An Inside Look at the Physics GRE

By John Conway | October 22, 2009 10:01 am

I am just back from Princeton where we held the annual meeting of the GRE Physics Committee of Examiners, a group of six, ahem, distinguished professors (we have grey hair) who sit around a conference table working through hundreds of potential and actual Physics GRE problem. Each year new exam forms are completed, new questions added to the pool, statistics reviewed, and a good time is generally had by all.

This was my last meeting – I have served on the committee for six years. The membership rotates roughly every two years. I had been an external reviewer and problem writer for a couple years, and was then asked to serve on the committee. I am sworn to secrecy about a lot of the details, for good reason, but let me try to tell you from my perspective as an exam writer how to study for this dreaded event in your physics education.

Firstly, there’s the format. The exam is 100 questions long, and you have 170 minutes to do it. This is, therefore, different from just about every other physics exam you have had in college, where you have, say, four to six problems in an hour-long exam. The GRE Physics problems (or “items” in assessment world jargon) are short, to-the-point questions, and just about all of them are short calculations, if any, and take little time once you see what to do. Writing such questions is a difficult thing to do, let me tell you. We are continually amazed how, after about six levels of review, we can find issues of clarity, reasoning, and even sometimes basic physics correctness in the items submitted to the pool. All the committee members spend a lot of time each year reviewing hundreds of problems, looking for flaws, but more often than you would think the face-to-face meeting in Princeton with the ETS folks reveals something previously overlooked. It’s a really interesting process.

For each new exam form we eventually arrive at 100 items that test mastery of a clear physics concept or idea, and there is, yes, a certain amount of memorization required in terms of the basic equations learned in undergraduate physics. But there are many problems that can be done using just concepts, and many that can be done with simple dimensional analysis. When there are numerical solutions (and many if not most are in that category) the numbers are chosen so as to allow easy arithmetic – no calculators are allowed.

My first piece of advice to students studying for this exam is to focus on reviewing the textbook from your freshman introductory physics course. In my years on the GRE committee, when I have needed to consult a text, it is that text at least 80% of the time. If you master every example in there and review the basic equations, you will do really well on the GRE. I have found that only a small fraction of the items on the GRE are actually from upper-level topics like stat mech, quantum, and special topics (solid state, nuclear, particle, cosmology, etc.) And presumably you have been studying the advanced topics more recently anyway. I think the single biggest mistake students make in studying for the GRE is to focus on too-advanced subjects.

The other piece of advice I give students is to be disciplined in your approach to actually taking the exam. You only have an average of 1.7 minutes per problem! If you get bogged down on a long algebraic calculation, you risk not being able to complete the exam, including items that you would correctly answer in a few seconds. So when you take the exam, read each problem, answer it if you can do so reasonably quickly and then put an X on the problem number. If you think the problem will take some time or a long calculation, put a circle around the number and come back to it in a second pass through the whole thing. But pIck off the easy ones first! It also helps build confidence as you go through.

Also realize that the GRE penalizes random guessing: your raw score is the number correct minus the 1/4 times the number incorrect. As a result it’s no better to guess than to leave an answer blank if you cannot eliminate some of the five choices. But if you can eliminate some, then by all means guess! Look carefully at the possible answers – sometimes just the units, or magnitude, or mathematical form can give you a way to guess more astutely.

So just what is the GRE measuring? A critic might point out that it measures the ability to work under pressure, memorization, and quick mathematical reasoning and calculation. Though these are good qualities for a physicist to have, they are by no means the only qualities required for a successful career. I would argue further, though, that the Physics GRE really does test knowledge about basic physics and the ability to analyze physical situations accurately.

So then how important is the Physics GRE for your career? It turns out that it is in fact quite important. Some of the top programs in the US even go to the extent of requiring a GRE score above some threshold for considering the applicant. I have served on our graduate admissions committee for five years now, and I can tell you that we regard the GRE as just one piece of information telling us how likely a student is to thrive in our program. We do see a clear correlation between an incoming graduate student’s Physics GRE score and their score on the other dreaded exam in a physics student’s career, the Ph.D. written preliminary exam, which is a very different beast. (There was, a few years back, some lore that the GRE Verbal score was a better predictor than the GRE Physics score, and there is a correlation, but not as strong as with the GRE Physics score.)

In considering an applicant we look at a number of things, including the applicant’s own statement, experience, letters of recommendation, and their undergraduate transcript, in addition to the GRE general and subject scores, to get an idea of the whole student. My own observation is that students below about the 30% level have a very hard time attaining a Ph.D., though this is by no means absolute. I am sure there are tons of very successful physicists out there who, for whatever reason, scored poorly on this peculiar exam and went on to great careers.

So, to of those of you facing this exam in a few weeks, I wish you good luck! Review your intro course, get a good night’s sleep before the exam, and make sure you pick off all the easy problems that you can!

CATEGORIZED UNDER: Academia, Advice, Science
  • okaasan59

    I wish all tests came with such useful and encouraging advice. After reading this I almost felt like trying my hand at the Physics GRE, despite never having taken a physics course in my life! But I will keep this advice in mind and pass it on to my son who is considering a career involving physics.

  • Dave Goldberg

    I run a “how to get into grad school and ultimately further your scientific career” workshop with our undergrads ever year or so, and I try to hammer home both the importance of the subject exam and also how different it is from just about every other test they’ll take, before or after. I’m going to make your entry required reading for this year’s class.

  • An Undergrad

    In your experience, around where do top-tier schools put that cutoff (for the schools that do use one), and what sorts of scores are cause for concern? For an undergrad just taking the test, it can be difficult to know what a score really means. Thanks for the post!

  • incredulous

    I found the experience of taking the Physics GRE to be one of the most pointless endeavors in my entire educational experience.

    As you say, John, 80% of the material comes from a first-year undergraduate textbook (by which I suppose you mean something of the caliber of Halliday, Resnick, and Walker). But many physics majors took honors-level introductory courses with more of an emphasis on derivation and methodology than reproduction of answers and formulas by rote – that is to say, learned how to do physics rather than answer physics problems. It is astonishing and borderline disturbing that, after four years of learning more and more about how to do physics, our academic futures are contingent on recall of material from a class that we were encouraged not to take.

    Really – does the physics GRE convey anything about one’s ability to do physics? To learn physics, to teach physics, or to research new physics? I would argue it does none of these things; yet, somehow, it is a rather large impediment to being able to do so.

    The physics GRE as a standardized test is not particularly unfair or biased; in that sense, the problem-writing committee is doing a fine job. But the emphasis on the standardized test is a bizarre and completely unwarranted relic that tests all the wrong aspects of physics and yet somehow carries real importance in the lives and careers of many physicists.

  • cmt

    I’m interested in the data correlating the GRE score and a student’s success in a Ph.D. program.

    One of the reasons I’m often suspicious of standardized tests in general is that the range of skills tested *seems* narrower, or at least weighted differently, than the range of skills necessary to succeed. For example, the emphasis on fast calculations in the physics GRE, especially with numbers, though a reasonable skill for a physicist to have, is not obviously more important in determining success than the ability to formulate a mathematical problem from a set of physical ideas. Nor is it even obviously related to the ability to solve a particularly knotty differential equation. Nor is it obviously related to the ability to design and execute a clever experiment. You could argue that these skills are correlated, and they may be. But it’s just so easy to find exceptions in both directions (high scorers who can’t think logically and low scorers who are quite good physicists) that I’m just not that confident that the scores mean very much.

    Full Dislosure: I didn’t do so well on the GRE compared to my peers (though I beat the 30% cutoff by a comfortable margin). On the other hand, I think it’s safe to say that I’ve been happily successful in my career as a theorist so far.

  • anonymous

    I have to say, when I took this exam, I found it to be a very poor measure of actual physics ability. As a full disclaimer, I did quite well on the exam itself, got into several excellent graduate schools and got my PhD, etc. etc. So please don’t read this critique as a disgruntled “The GRE ruined my career,” which it did not, it likely helped it.

    For one thing, the time pressure is rather absurd. At the PhD level, I’d argue the most relevant skill set is ability to think independently, break down a problem and reason through it. There is simply no way you can do that in 1.7 minutes for a given problem, and you end up penalizing many students who either don’t think fast, or don’t see the way through a problem nearly instantly. In addition, many students simply are not good at this kind of speedy, test taking, which in itself is a skill that really has little to do with physics at the professional level.

    The other issue I would have with the subject GRE is that, at least years ago when I took it, there were quite a few questions in more obscure areas. By obscure here, I don’t mean unimportant, but instead subjects that are not likely to show up in a typical physics progression for an undergrad. For example, I recall quite a few detailed optics questions (things you really couldn’t know without a full course in the subject), detailed nuclear physics etc. that just are not present in the majority of physics undergraduate progressions in any detail. While you can argue that these are still a small proportion of the questions, at the high end, for example if you’re applying in high energy theory to elite schools, every little point matters (and I’d argue remembering the formula for some lens behavior isn’t really testing anything beyond that).

    Now, these comments, coupled with your “threshold” for some graduate schools I find to be disturbing. Unless that threshold score is very low, I find that to be an incredible injustice to quite a few students who have great physics potential but may simply be poor multiple choice test takers. You mention that there is some correlation with performance on the GRE and the performance on the qualifier, but exactly how much? And then how much correlation is there between that performance and “success” in physics. There are several elite schools that do not have qualifiers, Harvard, Santa Barbara and UPenn come to mind and I think we can all agree that those places have produced quite a few excellent physicists. I’m also fairly certain there are plenty of students who have aced quals at Princeton and Chicago and then sort of petered out when it came to doing actual research.

    If the subject GRE represents one small part in admissions, that’s fine, but I think way too much weight is given to it and speaks to a need to have a “standardized” way of measuring something. I’m just not quite sure what its measuring and it puts an undo amount of pressure on already stressed out students.

  • Roban

    I largely agree with what comments above have said, and could rant for a long time about the shortcomings of the GRE and similar high-stakes multiple-choice testing.

    My main objection would be that there is nothing on the GRE that remotely resembles anything I do as a professional scientist. For instance, the bizarre ban on calculators: I am very slow and error-prone when I try to do arithmetic without a calculator. I find this to be a hindrance in my scientific life roughly twice a year. However it was incredibly debilitating, frustrating, and distracting when taking the physics GRE.

    The GRE does have some resemblance to other forms of testing, so I’m not surprised if it correlates with your comprehensive exam results. But that correlation doesn’t show that it’s actually useful in predicting success and productivity as a working scientist (or as an advanced graduate student, or as a student in courses that don’t rely on time-limited testing for evaluation).

    Obviously you won’t do well on the GRE if you don’t have some basic knowledge of the subject, but you can also do relatively poorly for a variety of unrelated reasons that might not have any impact on your ability to be a productive scientist. I think it is also troubling that there is evidence that suggests that the kind of high-pressure high-difficulty environment fostered by GRE testing is disproportionately hard on under-represented groups (see literature on the “stereotype threat”).

    Now that I’ve gotten all that off my chest, I applaud your contributions to making the questions as reasonable and clear as possible within the constraints of the system, and for your commitment to using multiple sources of information in admissions, and interpreting GRE results in the broader context of an applicants whole file.

  • Ned Wright

    “Also realize that the GRE penalizes random guessing: your raw score is the number correct minus the 1/4 times the number incorrect.”

    This is actually neutral. Guessing and not answering have the same expected value of zero.

    When I took the GRE I had to chisel my answers into stone, but I found that finding equivalent answers helped a lot. Because if A is equivalent to B, then neither can be right.
    And if A=B and D=E then C must be the right answer.

  • David Nataf

    One day, when there’s an equal amount of grade inflation or even no grade inflation in undergraduate programs, there will be much less of a need for the GRE. I recall in my last two years of undergrad, there were large fears among my peers due to rumours a lot of other universities give everyone a 4.0.

    There was a new story from a few years back that ~90% of Harvard students graduate with “honors”.

    It’s also a source of resentment among undergraduates, which I’ve seen in a few places, that people who take much easier classes will be rewarded for their higher GPAs.

    Basically, for all the points against the GRE, the GPA is much less objective. I did rather well in both before somebody assumes I’m biased.

  • John

    Ned: you are correct that subtracting 1/4 point for incorrect answers, on average, is neutral with resect to guessing. I updated that sentence in the post to clarify this.

    As for equivalent answers, we do try to make sure there are five logically distinct choices; rarely it may happen that we missed the fact that there are two or three mutually exclusive choices out of the five, in which case one must be correct.

    I agree with the comments about standardized testing as being an imperfect measure of skills which may only have partial relevance to a scientist’s actual career. This is true of the SAT, the GRE, and the graduate qualifier/prelim. But what is the alternative? We cannot admit every applicant to our Ph.D. program: we simply don’t have the resources. Transcripts from different undergraduate programs are even less reliable as an indicator of probability of success, and letters of recommendation vary wildly. I am sure we miss many fine applicants due to being put off by low GRE scores, but we do admit some.

    This having been said, I would say this to you, anonymous: the ability to think and work quickly and accurately is among my most prized skills as a physicist. I am under time pressure every moment of my career. :/

  • King Cynic

    I formerly served on the graduate admissions committee of a “Top Ten” school in the US that I will not identify here. The lowest GRE score of any student we offered admission to in my time on the committee was 650, but that was an exceptional case. I think we worried a little about anyone with scores below 750. The median among those offered admission was above 800.

  • hackenkaus

    It’s pretty amusing to read the sour grapes of people who took the test and were obviously disappointed their score did not reflect their own perceived level of brilliance. Objective tests can be brutal in that way for those who have coasted through on grade inflation and overly supportive professors. Also on those who think they’re really smart ’cause they took some advanced courses, to find that their knowledge is shaky enough they have trouble answering some freshman level questions. People who went to non-elite schools will often suffer from big fish small pond syndrome, and those who went to elite schools will feel they deserve a high score by virtue of having gotten into an elite school.

    Newsflash, the test is not that hard. Stop making excuses.

  • anonymous

    To hackenhaus, sorry I thought I was pretty clear I was not making excuses. I took and did quite well on the subject GRE. I’m generally good at test taking, as a skill so my comments are not sour grapes. I’m fairly certain that when I applied to graduate school, checking off that I wanted to do high energy theory was a larger impediment to admissions than my GRE scores (some of which were actually perfect). My comments were my attempt to be objective about a test I feel is flawed.

    John, I’m not against some objective measure per se, I’d just argue that the current format of a large number of relatively memorization based questions is really not the ideal since it has little to do with being an actual physicist (ditto the lack of calculators allowed). The time pressure you are under at the professional level, at least in my case is more of a long term thing, e.g. needing to get a paper out before a rival group does, that is not really on the time scale of 1.7 minutes. I’m not sure what the exact solution is, but I would like to see a test that has many, many fewer problems, or even gives credit for “showing your work.” If I were on an admissions committee, I’d rather have the student who understands the basic concepts but can also reason through a more involved problem than the one that can rattle off answers rapid fire nearly every time. There are also some countries that seem to overprepare their students for these tests, with the result that nearly every student gets a perfect score and and admissions committees have to “renormalize” for these.

  • O. E. Parker

    David Nataf: For what it’s worth, starting in 2005 Harvard placed strict limits on the number of honors degrees issued.

  • Dave Goldberg

    We’re all informed by our own experience, of course. Those who did well on the exam tend to get into top grad programs, and those in top grad programs tended to get good postdocs and faculty positions (i.e. the measures of a successful career). There is something tautological in this, and it’s why I make such a big deal about it to my own students.

    It’s also true that a priori, the Physics GRE is not typical of my experiences in graduate school or beyond. That said, I have to agree with David Nataf when he points out that it is currently the only objective standard we have of distinguishing between students from very different programs. And if I can be informed by my own experience, I will say that for the most part, over the last 8 1/2 years, I’ve seen a lot of students take the GRE, and there is a very tight correlation between other measures (GPA, performance in the lab, intellectual curiosity) and their performance on the exam.

  • John

    anonymous: I have a hard time disagreeing with you, so I won’t. :) One point worth mentioning is that the European institutions seem to get by without requiring the GRE…since I work with a lot of European physicists I should ask them how their admissions process really works.

  • loonunit

    Hi John, what’s the current status of the gender gap on the Physics GRE? I seem to recall hearing there was a nearly 30% median gap a little over a decade ago, but that the gap had shrunk considerably in recent years.

  • anonymous

    John, I’d love to hear how they are. As it stands right now, it just seems like a way to set a bar in admissions. I’m just not sure that bar is the right one. I mean, I could also take the people who could do the most pushups. That’s objective and independent of where they went to school, just not sure how relevant it is (and yes, that was an extreme example, I’m aware).

    As for what Dave G. says, if you take students who did very well on the GRE, I’m sure you’ve got a bunch of pretty smart people on the whole. To me, the more relevant questions are how many smart people are you missing because they aren’t good at this particular type of test, that we can all agree has little to do with actually “doing” physics at the PhD level? And why has this test become the objective way of measuring things? Why not pursue a better test? or something else? I fear that its become more about inertia, that this is the way its always been done, its objective so lets just keep doing that. I just don’t buy into that.

  • Eugene

    I semi-tanked my GRE Physics. But that’s because I didn’t have a physics degree when I took it, and was studying for it *while* holding a full-time job.

    For some reason, I managed to trick one of the regular bloggers on to take me as his student anyway.

    (Seriously : I think studying for the GRE was good for me, because I have no physics background. On the other hand, looking back, it is probably one of the least useful metric of a future student’s ability. I remember Sean posting something to this effect a while back. Bloggers-fight, commence!)

  • Jordan

    I’m a recent physics grad about to take the GRE.

    I certainly feel like there are better ways to measure my knowledge of physics, and my chances of success as a physicist. I don’t think I’ll ever need, for example, to have the resonant frequency of an RLC circuit memorized.

    On the other hand, I will definitely need to be able to focus on a single problem for longer than 2 minutes straight… and it just so happens that a lack of ability to focus on a difficult problem for hours on end hurt me in undergrad and would hurt my abilities as a physicist, assuming I survive grad school. This test won’t measure that flaw. Nor will it measure my ability to see the right way to attack a problem, or my ability to apply my knowledge creatively to a difficult problem. I think all of these are probably more important than remembering in less than a minute how to solve an elastic collision.

  • Sean

    I have probably mentioned this before, but when I was at Chicago we did a survey of our faculty about how well recently graduated students had turned out (as working physicists), then plotted it vs. physics GRE scores. There was a tiny island of people (less than 10% of the sample) who did very well on both measures — all were mathematically-oriented theorists. For the other students, there was no correlation whatsoever between GRE scores and success as a physicist (at least within the highly selected group that came to Chicago for grad school).

  • Jennifer West

    I read an article about 10 years ago by Howard Georgi (cannot find a link to the article, but a brief mention is here: who stated that if Harvard based its graduate admissions decisions on the subject GRE, EVERY STUDENT would come from the People’s Republic of China.

    He said that doing well on the physics GRE is a skill, and it can be taught and learned, and may or may not have a thing to do with whether you can do physics.

    I remember my own experience with the general GRE, back when the analytical thinking section was logic problems (along the lines of the LSAT). I scored an 800, which is perfect. Why? Because I trained myself like a rat, did many similar problems, and was eventually able to be very very quick in these type of problems.

    For the physics GRE, I got about a 50% I think, but at any rate it was a bad score. And my problem was that I didn’t study the freshman physics enough, so did not have the equations fresh in my mind, and that I actually get curious about problems, so start thinking. I remember I was looking out the window in a school in Boston where I took the test, in some dim corner of my mind registering squirrels running around being cute, while thinking about what some problem might mean. I know, it wasn’t smart.

    But I like Georgi’s article, and a lot, because he helped me understand that failing the GRE (and a 50% in any normal test would be a complete failure, though I did get into graduate school) MIGHT not mean that I am totally unsuited for science. That’s all.

    Personally, I would prefer oral exams. I would like schools to pick their top choices and fly them all out to talk with professors at the blackboard. It’s the same as how I feel with my own students now that I’m teaching in grad school – I can get a much better handle on how they think and what they know by having a discussion with them at the board, for one hour minimum. Plus it is fun.

  • John

    Sigh. I am just trying to help the students who, like it or not, in November will face this exam if they want to get into grad school. We can turn it into a debate about whether or not it should be used for that purpose but the fact of the matter is that it is used, and will be for some time to come. I have worked rather hard to help make it a clear and straightforward test of basic physics knowledge which all aspiring physicists should possess (yes, including the fact that RLC circuits have a resonance at a particular frequency – it’s applicable in the most unexpected situations…)

  • Schwa

    I’m a medical student who teaches for Kaplan test prep, and while I haven’t taken the physics GRE, I have taken the MCAT and general GRE. If these tests are absolutely unlike the physics GRE, ignore what I’m about to say.

    The first thing that occurs to me is that the aim of the MCAT is to identify people who will do well in medical school and on the boards. This is slightly different from identifying ALL the people who will do well in medical school – while I’m sure that the AAMC and ETS and so on would like to give everyone an equal chance, I’m not convinced that that’s their number one priority. I don’t think this is necessarily a good thing, but I do think that cries of unfairness to students who aren’t, e.g., good at mental arithmetic are sort of beside the point: if being able to do mental arithmetic is strongly correlated with being a competent doctor/physicist/whatever, there’s no reason to expect testing agencies to ignore such an easily implemented filter. I don’t write or do statistical analyses of tests, though, so I don’t know the extent to which Sean’s observation about Chicago undermines this idea.

    This gets into a slightly more interesting question: how do you identify people who will do well at whatever you’re trying to filter for? For content based tests like the MCAT, the two big reasons I can think of off the top of my head are:

    a) ensuring a minimum level of background knowledge so that students can understand how their equipment works (i.e. it’s handy to be able to look at a schematic of GLAST and tell whether or not the data it’s producing are relevant to your project). I know that this is huge for doctors, who work with a really varied array of equipment and need to not only be able to instantly tell when data are unusual, but also figure out what’s responsible for the abnormality.

    b) demonstrating that students can take a body of knowledge with which they have passing or minimal familiarity, and can quickly obtain an understanding sufficient to answer basic questions about that subject (i.e. can you look at a paper on a topic you looked into 4 years ago and be able to apply what you remember to this unfamiliar situation). Ideas from outside one’s area of expertise can often contribute a lot to solving problems within that expertise, but you can’t recognize and use those ideas unless you have a broad base of knowledge to work from.

    So anyway, expecting people to have a broad base of knowledge isn’t obviously stupid on ETS’ part, and it’d be interesting to know how many people with great GRE scores end up being total goobers in their professional careers.

  • An Ominous Coward

    So I’m about to graduate with the coveted physics degree, and I’m taking the physics GRE coming up here in a couple weeks. Like many of the other students, I’ve been diligently working through the four or so old exams that have been released.

    I took the first one. I think I scored in the 650 range. Practiced some problems, took the second one. Mid 700s. Did I learn any new physics between testings? Nope! I learned the following:
    1. Check the units on answers
    2. Look for answers that make no physical sense
    3. Think in really handwavy ways
    4. Look at the limiting cases of the answers.

    In particular and most importantly I’ve learned what I call rule 5:
    5. Solving the problem should be a last resort, after all the other “low hanging fruit” has been reaped.

    From which follows corollary 1:
    C1: Under no circumstances should learning more physics be the way to approach studying for the GRE.

    Granted, I’m graduating from a top ten physics department so my physics preparation should “presumably” be ok to begin with, but it’s not clear to me that this test really rewards knowing physics fundamentals.

    Ok, I’ll grant you: all other things being equal, having a better grasp of physics will probably cause you to do better on the GRE. But if you want to improve your score in a short timescale, it seems like learning the test and the tricks the test gods throw at you is more important. Or maybe the test is really supposed to be about cleverness, and not physics at all (cleverness is certainly a good trait for a physicist!).

  • eagle1879

    Thank you very much for your post about physics GRE. I’ve learned a lot from your post. Basic concept of physics is really important. And I think it must be difficult to choose the problems. Just as Ned saids. You have to avoid such condition.
    However, I am looking forward to a PhD of Astrophysics. And your post gives me many useful advice.

  • John

    I wonder how Newton, Feynman, Einstein, Watt, Rumford and others would have scored on the GRE.

  • Pingback: Ted Bunn’s Blog » Blog Archive » How to ace the physics GRE()

  • Peter Coles

    Just a short UK perspective. Thankfully, we don’t have the GRE or anything like it. Selection of graduate students is done via a written application (including details of grades, etc, as well as a statement of research interests), formal references from people involved in the undergraduate career (tutor, project supervisor, etc) and, most importantly of all, an interview. Of course, we’re just a small island so interviews are relatively easy to arrange. I doubt if it would work in the USA.

  • Dorothy

    Thank you John for the view from the inside, I find it fascinating. I realize that there’s no perfect way to select graduate students, and that this test should be one factor among many for a panel to evaluate. My own experience of the physics GRE was not a happy one: coming from a liberal arts college where no one I knew had taken it before, I was woefully ill-prepared (or rather, well prepared in precisely the wrong areas). Your advice would have been very helpful, as I’m sure it will be to the students preparing to take the test now.

    As it was, I scored worse in the physics GRE than I had ever done (or have done since) on any other test in my life, by some margin. The discrepancy between my physics score and general scores was almost comical — but in truth it was devastating. I have no idea if the top-tier US graduate schools I intended to apply to would have rejected me outright on that basis, or if the rest of what I considered a very strong application package would have at least made them take a second look. As it was, I was accepted into A Very Prestigious UK physics program that did not require my GRE scores before I even had to apply to the US schools, so I “luckily” removed myself from that equation altogether.

    I don’t want to be overly dramatic, but the damage done to my confidence because of that result had a real detrimental effect on my experience at graduate school and my perception of whether I deserved to be there. To be honest that feeling still resonates today despite a successful research career. I took a lot of solace from the comments by Harvard’s Howard Georgi mentioned by Jennifer West above, which came out about the same time I received my test scores (this article: ). All the same, I wish I’d never taken the test at all.

  • Insomniac Coward

    To An Ominous Coward:

    Far be it for me to defend the GRE, but I have to point out that the four things you mention:

    1. Check the units on answers
    2. Look for answers that make no physical sense (I presume you mean “Eliminate …”)
    3. Think in really handwavy ways
    4. Look at the limiting cases of the answers.

    are all absolutely, positively essential tools in research. In fact I think I use each of them every single day in my research and certainly far, far more often than I ever need to remember, say, the definition of Reynolds number or how to calculate the anomalous magnetic moment of the electron.

  • Ted Bunn

    For those who think that complaints about the GRE are just sour grapes: I kicked the Physics GRE’s ass when I took it, and I still think that most of the criticisms of it are right on. It rewards speed and having learned its particular test-taking tricks, neither of which is predictive of very much useful.

    But none of that is John’s fault, and I’m very grateful to him for passing on this advice, which I will certainly pass on to my students.

  • Adrian Soler

    For people like me taking it this next nov. this article is of great help, is true a single test won’t put a permanent positive or negative mark in your career but it adds to the general perception a grad school will make of you. I do agree for the GRE you don’t need to know more than basic physic concepts but examination skills instead.

  • fh

    John, you asked about Europe, I can tell you about my experience which, being 4 years old and the European systems being in an enormous state of flux might be outdated already.

    In Germany after you leave high school there are no standardized tests anymore. During the old Diploma system you would have 2 years of basic lectures (including a very solid foundation in mathematics on the same level as a basic maths degree), 2 years of specialized lectures and a 1 year research project involving fairly independent work (no more lectures required) more akin to a first year PhD student in the anglo saxon system, which should result in a Diploma thesis.

    How the final grade is determined in the Diploma system varies from university to University, typical is that you have 4 oral exams which are about hour long 1 on 1 sessions with a professor. The topics are Theoretical Physics, Experimental Physics, Physics Specialization, Minor Specialization (typically Maths for theorists). The grades from these 4 sessions are taken together with the grade on the Diploma thesis. As there is a serious amount of grade inflation the final mark is a rough indicator, and its meaning does vary from University to University, but averaged data is available. As our Universities are far less segregated into tiers of quality, comparability is still given.

    As people have research experience when applying for PhDs they will tend to apply more focused for doing a PhD with a certain Professor, there is no notion of general graduate school. A Professor will have a number of applicants and decide based on whatever fancy strikes him/her whom to take. Usually of course a combination of factors: Recommendation letters, previous research work, personal interview, etc…

    I am not sure how the system has changed under the Bologna process.

    Obviously our system tests a very different set of skills, in particular being capable of personally conversing and discussing fluently and convincingly about specific topics in physics. This certainly played to my personal strengths.

    The GRE was one of the reasons I didn’t apply to any North American grad schools. It seemed unnecessary and irrelevant chicane to me, and the people I wanted to work with were in Europe anyways. That said I don’t think it would have been an impediment for me had I seriously wanted to apply.

  • onymous

    Ominous Coward gets the way to prepare for the GRE exactly right. My score improved by ~200 points between a practice test and the real thing, with minimal studying, just by taking those lessons to heart. John’s advice is also good; looking over the end-of-chapter formula summaries in a first-year textbook the evening before the test probably got me 10 questions that I would have otherwise had to spend too much time deriving a result to answer.

    The test is annoying and has little obvious relationship to skill in research, but there is something to be said for the lesson of always thinking of as many simple checks like “does this formula make sense when I send this variable to 0 or infinity? what about the limit where objects A and B have equal mass?” as possible when looking at the result of a calculation.

  • Sean

    John, I hope that my comment didn’t contribute to your sighing — although I can see how it would. I should add that, despite my deep skepticism about what the GRE actually measures, it is and will continue to be a significant component of the admissions process, so making the test as fair and useful as possible is an incredibly important job. If a committee is faced with two applicants with identical grades/letters/experience, they’re always going to choose the one with higher GRE scores — and why wouldn’t they? (The trick is comparing students with good GRE’s to those with good something else.)

    Especially for students from smaller or less-recognized schools, working hard at acing the GRE is probably the highest-impact activity you can devote yourself to if you’re a college junior/senior who wants to get into a good graduate school.

  • John

    Okay now to reveal a bit of a secret…not all grad physics programs require GRE scores, or GRE subject scores. When I am reviewing applications and there are no GRE Physics scores, I look at the verbal, since most physics students score very high on the analytical and quantitative parts anyway. But in the absence of a GRE Physics score, the student’s statement and letters and their transcript have a good deal more weight.

    So if you really hate the GRE for any or all of the reasons discussed above, and don’t necessarily want to get into a program that requires it, and has a high threshold, then you might do better not to send your scores or take the exam at all. (I can’t believe I am saying this.)

    One other piece of advice though, about the personal statement. Keep it focused on the physics that you are most interested in, and stay away from a personal biography, or grand observations about the majesty of the universe, etc. If you are not sure of what field you want to pursue, name a couple anyway, and mention possibly doing experimental physics (trust me on this one). Good luck!

  • max

    I really don’t understand. If we all hate PGRE (and that seems to be the consensus), then why don’t we replace it with something better?

    My problem isn’t so much with the test itself, although I do largely agree with the above negative posts, but with the business practices of ETS. They charge exorbitant amounts to actually take the test, more to send scores (which I’m sure have a marginal cost of pennies for them), and then they do things like make the late registration deadline for the November test the day BEFORE the October test so that students have to pay double to insure against some calamity on the first test day. The fact that they get away with calling themselves a non-profit is vaguely sickening.

  • jackal

    In my graduating physics class (as an undergrad) at a top-5 institution it was nearly universally agreed that the physics GRE was a scam to pad the coffers of the ETS. Our hostility to it was based, in addition to all the criticisms highlighted above, to all the random fees and our being assigned to random locations requiring 2 hour drives; and being charged to change location. I forget the exact details now, but I remember some 20 of us writing and signing a sternly worded, angry letter to the ETS over the whole fiasco.

    Beyond that, most of us hadn’t answered questions like those on the GRE since high school (multiple choice, etc), and even though we’d all obviously done well on these tests back in high school, through college we’d been trained to answer problems in depth and often more complex exams that were take-homes. I mean, the premeds taking the premed physics class had far more practice doing this than us actual physics majors, grr. So really, we thought it a giant waste of our time (and our departmental advisors mostly agreed..).

    As for correlations, I’d say it mostly had to do with those that had less stressful/ less research-heavy senior years and had some time to prepare. I did fine, and got into a top program that was my first-choice. A couple of smart friends didn’t prepare much, and did pretty badly, but got into the top programs of their choice anyway and are rocking research-wise now.

    We were all of course thrilled to donate money to ETS for the privilege… In retrospect it is amusing how annoyed we were over the whole location/fee disaster (not to mention the test). I guess the exam survives since as grad students we quickly forgot about it all..

  • yowza

    lol @ this article…..

    the physics gre is a pathetic joke and the ETS company should fucking burn.

    go back to the way it was in the 50’s BEFORE this piece of shit existed, plenty of fine scientists were generated without the help of the physics gre.

    also, this comment by the author:

    “As for equivalent answers, we do try to make sure there are five logically distinct choices; rarely it may happen that we missed the fact that there are two or three mutually exclusive choices out of the five, in which case one must be correct.”

    is complete bullshit, answers are grouped in either sets of 2 or 3, and you can easily eliminate one set if you’re using your brain.

    i scored a 900 on my april test.

    all of you losers who think more and more people need to go through this should jump off a bridge and rid the world of your ignorance, we need MORE scientists, and this test IS NOT helping gather them. some of the most brilliant scientists and researchers are people who cannot perform well on these exams.


  • Claire C Smith

    Was about to lay off commenting for while, real life takes over every now and again (I love Scov Mag) but will return. So, I am just halfway through reading these comments and will finish reading them after I have typed this out. Wanted to comment about this because it’s so interesting:

    This post and the comments here rung a bell. First, I am not a scientist in my day job, far from it! I love physics in my spare time (art too)- I think physics is hard but very<- interesting. Anyway, I have studied physics at college in the UK at night school (did do it early secondary but didn't take main exams later) bit did take exams when at college. Know a bit about it – you know, gravity, electromagnetism. So, while I was at college studying physics, I realised something very important. That doing the homework took me about three times as long as everyone else. The maths (mental arithmetic) in the physics was harder than the physics, which I loved (the Physics). I found this out by asking students regulary how long it took them to do theirs. Answers – mostly sooner then me. I scored less on tests (mocks) but scored around a good 85% from project work and homework quite often. But just before I return to this, going back to an article on this mag a few weeks ago, it was about Roger Penrose. This article struck another bell/nerve. It related to this topic. What I didn't know and what it said in the article, was that he actually struggled with mental arithmetic before becoming who he is today – his hard studying of mental arithmetic often went into the next sessions, yet if you see, he turned out to be a really good scientist. They even say he is a mathematical physicist! Of course this isn't something that happens very often, but it makes you think.

    The education system in the UK has a strategy that relies upon measuring (quite alot) a specific type of learning – in which the data from it comes from students taking exams. This is the best way to do this , it's correct. The problem is that, although learning a subject like Physics means including taking tests to measure test competence – which it rightly should, it is that actual learning system that is also extracted and used from just about any other subject, that makes it appear that the learner is the doing the subject, when really it might not be that simple. Eaxample, any subject that requires a question to answer module for a specfici thinking method, without considering what each part may represent, shows up as a good academic measure. In this sense Physics becomes a vehicle in which learning is measured (like the others subjects) This is good but you ask the question, are the students then actually learning Physics or learning how to pass tests. Rote learning and memory recall is obviously important but it misses stuff. The problem is, you can't measure curiousness or wonder! or inquitiveness, so it is deemed less profitable in a data set for an institutional setting, which is fair enough if you think about it, but, it makes you think what the real motives of Education really are. At the end of the day it's about balance – a mixture. Those students who are slow but are good thinkers (as mentioned in a comment here) can become strangled out, but what was it thay say about life?

    It's tough! LoL!


  • Haelfix

    The trick to getting a good GRE score is to have some variety. There are a ton of super easy questions that are like 5 seconds (you either know the definition or answer or you don’t). The catch is you will have like 2 of those per subject. If you answer all of them correctly, and simply grind your way through the rest, you are already in the 50%ile range.

    So if you only know quantum mechanics but forgot your optics in highschool, you will get the gimme ‘fermions are antisymmetric, bosons are symmetric’, but miss the ‘the image is real and inverted’ question. That basically buys you the extra time needed to actually calculate things (b/c 2 mins per question is impossible for all but the smartest human calculators)

    The ridiculous part of the exam is cracking the last 5-10%ile range. The first practise test I did I was roughly in that regime, and to do better required about a month of pure memorization and doing Halliday and Resnick problems such that I knew the answers/formula for most of the questions by heart. That truly sucks, and was a ridiculous waste of time, even though it did allow me to break into the elite scorers and get into the good grad schools.

    Overall, I don’t buy the veracity of the test at all. The competency exam is far better and more of a real measure of physics skill.

  • Schwa

    Yowza, what you’re saying only makes sense if the factor limiting the number of scientists trained is how many can get into graduate school at all, not how many spots there are in graduate schools. I don’t know many professors who would rather have no students than students who didn’t do well on a standardized test, and when you look at programs which do have trouble filling their ranks with qualified applicants (e.g. art history graduate programs), the problem is generally not the test scores. In short, your complaint would make sense if we lived in a world where training resources (funding, faculty, facilities, etc.) were more plentiful than the students wanting to enter graduate programs.

    Meanwhile, when we look at the relevant shortages of trained personnel, the gigantic hole isn’t experimental physicists, but high school physics teachers and rad techs. Again, I’d be super surprised if the main barrier here was the physics GRE, which isn’t even necessary at many schools. It’s like arguing that the NCLEX is keeping people out of nursing.

  • Schwa

    Oops, meant to say the NET. NCLEX is the licensure exam.

  • yowza


    you are wrong in so many words…..

    read here:

    why do you think there arent enough jobs for these people? its probably because the people who would be supporting them have been too busy giving a reach around to inflate the dollar. i agree on teachers and medical techs, whatever, great……but the fact is that we need more scientists.

  • Arun

    If there is a shortage of scientists, then scientists’ wages should be rising faster than inflation – at least, as per my understanding of economics.

    The collaborators on this blog can perhaps poll themselves whether their incomes are rising rapidly.

  • Jimbo

    “I come to BURY the GRE, not to praise it (or `improve’ it) !” – Apologies to Mark Antony

    May the GRE adv. physics test, and GRE gen. tests fade into antiquity, for like the SAT, LSAT, ad nauseam, these attempts at SCREENING people’s futures have caused much more harm than good. Not to get off topic, I will stick to the GRE adv.

    My primary complaint is that this `measure’ of a student’s knowledge of physics comes complete with the same mindset as do most grad prelims & qualifiers: Life OR Death, at the mercy of a STOPWATCH ! Enough stress to stymie many a bright person into utter stupidity, all in a format which is totally antithetical to the manner in which REAL world physics gets done: weeks, months, years: Not Seconds !

    The results beget a ludicrous intellectual stigma, which follows the person on thru grad school, and into teaching. Why must only theorists be allowed to teach the graduate core courses, and experimentalists relegated to undergrad courses ? No doubt, some chairs conclude, because of `abilities’. Where do they get this impression from ? A trail of numbers….Still, one would think that theorists could uniformly & accurately convey the essence of graduate-level physics such that there would be no need for the apply a testing filter, only a passing grade would be required. Indeed, some universities do !

    This absurd testing format has been completely dispensed with by two prestigous physics depts., at Colorado & Rice, who have abolished their written PhD exams. One would hope they would precipitate a snowball effect, and others would follow their lead, but there is an unwillingness to capsize an already leaking & sinking ship, commandeered by a peverse crew & captain. Someone has an ulterior motive for sustaining this inequitable system on life support, and I’ll bet that none of them are students.

  • Kaleberg

    “If there is a shortage of scientists, then scientists’ wages should be rising faster than inflation – at least, as per my understanding of economics.” – That kind of thinking reflects our current surfeit of economists. The problem is that scientists are absolutely essential, but employing a scientist doesn’t mean you will ever see a dime, even if he or she comes up with something really important. That’s why smart kings had a court astronomer.


    Way back, when I shared a house with way too many physics graduate students, they were all sweating over their PhD orals. Much of that sweating was done over the open pages of their freshman physics texts because the orals could cover anything from basic mechanics on up to whatever world shaking thing they were doing as their theses. I asked about this and was told that this was because their professors wanted them to demonstrate that they could think like physicists. Think Enrico Fermi with his slips of paper. That meant getting the units right, making sure things made physical sense, waving ones hands to indicate ellipsis, exploiting the limiting cases and using one’s knowledge of physics to take a shorter path to a solution.

    As the xray astronomer explained, when push comes to shove, it all comes down to F=ma. (He was the wittiest.) The IR spectroscoper agreed; it was all about conservation of energy, er, mass-energy. For the fusion physicist it was all about no magnetic monopoles and yes charge density. (As I said, way too many physicists.) All the rest was just icing on the cake.

  • Schwa

    Yowza, did you even read the article you’re linking? The fact that we have a pile of engineers who are calling *themselves* overqualified for currently available jobs suggests that the problem is not that we’re keeping people out of graduate school. Moreover, even if it were true that companies and universities were slavering over anyone with a PhD in a hard science, that wouldn’t change the fact that there’s not enough funding or professors to double (or whatever) the number of PhD graduates without either providing inadequate mentoring or making professors’ lives hell. Unless you solve THAT bottleneck, any filter that gets quality students in (even if it keeps other quality students out) isn’t harming the quality of PhDs getting trained.

    To show that the physics GRE is not adequate in our current situation, you either need to show that there are significant numbers of people getting accepted to physics graduate programs on the basis of strong GRE performance who turn out to be professional failures, or you need to show a compelling reason that (usually taxpayer funded) physics programs should care more about some deontological conception of fairness than keeping a system which helps them train competent physicists.

    When there are more acceptable applicants than places to accept them, what will determine the quality of graduates is how effectively your application process weeds out the unqualified, even if it accidentally rejects some acceptable applicants. Going to graduate school is a privilege, not a right.

  • I.P. Freeley

    Blog post title: “An Inside Look at the Physics GRE”
    Me: And I thought it looked bad from the outside.

  • ian

    It sounds like the European and pre-1960’s method of grad school admissions works fine and makes more sense. So why is it worthwhile to give ETS millions of dollars each year?

  • David Nataf

    Some advice to anyone taking the test:

    Serway Beichner Jewett “Physics for Scientists and Engineers” is an excellent book on Freshman physics. I highly recommend going through the quick quizes in the chapters, each yields a significant qualitative insight.

    Also, a perfect score is 85 correct answers and 15 blanks. If you’re studying and you find yourself not knowing 2 topics, you only need to learn 1. Except classical mechanics, that one will yield a lot more than 15 questions.

  • Jordan

    I didn’t mean to sound critical of you, John- just of the test. Thank you for the advice; it is accurate and appreciated.

  • Charon

    Useful advice, from my perspective as someone who took the physics GRE several years ago. I wish I had had such clear advice.

    As pointed out above, many physics majors (including myself) never took a Halliday & Resnick or Giancoli type course. We therefore had _never_ solved problems anything like those on the GRE, even in first-year physics. We were warned of this, as we approached taking the test, and told horror stories like that of the student a few years ahead of us who was an amazing student in research and classes (and this was at a very good school), but was rejected by Berkeley because of his GRE score. Well, he got into MIT though, so it wasn’t a big loss.

  • Yvette

    I once met David Griffiths of textbook fame who says he used to be on the Physics GRE committee, probably the same one John is on. He said something on the lines of how he went in excitedly thinking he was going to change the test for the better but left a bit bitter about the whole process, in short. He also had a story about how one year they brought in a student who had gotten all 100 problems correct, excitedly asking him how he’d learned all that physics, but the student was confused because all he’d done was apply tricks like taking the limit and checking units etc. Yay?

    I will note at my undergraduate institution the majority of time students spend studying for the GRE is sitting down with a flashcard deck to memorize about 150 equations, which is the most logical thing to do if you’ve never been required to actually memorize the equations before. Put it this way, how many physicists could recite right NOW the differences between the moment of inertia between a disc, ring, sphere, hinge…

  • Mary

    I will fully admit not being prepared the first time I took the physics GRE– all I had done was take 4 practice tests, which I scored fairly well on. Then came the exam and I have no idea what happened, but it was BRUTAL and definitely scarring. I think my score was the reason I was waitlisted at as many schools as I was. In the end, I wound up at a program that wasn’t suited for me (due to aforementioned waitlisting).

    While there, my physics abilities VASTLY improved. Things somehow clicked. I could solve problems better & faster than before. But I didn’t like my program. So with a better grasp of physics & more time to study, I tried to tackle the GRE again.

    Though not quite as brutal, the exam did not go well. And I still don’t understand why. I was sure my skills had improved at least 4 fold (this is not an exaggeration, I’m underestimating here), and my scores improved and alotted me a meager 8 percentile over my last score.

    Oh, and after all that, ETS managed to send out all of my old scores to my new schools (which I only found out about in March after having been waitlisted AGAIN). After about 10 hours of phone calls later, they agreed to send out the new scores free of charge, though it wasn’t a mistake on their parts. GAH.

    I thank you, though, John, for your advice. Wish I had a resource like this (and the comments) when I took the exam years ago.

  • John

    David Griffiths was indeed on the committee, but we did not overlap.

    You cannot do all the problems on the GRE by doing limiting-behavior checking and units and so forth, though some can be narrowed that way. You really need to know some basic undergraduate physics. Memorizing some equations is definitely helpful.

    A perfect score is most definitely *not* 85 questions. It varies from exam to exam.

    ian: ETS actually loses money on the subject tests. And, with about 4500 students per year taking the exam, you can quickly calculate that it’s far from millions of dollars.

  • Carl Brannen

    This opened my eyes. About 30 years ago I took the physics GRE and ended up at the 84% level. I remember it still because that was the worst score I ever got on a standardized test. I had no physics degree, or even physics minor, but I was nearing the end of an MS in mathematics and thought I should do much better.

    The worst subjects for me were electricity and magnetism, thermodynamics, and optics as I’d had graduate classes in some subjects but no undergraduate classes other than freshman physics and quantum mechanics. And the thought of memorizing moments of inertia never occurred to me, it’s just a matter of integrating quickly. But if your mind is better at memorizing than calculating, it’s definitely a way to get by.

  • Jerry Vinokurov

    I took the physics GRE and did ok, well enough to get into grad school, anyway. I’m glad so many commenters before me have pointed out what is so wrong about this test that I don’t need to do so again, but to summarize: putting people on the clock and making them solve 100 simple problems isn’t really measuring anything like physics knowledge. It measures basic understanding, sure, but beyond that it also measures how fast you can speed through the problems using various tricks. I find it pretty offensive that, after many years of doing just fine in physics classes at a respectable institution with a minimum of grade inflation I had to completely reverse my way of thinking in order to do well on this test.

    I don’t understand why the physics subject test cannot look like the prelim I took in grad school. We had 5 hours to solve 5 problems, one from each major area of physics (stat mech, classical mech, E&M, and 2 QM problems). I think a test similar to that would be a far more accurate reflection of one’s ability to do physics than being put on the clock and having to worry about bubbling in the wrong bubble or whatever.

  • John

    Jerry, how much would it cost to individually grade five written problems for about 4500 students? How do you standardize the grading process? How long would it take? With so few problems, the statistical fluctuations due to the fact that a student either knows or does not know that particular thing (like, say, displacement current in a capacitor, or orbital mechanics) would mean that the exam is hit or miss, with their whole potential graduate career riding on it. It’s not a sensible approach to the problem, sorry.

    We really, truly, honestly go out of our way to avoid problems being solved just by “tricks” that have no relevance to doing physics or being a physicist.

    And, by the way, we give moments of inertia of common objects when needed, unless we are testing the ability to so a simple integration to find on. So, GRE students, please don’t memorize that, but do memorize the formula for the integral to obtain the moment of inertia!

    Actually there is a fair bit of mis-information in the comments above, and I fear that students will believe it…my hope was to help students prepare effectively for the exam.

  • Jerry Vinokurov

    John, obviously there are problems with my proposal. I’m not wedded to it by any means, but it’s something that sure works at the level of the department, where it’s done every year. Is it a hard thing to implement? Yes, absolutely. However, the alternative, in my view, is no better; it’s a test that basically measures how fast you can do a small subset of trivial problems without making computational mistakes. The GRE doesn’t care that you made a trivial arithmetic mistake, it treats that equivalently to a complete failure to understand the question.

    By the way, this is something that already exists at the high school level: namely, the Physics AP test. It has both a multiple-choice section and a free-response problem section, and many, many more high school students take that test than take the physics GRE. So if the AP system manages to make it work on a much larger scale, I don’t see why we couldn’t do the same thing as well.

    Also, I want to address your point about statistical fluctuations. First of all, I think it’s rather easy to identify the core things that students should take away from classes in any given area of physics. In classical mechanics, that should be something like the ability to write down a (possibly complicated) Lagrangian and then find and solve the associated equations of motion. In statistical mechanics, it would be the ability to construct a partition function and solve problems based on that. In quantum mechanics, it’s variants of the Schrodinger equation and maybe some perturbation theory. None of those require one to know anything about displacement currents in capacitors or the intricacies of orbital mechanics, and indeed, I think there would be little point to my suggestion if all it did was to encourage memorization of some other set of equations. Second, it seems to me that you are saying that having fewer problems would provide less of a discriminator between students, but I don’t know that this is the case. Certainly, tests in my undergraduate classes seemed to do a pretty good job of discriminating between students, and they were all based on solving problems. Anyway, that seems like a peripheral question to me; what we should be concerned with is whether students understand physics, not whether they can do trivial computations really fast.

  • John

    You make an excellent case, Jerry, I must say. The AP Physics exam does indeed manage to do it somehow…it would be a better exam than the GRE. Perhaps a mix of the two approaches, as found on the AP Physics exam, is ultimately the best approach. I can’t really say how much more expensive it would make the exam, but it would be more expensive, I am pretty sure, to grade it. Maybe it would be worth it.

  • Charles

    Hey thanks for the great article! I was wondering what the author and commenters recommend to be the best undergraduate book to use in preparing for the Physics GRE.

    Some popular titles are:
    halliday resnick walker—Fundamentals of physics
    halliday resnick krane–Physics
    young freedman–University Physics
    Serway Jewett –Physics for Scientists and Engineers.

  • Brian Beverly


    I have a physics/math BA. Do you think the measure of a physicist is their ability to solve a physics problem in 1.7 minutes? Is that why all of my physics tests were two hours long and only had four questions? What about creativity and a passion for the subject? I love physics, but grad schools need to get real. They want stellar GPAs, research experience, letters of recommendation and then the entrance exam is the complete opposite of any ‘normal’ physics test. Sure you have a fancy metric but the philosophy is rotten to the core. Everyone hates this exam for the same reason because it is a waste of fucxing time. The science is slowly dying and hopefully the admissions process for physics graduate programs is overhauled before it goes extinct.

    Graduate schools have turned into legalized indentured servitude; although students do get to experience the joy of paying off their undergraduate loans on a pitiful stipend. I’m not buying into the GRE propaganda nor should should anyone have to pay for the privilege of taking these exams. Do you have a PhD in education? Is there only one type of physics student or does the GRE create a group think monoculture? If you want me to go to graduate school then beg. I’m happier studying physics as an independent autodidact.

    Brian Beverly

  • Grant Entwistle

    So I took the physics GRE’s November 7, and I thought this advice was very sound. I focused on reviewing the basics, and I think it may have paid off :)
    Thanks for the article!

  • Rena

    Jennifer West—I believe the Howard Georgi article you referred to was in the Status of Women in Astronomy newsletter, January 2000, available at His comments on the GRE may be of interest to others here.


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Cosmic Variance

Random samplings from a universe of ideas.

See More

Collapse bottom bar