Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Wondering about GPA requirements

  1. Dec 14, 2008 #1
    I would like to know how competitive entry into Physics Grad Schools is exactly; I would really like to shoot for a high-end grad-school but this semester has made me worried about things. I took Math 25 and Physics 16 this year (Harvard freshman), and found both classes really difficult; I was adjusting to the competition (kids in my class did IPO, IMO, ect), the "living" on my own, and the huge difference in the difficulty between these courses and Physics/Math I'd done in the past(Crappy public High school in TN + independent readings) and I'm not sure what my GPA will look like this semester... I definitely don't think I'll get A's in either of them, and I've heard it's basically impossible to get into any really good Grad School's without a 4.0... Is this true? Should I be worrying about this or am I just being ridiculous?

    I really had no idea what I was doing when I applied to College; and I don't want to make the same mistake this time... so what is it going to take to have an earnest shot at top grad-schools?
  2. jcsd
  3. Dec 15, 2008 #2


    User Avatar
    Homework Helper
    Gold Member

    That's not remotely true. I don't think you should really worry about graduate admissions in your first year! Enjoy yourself, try taking courses you're interested in, and talk to professors on whether you can be a part of a research group, to experience the type of work that's typically done in grad school.

    The most important thing that grad schools look for in potential candidates is prior research experience. Good recommendation letters are valuable too. Grades and test scores are of secondary importance.

    Don't worry too much, and have fun! Harvard is a great place to study Physics.
  4. Dec 15, 2008 #3
    very few people on this planet have 4.0s, especially in something like math and physics from harvard. id say relax and dont worry about it. yes, improvement is certainly neccessary, but you have a lot of time left for that. you will need to work harder to get up to speed with your peers who have better background. grades are important as hell, and its good that you are aware of it. but as long as your not pulling Ds your okay.
  5. Dec 15, 2008 #4
    It's not just the grades I'm worried about;I'm definitely enjoying the courses; I think my anxiety is just, I'm not understanding things the way I used to (In Calc BC, ect.) I just feel like I'm not having the best of retention. I've absorbed about half of my Math class (It's introductory topology/Analysis (We're using Rudin), theoretical linear algebra (We used Axler, and are now finished with that.) Even when I feel like I understand the problems (maybe 2/3 of the problems on a P-set), I'm still having a really hard time proving certain things; like I don't really know how to do theoretical math; How to approach a complicated proof, ect. It seems very technical and requires all these tricks I never picked up on; I'm not sure how to fill-in the gaps, since everyone seems to think Rudin is a pretty basic text, I don't even know what I need to do to fill in the gaps. I really enjoy the problems, but it takes me forever to do these problem sets because it takes me Days sometimes to figure out how I might be able to do something with proper rigor.

    Physics just skipped a lot of physics I didn't learn (We did like Moment of Inertia Tensor, Motions of Solid/rigid bodies, Lagrangian Mech, Normal Modes/Oscillations, and Special Relativity/the Invariant Interval, ect and this week we did some Legendre/Hamiltonian thing I didn't fully understand because it involved some Math I've never heard of). It's all incredibly interesting but it's just so much content I don't feel like I can wrap my head around it. And that worries me; Because this is supposed to come easily for physicists; it's intro level classes.
  6. Dec 15, 2008 #5
    Seriously? That's a lot for a freshman class. I hadn't done much of that stuff until... well, actually the following year. But still, that's a lot.
  7. Dec 15, 2008 #6
    Yeah, that was what we covered this semester. It's the intro level class we are advised to take if we took High School Physics.
  8. Dec 15, 2008 #7
    Rudin and lagrangian mechanics / waves in freshman?

    These are things typically done in 3rd year. I dont know what the hell Harvard is thinking. It must be on a very shallow level. For someone to go from no-proof highschool to Rudin is just madness.
  9. Dec 15, 2008 #8
    It's sort of weird; We have the following Math classes for Freshman

    Math 1a - Crappy Calculus class for people who don't like Math
    Math 1b - Part II of crappy Calculus class for people who don't like Math
    Math 21a - Computational Multi-var Calc (They use stewart I think)
    Math 21b - Computational Linear Algebra (Not sure)
    Math 23a/b - Mixing Analysis/Linear Algebra, they do 26 proofs and they just memorize them. They don't actually prove anything from scratch.
    Math 25a/b - It's a theoretical proof-based course; We do Analysis and Linear Algebra (From Rudin and Axler) - This is the class I'm in. It was the lowest level proof based freshman class.
    Math 55a/b - Theoretical proof based course - They do Linear Algebra (Axler), Group Theory (Artin), And Real/Complex Analysis. There's like 7 people in the class and it's specifically designed for IMO kids; that sort of thing.

