you need beforehand). We might be given something like the solution to the square well anyways for a question that involves a modification of that potential, in some way.
And I definitely don't agree with homework being a majority of your grade, as students can easily use the internet to find all the answers they need.
This one is tricky since most major textbooks do have solution sets online (e.g. Griffths, Jackson, etc.). But, so what? If a student "cheats" on all of his/her homework then they will not do well on the midterm and final exam. Even without internet solutions, a student can still "collaborate" with others and there's no way to tell a student who contributes to their study group vs. a student who uses it to get answers. But, in the end, they still need to know what they're doing to get a high grade. In addition, most of my senior courses involved a term paper or project. It's pretty hard to get something like that off the internet! I think projects and papers, especially in graduate school, are a much better way to testing whether or not the student has understood the material. In addition, I think the ability to do something like a research project or to show your knowledge through an exam or homework (i.e. where the marker can see your thought process and evaluate you on that) is a much better indicator of grad school potential than the PGRE.

But GPA is tricky since different schools, departments, and profs all grade differently! This is why I think your undergrad school matters (original topic of thread?). People can use the school's reputation (and especially the performance of past students they've looked at/accepted from that undergrad school) to judge the "quality" of a student's education and thus how to interpret their GPA.

I definitely agree with the thoughts that neither PGRE nor GPA is not a good indicator of physics ability by itself. I actually think things like research experience and LORs are more important than either PGRE or GPA.

It is also interesting to note that the PGRE is really only required for US Physics graduate programs. Canadian and European programs don't really use PGRE scores. My undergrad university (Canadian) uses GPA as a minimum cut off (Faculty of Grad Studies regulations; but the rules could probably be bent on a case-by-case basis) while others do not have any cut off. All schools will evaluate the entire application package though.

Re: Why does your Undergraduate School Matter?

Posted: Wed Jan 16, 2013 8:34 pm
by twistor
You asked a question:
If some student went to a no-name liberal arts school, got a 990 on the PGRE, and spent a summer studying at SLAC, what does he lack when compared to a student who went to a top-tier physics school (such as MIT or Bob Jones University )?
We answered it. What more do you want? You're free to contemplate the philosophical implications while you prepare your applications, applications that include your school, GPA, and test scores.

Re: Why does your Undergraduate School Matter?

Posted: Thu Jan 17, 2013 1:43 am
by Meteorshower
I struggle to comprehend how PGRE can be considered even vaguely comparable to the importance of GPA.

An absolute majority of the PGRE was taken from courses I sat in 1st or 2nd year. It's also a multiple choice test! Sure it's probably about as good at testing physics as a multiple choice test can be, but there is no scope for testing any skill that takes more than a couple of minutes to exhibit. It also focuses on some areas that to me seem fairly trivial - formula memorization, not making sign errors etc and even finding order of magnitude quickly isn't that important to any physicist that can work a calculator.

Also, it's just one test! Presumably you take many physics courses, where having a bad day wont markedly affect your average. Sure it can be sat multiple times, but we can't all afford that.

Re: Why does your Undergraduate School Matter?

Posted: Thu Jan 17, 2013 10:56 am
by Throwaway1
Agreed. I think the PGRE is a poor measure of aptitude compared to GPA. And if we want to talk about physics aptitude, the most relevant parameter is research experience, in my opinion.

I find it ironic that "goingnuclear" accuses others of playing the academic system when the PGRE is all about playing the test.

Moreover, just because you have a cheat sheet on an exam, that doesn't imply you know how to use it. Some of my exams were even open book and they were some of the hardest exams I ever took.