1. Not finding help here? Sign up for a free 30min tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Courses Physics Graduate School Without Physics Courses

  1. Nov 25, 2011 #1
    Long story short, I am currently a junior mathematics major, and I plan on going to graduate school for physics. My college offers but two, non-calculus based physics courses covering classical mechanics, electricity and magnetism.

    Correct me if I'm wrong (please!), but I believe that if I do well on the GRE and GRE Physics subject test, I will be able to get into a fairly decent graduate program (top tier would be nice) without having majored in physics.

    I believe that if I studied independently, a few hours every day, that I will have a fairly decent grasp on the material before I take the exam(s).

    In essence, my question is this: What books/sources do you recommend I use?

    Below is a direct quote from ETS about the subjects covered in the GRE physics test. I would like, at the very least, to know these materials thoroughly, but am very open to any additional materials you should recommend.

    Thank you!
    Alex

     
  2. jcsd
  3. Nov 25, 2011 #2

    Pengwuino

    User Avatar
    Gold Member

    After those 2 classes, you'd be lucky if you were able to answer a single question on the Physics GRE unless studying "a few hours a day" is followed by "for about 2-4 years".

    No graduate school is going to accept you if you've never taken a real physics course, have a near 0% Physics GRE score (which you will), no letters of recommendation from physics instructors (which you probably wouldn't be able to get), and no physics research experience.
     
  4. Nov 25, 2011 #3

    jtbell

    User Avatar

    Staff: Mentor

    I suggest you read this recent thread which was started by someone who wants to do something similar, skipping an undergraduate degree completely:

    https://www.physicsforums.com/showthread.php?t=546911
     
  5. Nov 25, 2011 #4
    ^This. Studying "a few hours a day" is a pretty normal pace for someone actually pursing a bachelor's in physics (including lectures and homework), and that usually takes at least 3 years.
     
  6. Nov 25, 2011 #5

    Choppy

    User Avatar
    Science Advisor
    Education Advisor

    If you really plan on going to graduate school for physics, why not go to a school that has a good undergraduate physics program? It seems like you're trying to do this the hard way and in doing so are setting yourself up for disappointment.
     
  7. Nov 25, 2011 #6

    micromass

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor
    2016 Award

    OK, what I don't get is that you plan to go to grad school in physics, but you don't actually major in physics... Or you don't plan to actually take physics courses...
    Most people spend four years getting a physics degree. What do you think they do during that time, nothing?? You plan on taking two courses and doing as well as people who are putting in four years of physics. This does not compute...
     
  8. Nov 25, 2011 #7
    @Penguino, bcbwilla

    Perhaps I should rephrase. By a few hours a day I mean, potentially, 5-6 hours a day, Sunday-Saturday, throughout the entire year. I am already fairly confident in my classical mechanics, and am currently studying electromagnetism. That should save at least a little bit of time.

    It should also be noted that I am a 4.0 student, and that I am used to fast-paced learning. I learned Pre-Cal, Calculus I, II, and III in roughly 1.5 months.


    @Choppy
    I was accepted to UIUC as a transfer student last year, but was unable to attend due to unexpected medical bills in the family. I have considered finishing my degree in mathematics and then taking some undergrad physics courses at another institute before applying, but I would much rather save the time and teach myself.


    @jtbell
    Thanks ;)
     
  9. Nov 25, 2011 #8
    Yes, you are wrong.

    No, you won't

    The material I recommend is an undergraduate degree in physics. The GRE is not a walk in the park, and even majoring in both physics and math, I can tell you that they ask you questions that you would have never studied before as a physics major (unless you are R. P. Feynman). I am sorry, but you kind of offend me thinking that a couple hours a day will be equivalent to all the heart wrenching, head banging years I spent to just get my degree, let alone getting into grad school. Unless you apply oversees where they don't take GRE scores, you will still need a physics professor write you a letter of rec (like was stated earlier).

    I also kind of find it weird how people think "hey, I'm really really good at math, physics should be easy" because its a completely different skill. You have to build a physics intuition from the ground up. I didn't go to math grad school because I am horrible at proofs. Its a completely different skill, and most "proofs" in physics would give mathematicians heart attacks, and most proofs in math seem useless in physics when you can "show" its true without a proof.

    I don't mean to burst your bubble, but it will take way more than a few hours a day for a year.
     
  10. Nov 25, 2011 #9
    I don't think it's the actual degree that is the big deal. The issue is recommendation letters. Also, it helps to take classes and be able to say you got an A in such and such class and 3.9 physics gpa, etc.

    I was planning on doing something similar, but I had planned to take more physics classes. I ended up going for math, instead. I am still trying to learn physics on the side because it's close to the math that I am doing, but grad school in math didn't leave much time for that. So, I think maybe I'm still about 3 classes short of having the equivalent to an undergraduate degree in physics after about 7 years of studying physics in my spare time (although I know bits and pieces of some graduate level topics in physics). So, as yet, I have been unable to do both, to the extent that I had intended.


    I don't think this is quite true. It's a different skill, but not completely different. There can be a lot of overlap between physics and math, depending on what you do. As I see it, there's no sharp division between math and physics. It's more of a continuum with a lot of variation within each field. Theoretical physicists are closer to mathematicians, for example, and mathematical physicists ARE mathematicians who study physics or problems motivated by physics. And there are people like string theorists, who, I think, are basically non-rigorous mathematicians.

    Of course, part of the problem is that physicists and mathematicians don't talk to each other as much as they should.

    Math majors usually make good physics majors. The trouble is that it takes a lot of additional effort to learn physics, not that they aren't good at it. In some cases, people find they are just better at one than the other, though.
     
  11. Nov 25, 2011 #10

    Vanadium 50

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    A BS in physics is ~40 hours of work a week for 32 weeks a year for 4 years: a little over 5000 hours.

    5 hours a day for 7 days a week for 52 weeks is 1820 hours. So you're arguing that without instruction, you can still learn the material 3x faster than a typical person who goes on to get a PhD in physics.

    While I won't say this is impossible, I sure wouldn't bank on it. Just as I wouldn't bank on winning the lottery as a financial plan.

    If you want to go to grad school in physics, you need to take physics classes as an undergrad.
     
  12. Nov 25, 2011 #11

    Pengwuino

    User Avatar
    Gold Member

    Well now you're being half-way realistic. Yes, you're looking on the order of years (plural) in terms of studying. One thing that must be pointed out, however, is that I have NEVER EVER EVER EVER heard of anyone being able to do this. I think most people will say that most of their real physics understanding doesn't come from just going through textbooks, but through interacting with professors, other students, and doing experiments as well (oh by the way, do you plan on doing purely theoretical physics?). I mean, getting an answer from solving a problem is nice, but over the years I found myself constantly asking others and professors "what the heck does this mean?"

    There is a fantastic way of figuring out whether or not you stand a chance. Go to ETSs website, find the practice physics GRE tests, and just have at it (Actually, here is the link: http://www.ets.org/Media/Tests/GRE/pdf/Physics.pdf [Broken] ). If your confident in your classical mechanics, you should find the classical mechanics problems fairly easy (and by easy, finish each within 60 seconds). This it not a joke either, you have about 1-2min per question on the physics GRE to do decent. You can also take the entire test, under test conditions (180 minutes I believe?). They give you the answers, you score yourself, and see how well you do. If you got a 50% or so given their percentile calculations for that practice test, great! If you get single digits, you are not even close to prepared.

    That's really not relevant. Were you formally tested on all of those classes?
     
    Last edited by a moderator: May 5, 2017
  13. Nov 25, 2011 #12

    Vanadium 50

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    I took a look at that test - I would say the practice test is substantially easier than the real deal.
     
  14. Nov 25, 2011 #13
    I'm working with one of my professors to get an internship setup. It would be physics research at another school.

    I did, and do. Though I only went through about half of them.

    Yes. I made perfect or nearly perfect scores on all of them.

    Thank you for actually using numbers. I realize it will be extremely, extremely hard to accomplish. I'm hoping that since I already have the mathematics, applying it to concepts will go faster than if I were learning both simultaneously.
     
  15. Nov 25, 2011 #14

    vela

    User Avatar
    Staff Emeritus
    Science Advisor
    Homework Helper
    Education Advisor

    There's a big difference between lower-division and upper-division courses.

    Since you're still in your junior year, why not just add physics as a second major or possibly as a minor? You could probably cram everything in before you graduate. At the very least, you should take classical mechanics, e&m, quantum, and statistical mechanics.
     
  16. Nov 25, 2011 #15

    atyy

    User Avatar
    Science Advisor

    I'm a biologist, but I did try the physics GRE years back (much easier than the Biology GRE). I 've no idea about physics grad school (probably much harder than biology grad school), but if you just want to try taking the test, maybe:

    Basic Mechanics: Kleppner & Kolenkow (the absolute best)
    Fancy Mechanics: Fetter & Walecka (just the first half), supplemented by Landau & Lifgarbagez when Fetter & Walecka are confusing
    E&M: Dugdale
    Waves: Lewin
    Thermo & Stat Mech: Schroeder, supplemented by Kardar when Schroeder is confusing
    Modern Physics: Beiser (just read quickly), supplemented by the special relativity parts of Rindler for concepts
    QM: Griffiths, supplemented by Peebles and Shankar when Griffiths is confusing
     
    Last edited: Nov 25, 2011
  17. Nov 25, 2011 #16
    Dear SecretNile


    I posted a long sequence of what books/knowledge/skills you should have for grad physics and beyond at: https://www.physicsforums.com/showthread.php?t=540829
    At the top there is a short 8 page document for those at about the junior level interested in particles and fields (with a Word document), and then there is a long sequence of posts of a long Word document for going from junior level math/physics to grad physics and beyond (math/physics texts/references/key ideas/key skills). This, later, sequence is the sequence I have edited, thanks to great feedback and have available now. I haven't posted the reduced material yet, but I can email it to you (akalaniz AT gmail.com).

    (Both documents are there in that post, just scroll down.)

    (Note: Part 1 of the long document is on the limits of physics and math. Then Parts 2 & 3 are about the book lists and topics lists.)

    I feel the pain of some the other people who have replied. I got a piece of crap BS in physics from a liberal arts college, and got my butt kicked in grad school UT Austin. I ended up getting a pilot's license and joining the Air Force. Later, at the Air Force Research Laboratory in Albuquerque, I spent 4 years relearning my undergrad physics and undgrad math with the tutelage of a bunch of PhDs. It was mostly me that did self guided learning, but thanks to the PhDs, I had a heads up on what was required in grad school.

    After 4 years of self learning, I left the Air Force and knocked out a 4.00 MS Math, 36 hour non-thesis.

    Then it was back to physics grad school. The math helped, but not as much as you would think. Physics intuition/skills takes a lot of hard work. For example, Griffiths is a typical author of junior level E&M, and Jackson is the gold standard for graduate school. There's not much more in Jackson than in Griffiths when it comes to theory, but applications are far deeper in Jackson. It took me years working at an accelerator after my PhD to begin to appreciate the depth and beauty of Jackson.

    That file I can send you is also about filling the gap between pure (almost useless) math and math that physicists really need but can't get for lack of good sources.

    I hope this is helpful,

    Alex
     
    Last edited: Nov 25, 2011
  18. Nov 25, 2011 #17
    Thank you! That will be incredibly helpful. :)

    Thank you! I will email you shortly.
     
  19. Nov 26, 2011 #18
    I have a suggestion of something that would probably be much more straightforward. If it is theory that you are interested in, you might try to apply to graduate school for mathematics to departments that have a lot of mathematical physicists, or professors that are both physics and math professors, etc. Then you can keep pursuing your math degree, take the math GRE, and then learn physics in graduate school. Furthermore, some departments may let you have someone in the physics department as your adviser.
     
  20. Nov 26, 2011 #19

    lisab

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    It's very possible that should you be accepted to a physics grad program, they will require you to complete a bit of "catch up" work before starting grad-level classes. So taking a few classes at another institution may not be a waste of time. And it would likely help your PGRE scores, too.
     
  21. Nov 26, 2011 #20
    The most important thing is not how much time you spend learning the material. It's the quality of that time.

    At least in my university, if you were a "good" student : going to all the lectures, taking all the notes and doing all the homework, you would be wasting a lot of time with minimal benefit. Only a part of the lectures are useful, only a part of the books are worthy to read, only a part of problems are worthy to be solved. Hence, if you spend your time wisely, if you're good at self learning, I think you could easily self learn all the necessary physics in a year.

    As of textbook recommendations, I would suggest picking books which cover mostly the most important, fundamental topics, without going into details. However, those fundamentals should be explained in a deep fashion. What is more, the books should be written in a formal, mathematical way, with just a few sentences of additional explanation. This would make the books (or, actually, top universities lecture notes might be even better) with relatively small amount of pages, which would give you some extra time, however, that time should be used for slow reading, thinking extensively about all the fundamental concepts. In my opinion, not only for GRE, but for physics in general, it's not how many concepts, how many equations and how many various details you know that matters, but how deeply you understand the very fundamentals. With them you can derive the details when needed yourself.

    For example, for mechanics my suggestion would be :
    http://www.damtp.cam.ac.uk/user/tong/dynamics/clas.pdf
    It seems it's from Cambridge, Mathematical Tripos, hence it should be designed for mathematicians. It also looks clear and short, but at the same time deep and it seems it covers all the most important concepts.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook