1. Limited time only! Sign up for a free 30min personal 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!

GPA for top schools

  1. Dec 20, 2011 #1
    I'm an undergrad math/physics major and an aspiring theorist. I, like many others, would like to go to a top physics school for my post-grad education. I have a couple of questions about admissions to these schools.

    First, purely speaking in terms of GPA, below what GPA would these schools (Stanford, Berkeley, Princeton, Harvard etc) start raising eyebrows? I know that GPA is viewed in the same manner as SATs are viewed in undergrad admissions (a high one doesn't guarantee anything but a low one probably means you're out). What mark do you think this would be? Of course other factors also count for a lot but here I'm taking about the "danger zone" in terms of GPA.

    Secondly, does GPA as a pure number mean a lot or do schools place more of an emphasis on the actual transcript? As in, if the classes bringing down my GPA are say intro E&M and ODE's, would the fact that I did well in upper-level E&M and PDE's count in my favor? Would someone who followed the standard track and ended up with a 4.0 be viewed as a better student than someone who jumped ahead and took higher level classes earlier on, but ended up with, say a 3.8?

    Thanks for answering.
  2. jcsd
  3. Dec 20, 2011 #2
    I read everywhere that above a 3.5 should allow you to find a paid grad-program somewhere. For the top programs, I would imagine that everyone who is accepted is somewhere in the 3.8-4.0 range. From reading "So you want to be a physicist" (located on this forum), it seems as though there is more of an emphasis on research when it comes to getting accepted. I would shoot for a 4.0 obviously, but from my understanding, the GPA is only one part of the picture.
  4. Dec 20, 2011 #3
    the idea is that there are more qualified applicants than there are spaces available, so you need to show your love of research and ability to complete the work required on time.
  5. Dec 20, 2011 #4
    When it comes to research my dilemma is that I'm primarily interested in theory. I'm only a sophomore however, so I can't see how I could be of use to any theorist until sometime in my third year. I could work for an experimentalist (and I have tried talking to some) but the kind of projects I was offered didn't interest me all that much, and if I'm not passionate about what I'm doing, how can I make a decent contribution and impress the professor I'm working for? If/when I do research I want to make an actual contribution rather than just moving tools around for a professor.
  6. Dec 20, 2011 #5
    you literally have to do research in any field, publishing in scientific journals, no matter what you plan to study in graduate school to be accepted at a top school. all you do in grad school is research, so grad schools need see that you are 100% committed to, and capable of scientific research. the top grad schools are able to pick among students who excel at this task while achieving near perfect / perfect scores. you need to be one of these students.
  7. Dec 20, 2011 #6
    I guess I should find a project that I'm interested in, in that case. However, I doubt the classes I intend on taking next semester (QM II, E&M I, Intro to GR, Abstract Algebra and PDEs) will allow a lot of time for research. In such a situation, which should be given preference? Should I lighten my load and focus more or research or should I delay research and continue with that load?

    My original planning was to take a large load of theory classes, so that I'll be competent enough to do some theoretical research sometime in my undergrad career.
  8. Dec 20, 2011 #7
    it's hard to say, but if your grades are good, you might want to make room for research and start cutting out more time for it. the idea is to have at least one publication, even if it's a co-author, as an undergrad.
  9. Dec 21, 2011 #8
  10. Dec 21, 2011 #9
    First of all, physics Ph.D.'s don't work via tiers. You will be doing good if you can get in anywhere. One thing about physics graduate schools is that the standards for the "non-big-name" schools aren't that much different from the big name schools.

    Hard to say because a lot depends on your transcript and what school that you went to.


    Hard classes with a 3.8 beats easy classes with a 4.0.
  11. Dec 21, 2011 #10
    Well if you do real research and you find out that you don't really like it, that's a sign for you to do something else. One really important part of undergraduate research is not some much that it will help you get into graduate school, but it will let you know if you really want to go to graduate school. Conversely one reason that graduate schools like students with research experience is that no one wants a student to find out two years into a graduate program that they really don't like research.

    90% of research involves moving tools around. Most research is boring and frustrating, which is why it's important you get exposure to it early to see if you really like it.
  12. Dec 21, 2011 #11
  13. Dec 21, 2011 #12
    How can the school one attended affect the admissions process? (about the OPs question concerning GPA, and anything else where attending X school would be a horrible or brilliant idea)

    Gold? Wait till you see the Barakley post.
  14. Dec 21, 2011 #13
    a 3.6 at MIT probably will be looked at differently than a 3.6 from State U
  15. Dec 21, 2011 #14

    Vanadium 50

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor
    2017 Award

    False. Yes, research helps, particularly because it allows people to write letters of recommendation with some meat in them. But there are plenty of students, particularly from small colleges, who have done no research but nevertheless get in. As for publications, I think everyone understands that at the undergrad level whether the research leads to a publication or not is most strongly influenced by luck.

    Have you sat on any admissions committees?

    It can cut both ways. If a student is at MIT, and does not participate in research, this will seriously hurt the candidate. To get a degree from MIT without having done some undergraduate research is something you have to actively put effort into. This will cause the committee to wonder two things - why is this person applying to grad school at all, and why did they not avail themselves of all the opportunities they had as an undergrad.
  16. Dec 21, 2011 #15
  17. Dec 23, 2011 #16
    I understand. Thank you.

    This explains a few things I observed on the physicsgre.com 'Applicant Profiles' threads. More specifically, two students from MIT and CalTech respectively, both with GPAs of about ~3.6, who most probably, because of their significant research experience (looks like they were involved in quite a few projects), got accepted into the "big schools".

    It also seems that the PGRE is of high importance, especially when it comes to students coming from overseas institutions. These posts (although they're a somewhat limited data set), lead me to believe that one needs 950+ on the PGRE (and lots of research experience) to have a realistic shot at the top programs within their field. Ironically enough, somebody on there, got rejected everywhere in the USA, except for Stony Brook but got into Oxford. It would seem that it's because of his PGRE which was relatively low and that UK universities don't consider it and that they place a higher value on grades.

    Anyway, this thread can prove to be useful. All this talk about "where one does their undergrad" can get a little worrying, if you ask me...
  18. Dec 26, 2011 #17
    Yes and curiously it would be considered much less impressive. Both MIT and Harvard grade *WAY* less harshly than most public state universities, and Harvard grades less harshly than MIT. MIT also as a "no record" policy in which any classes you fail freshman year are not recorded in the transcript.

    I vaguely remember that the different between MIT and most state schools could be quantified as 0.2 and 0.3 GPA (i.e. you add 0.3 GPA to a state school GPA to get the MIT GPA). I can't find the reference, but that seems right to me.
  19. Dec 26, 2011 #18
    Not sure what that means. Doing undergraduate research at MIT is pretty standard. Frankly, if you aren't going to do undergraduate research, then there is no real point in going to MIT. The teaching quality is uneven, and if you prefer to learn things in a classroom, you can much better off going to a small liberal arts college.

    Now if you are a "drop me in the ocean and I'll see if I can swim" sort of person, then MIT works pretty well.

    And it's looking at the wrong thing. My experience has been that people who worry too much about grades end up burning out to the point that they stop doing physics and so are in no way able to get into physics graduate school. You will get into trouble if you don't worry about grades at all, but my experience with undergraduates is that many of them worry too much about grades, and so burn out.
  20. Dec 27, 2011 #19
    I was going to respond, but Vanadium said what I was going to. The one addition I might make is that it is fine for one's research to not end up solving an original problem or resulting in publication, even at MIT. I know for a fact that at mathematics programs, a candidate without publications can easily be taken in favor of one with publications. The reason is that mathematics is a particularly "slow" field, and jumping the gun and publishing something does not always or even usually indicate superior potential. This is true even at many "top schools", but one cannot bet on anything with those in the first place; nearly everyone, research or not, will have a very tough time with those.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook