The Physics GRE, thoughts? Opinions?

[Pengwuino]

So a professor of mine and another student were having a discussion about the PGRE the other day. While it does seem to be of the overall opinion that it is only good for determining how well a student can take the PGRE and not really how good they are at being a physics student, I've come to a different conclusion. I've noticed that the people in my department who can't do physics worth a damn and who should have picked a different major a long time ago are the same ones who can't score beyond 20th percentile even while they're doing their masters degree at the time of taking the test.

For example, one of these students I posed a question to: If you drop a ball off of a building and simultaneously let another ball roll off a ramp connected to the building at the same height, what would reach the ground first? They couldn't figure it out... and these are people who want to go get their PHD and teach this stuff.

So most people say you shouldn't judge someone based on how well or poorly they do on the PGRE. My contention is that if you can't do well on it, you have no business trying to get into a PHD program. Your thoughts and opinions?

 

[magpies]

if that's exactly how you worded the question and some how I bet it was even more vague when you asked it then I can understand why they didn't give you a good answer.

 

[billiards]

I don't know about the physics GRE but the general one I did. The verbal bit seemed to basically be a vocabulary test. It seemed to be more about what you know, rather than how you think.

Also I hate that you can't go back to answer questions. I think that plays to the advantage of people experienced in this kind of testing, again, not so much in quality of thought.

 

[cronxeh]

For example, one of these students I posed a question to: If you drop a ball off of a building and simultaneously let another ball roll off a ramp connected to the building at the same height, what would reach the ground first? They couldn't figure it out... and these are people who want to go get their PHD and teach this stuff.

What is the friction due to air and drag? What is the height of this building, and is there another massive object around the two balls? What kind of material is the ball made of? Is there a wind breeze, or perhaps freezing temperature variations that could affect the friction rates of the ball on the ramp, and what is the deal with airline food?

 

[Pengwuino]

if that's exactly how you worded the question and some how I bet it was even more vague when you asked it then I can understand why they didn't give you a good answer.

No, this happened half a year ago, I don't remember exactly how I worded the question but obviously it was not a problem of vagueness. It was your typical idealized point particle frictionless world. Even so, absolutely irrelevant. The fastest path is straight down along the gravitational field regardless of moment of inertia or friction or what have you outside of one object being shielded from the hurricane that was tearing down the building and only one of the objects.

 

[flyingpig]

No, this happened half a year ago, I don't remember exactly how I worded the question but obviously it was not a problem of vagueness. It was your typical idealized point particle frictionless world. Even so, absolutely irrelevant. The fastest path is straight down along the gravitational field regardless of moment of inertia or friction or what have you outside of one object being shielded from the hurricane that was tearing down the building and only one of the objects.

would it depend on how slanted it is?

 

[Chunkysalsa]

I'm no physics student (yet) but what's the answer to that question? Assuming that everything i.e friction is negligible then won't they hit the ground at the same time.

 

[cronxeh]

I'm no physics student (yet) but what's the answer to that question? Assuming that everything i.e friction is negligible then won't they hit the ground at the same time.

No. The ball travels a longer distance on a slant, and the force of gravity is multiplied by the sine of the angle of the slant, hence the smaller the angle the slower the ball moves. The maximum acceleration due to gravity is achieved at a 90 degree (vertical) angle or simply free fall.

If the problem was dropping a ball from height and shooting a bullet horizontally from the same height at the same time, in theory the bullet should hit the ground at the same time as the ball because there is no obstruction for the bullet to go downward vertically as is the case with the ball on a ramp.

 

[Chunkysalsa]

Ah I see, I was thinking of the gun example. I just thought the freefall option was too obvious, got tricked by thinking it was a trick lol.

Well I have lots to learn when I start physics next fall. Never really cared for the subject back in high school.

 

[Leptos]

No. The ball travels a longer distance on a slant, and the force of gravity is multiplied by the sine of the angle of the slant, hence the smaller the angle the slower the ball moves. The maximum acceleration due to gravity is achieved at a 90 degree (vertical) angle or simply free fall.

If the problem was dropping a ball from height and shooting a bullet horizontally from the same height at the same time, in theory the bullet should hit the ground at the same time as the ball because there is no obstruction for the bullet to go downward vertically as is the case with the ball on a ramp.

Common sense tells me at least two things:
A) The ramp is like a hypotenuse of a triangle whereas the length straight down is like a side.
B) Ramps decrease the rate of work being done.Anyway, I'm currently in my first semester(Spring 2010) of college, when should I plan to take my GRE? I'm thinking that I will probably take an extra semester or two to graduate because I want to double major, but I'm not sure if that will hamper my chances of getting into a good grad school.

 

[Pengwuino]

No the question wasn't the bullet out of a gun example you hear so much about. The ball hits the ground first that is dropped straight down.

Whatever. That isn't the point here, the question is the usefulness of the PGRE and everyones opinion on it.

 

[Vanadium 50]

If you drop a ball off of a building and simultaneously let another ball roll off a ramp connected to the building at the same height, what would reach the ground first? They couldn't figure it out... and these are people who want to go get their PHD and teach this stuff.

Note that this doesn't require any knowledge of physics - just observations about what happens with ramps. (And "ramps" that we see every day - we call them "hills") But I agree that someone who has trouble with this idea will have trouble in grad school, and will - or at least should - have trouble with the PGRE.

 

[Klockan3]

Whatever. That isn't the point here, the question is the usefulness of the PGRE and everyones opinion on it.

After reading the questions it seems like it is a good measurement of the students physical intuition which while it isn't the be all end all skill to have for a physicist it is certainly extremely handy.

Some are just awful at it which is why the test is needed, it isn't something you can see from his grades or courses or anything really.

 

[Gokul43201]

After reading the questions it seems like it is a good measurement of the students physical intuition which while it isn't the be all end all skill to have for a physicist it is certainly extremely handy.

Some are just awful at it which is why the test is needed, it isn't something you can see from his grades or courses or anything really.

The Physics GRE is most definitely a lot more than a test of intuition. It is a test of your knowledge of the most basic concepts and principles in a number of subfields. If you can't get at least half those questions right, you have not been exposed to enough physics or haven't absorbed enough. You need to spend more time with Physics before getting into Grad School.

Similarly, I believe that if you can not score over 750/800 in the Quant section of the General GRE, you haven't been absorbing enough math for Grad School.

 

[Klockan3]

The Physics GRE is most definitely a lot more than a test of intuition. It is a test of your knowledge of the most basic concepts and principles in a number of subfields. If you can't get at least half those questions right, you have not been exposed to enough physics or haven't absorbed enough. You need to spend more time with Physics before getting into Grad School.

Well, the problem with your argument is that there are people who can do all of the calculations and would be able to answer everything on that test but would do it too slowly to get a good score.

Of course I would argue that then they don't understand what they have done but that in no way means that they are not able to do well in grad school or in academia in general.

 

[Pengwuino]

Well, the problem with your argument is that there are people who can do all of the calculations and would be able to answer everything on that test but would do it too slowly to get a good score.

Of course I would argue that then they don't understand what they have done but that in no way means that they are not able to do well in grad school or in academia in general.

There are a lot of conceptual questions, however, which I like (even though for some reason most the people I know hate the idea...). Other then that, if someone realizes that ETS designed the questions to take less then 2 minutes and they're solving a mechanics problem through say, the entire Hamiltonian formalism, it's a good indicator that they don't have a lot of intuition about the problems.

For example, a popular question with them seems to be 3 masses connected linearly on a spring all with the same mass, same spring constants. It looks like:

OxxxxOxxxxO

They give the tough frequency of oscillation and then would ask what is the other oscillation mode? I know a few people, when given that question, derive the whole problem through Lagrangians and small angle approxmations and all that and don't realize, even though they've seen and done the problem before in a class, that you just get your 1-mass frequency when you hold the middle one and let the 2 side masses oscillate by themselves.

 

[twofish-quant]

For example, one of these students I posed a question to: If you drop a ball off of a building and simultaneously let another ball roll off a ramp connected to the building at the same height, what would reach the ground first? They couldn't figure it out... and these are people who want to go get their PHD and teach this stuff.

Unless they've changed the PGRE radically in recent years, this sort of question is not the type of question that the PGRE is focused on.

Also, someone ask you this sort of question, you get it wrong. The next time you get asked the question, you get the right answer.

 

[twofish-quant]

Other then that, if someone realizes that ETS designed the questions to take less then 2 minutes and they're solving a mechanics problem through say, the entire Hamiltonian formalism, it's a good indicator that they don't have a lot of intuition about the problems.

They may have changed the test in the years that I took it, but when I took it, I found that intuition was pretty much useless in taking the test. The problem is that the with intuition you could narrow down the problem to two answers on the multiple choice, but once you got to those two answers, you had to do the full calculation to figure out which one of the two was correct.

EDIT: I just looked at the latest PGRE, and it doesn't seem that different from the one that I took. I hated the test because I found that I couldn't get very far with physical intuition, and I'm not that good at calculating things quickly.

The other thing is that if I had to take the test right now, I probably wouldn't do that well on it. Give me two weeks of prep work, and I'll do fine. This suggests that the test measures something other then simple physical intuition (which is not necessarily a bad thing).

 

[twofish-quant]

So most people say you shouldn't judge someone based on how well or poorly they do on the PGRE. My contention is that if you can't do well on it, you have no business trying to get into a PHD program. Your thoughts and opinions?

First of all, there is a difference between can't do will and didn't do well. A huge amount of getting a high score on the PGRE involves just test taking skills, which can be learned. If someone gave me the PGRE right this moment, I'd totally bomb it.

The thing about test taking skills is that are minorly important in grad school, but a lot less important than other skills. Also depending on the topic, physical intuition may or may not be important, and in any case you can teach intuition.

 

[Klockan3]

EDIT: I just looked at the latest PGRE, and it doesn't seem that different from the one that I took. I hated the test because I found that I couldn't get very far with physical intuition, and I'm not that good at calculating things quickly.

What you need is to quickly isolate what is important, then you can solve most of the questions in ~20 seconds by just plugging some basic formulas that you should know. (And 20 seconds is on the far side, isolating what is important will take the majority of the time)

At least I call that intuition, seeing the structure of the problem without doing it. It can be manifested in pictures such as "Which of these represents the correct acceleration in the following points" or it can be through problems that if you do it the normal way you need to calculate a lot but if you see through the problem then it is basically already done.

But of course as said this is not a skill that is pivotal to success within physics grad school, it just helps.

 

[ctg7w6]

How much should you have studied before you take the PGRE? As in... I just took the GRE at the beginning of this year. I have one year left, but I am doing almost my entire physics degree in this last year. So if I want to apply to grad school in physics I need to take the PGRE early Fall semester 2010. I have only taken the two intro calc based physics classes. I will take all the upper division classes next year.

If I study can I do well, or do I need more classes? Or is it a test of basic physics skills, and I do mean pretty basic?

 

[Gokul43201]

Well, the problem with your argument is that there are people who can do all of the calculations and would be able to answer everything on that test but would do it too slowly to get a good score.

Again, if one can not answer even half the questions correctly in the time allocated for answering them all, I would recommend spending more time with basic physics skills before going to grad school.

How much should you have studied before you take the PGRE? As in... I just took the GRE at the beginning of this year. I have one year left, but I am doing almost my entire physics degree in this last year. So if I want to apply to grad school in physics I need to take the PGRE early Fall semester 2010. I have only taken the two intro calc based physics classes. I will take all the upper division classes next year.

If I study can I do well, or do I need more classes? Or is it a test of basic physics skills, and I do mean pretty basic?

You may be better off taking all the necessary classes. One way to know what's necessary is to compare the PGRE syllabus (from the ETS website) with the syllabus of material you've covered so far, and what is left to be covered next year. Or you could talk to an academic counselor.

 

[eri]

I'll admit I didn't spend too much time studying for that test, and my scores reflected that. However, students coming from liberal arts colleges tend to do poorly on the PGRE anyway, and they tend to blame that on idea of 'concepts over equations'. Most people from my liberal arts school who went on to grad school in physics did just fine. My grad school doesn't even require PGRE scores any more because they didn't see a correlation between high scores and who finished a PhD. I'm going to finish the PhD in a few months (assuming I get a job offer soon), passed the quals with no problem (but actually studied for those) and have a very high graduate GPA despite low PGRE scores. It's not always a good predictor, and it's unfortunate many schools use it that way.

 

[twofish-quant]

What you need is to quickly isolate what is important, then you can solve most of the questions in ~20 seconds by just plugging some basic formulas that you should know.

Didn't work that way for me. The problem was that with intuition i could only narrow things down to two or three answers, and then couldn't figure out what to do from there. It's also not good that I'm quite bad at arithmetic. The typical problem that I totally hated was when you had four answers and the only thing that was different was the signs.

Something that did help me was that I was able to triage questions. After spending X minutes on a problem, I was able to just give up and go to the next one.

 

[twofish-quant]

How much should you have studied before you take the PGRE? As in...

Take some practice tests. Also once you do well on the practice tests, I'd recommend that you schedule the tests so that if you bomb once, you have another bite of the apple..

If I study can I do well, or do I need more classes? Or is it a test of basic physics skills, and I do mean pretty basic?

Like all tests, you will do better if you study the test. They tend to have the same sorts of questions, and if you get good at cranking out the answers to those questions, then you'll do better on the PGRE.

 

[twofish-quant]

One other thing. Learn SI units...

I did somewhat worse that I should have because my physics curriculum was all in CGS, and so I had no idea what the conversion factors were when I did the PGRE.

 

[darkchild]

my physics curriculum was all in CGS

Really? Was this in an engineering program or something?