    What I really needed, I think, was a good entirely proof-based course in Calculus; I understand it well conceptually but I'm not used to Mathematical rigor; so I chose the option which seemed to be the "You did well in High-school math and want to learn to write proofs" option... but they start with texts which seem really advanced and require a lot of tricks and stuff to do some of the problems (I have spent six hours tonight doing about 3 problems in Rudin... and that just shouldn't happen)

    And the last of those is an amazing class. I've sat in on some lectures and Curtis McMullen is just amazing; but I'm horribly confused... and it's obvious if this class is this hard, I'd die there.
    Last edited: Dec 15, 2008
  10. Dec 15, 2008 #9


    User Avatar
    Homework Helper
    Gold Member

    For many university courses, it's not uncommon to spend an entire weekend on a single problem sheet. The effort you put in now will be absolutely worth it. Also, the only way you'll really understand something and get familar with it is by doing problems and struggling with them. I can remember my first linear alegbra course, which used the text by Hoffman and Kunze, which was also my first experience with formal proofs. It does seem intimidating initially, but as you spend time working through your problem sets/homework, you'll feel more comfortable.

    Have you tried forming or joining a study group? Discussion and exchanging ideas can help you understand things better and is also enjoyable.

    I'm guessing that's the calculus of variations? If you know calculus, it's not hard to understand the relevant application to Lagrangian mechanics. There are many books which give you a short & practical introduction to the subject.
  11. Dec 15, 2008 #10


    User Avatar
    Homework Helper

    In physics it is important to pick up some of the math on the fly, so much is used that it would take a long time to learn it all in advance. Keep a positive outlook. Math is hard, but also rewarding. This is especially true when making a transition. You have identified a problem and worked to correct it. Which part do you find problematic? Is it understanding, following, or evaluating a proof? Is it inventing proofs on your own? Is it thinking abstractly? The good news is all three areas get easier with practice. It seems your circumstance has you making the transition quickly. It would of course be easier if you had a year or two (say in high school) to get used to proofs and such. People learn differently. For some once they have seen a certain number of proofs it gradually becomes obvious what constitutes a proof and how to construct one. For others it helps to systematically study proofs. I can't recommend a "intro to proofs" type book because most I have seen are garbage, but you might flip through a few and see if they are helpful, I hear tell Harvard has a nice library. Another possible problem is you are learning proofs of new material. It might be worthwhile to revist some material you know well. Either on your own or with a rigorous high school type book, prove some basic things for practice. Do you understand well proofs of things like
    sqrt(11) is not rational (or an integer)
    there is no largest prime
    there is no bijection between Q and R
    sin(x) exist and is unique
    exp(x) is not a (finite) polynomial
    if f is injective f(x) is in f(A) if and ony if x is in A
    if n=ax+by gcd(x,y)|n

    Where at West Pickle Univelicity?
  12. Dec 15, 2008 #11


    User Avatar

    Staff: Mentor

    It's intro level classes at Harvard. :wink:

    At "normal" colleges and universities, you get that stuff in intermediate and upper-level classes.
  13. Dec 15, 2008 #12
    Not Calc of variations; that wasn't too hard to follow conceptually; you're just maximizing a function whose value depends on another function; which is hard too intuit, but not too difficult to understand formally.

    It was stuff about Poisson Brackets & Commuter operators on Hilbert Spaces making some phase space fuzzy (I think p -> i(hbar)partial(f/x) and q->q) and q->; I don't think we're responsible for it (It was like a, "this is cool look what we can do with Hamiltonians" thing, but it's something that's on my check-list of things I need to spend personal time with after the pre-final crunch.

    I've actually checked out Hardy and Courant and have been going through them for a few weeks; I don't make a terrible amount of progress because of the amount of time I spend on my Physics and Math Psets, but my plan is to slowly re-build the foundation I should've had.

    And I understand all of those proofs/believe I could prove all of those things, except that e^x is not a polynomial.
  14. Dec 15, 2008 #13


    User Avatar

    Think fundamental theorem. ;D
  15. Dec 15, 2008 #14


    User Avatar
    Homework Helper

    See things are not so bad. Most people need to grapple with these ideas a while to understand them. That is while colleges have yourepeat material several times with more details each time. Next year when you take Physics 151 you will have an advantage over those who took Physics 15a.

    For e^ is not a polynomial try either
    if p and q are polynomials deg(pq)=deg(p)+deg(q)
    with exp(x)exp(x=exp(2x)
    if p is a polynomial there exist a natural number n such that
  16. Dec 16, 2008 #15
    I meant to say 2nd year. No, it is standard even among best schools (cough MIT) to teach waves and lagrangian mechanics in second year, if not later. The same goes for real analysis. Unless Harvard grads have exceptional backgrounds, which I doubt given the US pre-college education system.

    That sounds dissapointing for Harvard. I thought they were supposed to be a good school. Not another weedout center where you can claim you survived.
  17. Dec 16, 2008 #16
    Yeah, but when's the last time Harvard accepted a student that only had a regular public school background? They usually have to show some initiative with learning on their own and junk.

    But still, you are right, it's not very likely because there is no telling of where each student is that way. I don't know. He also said he went into Hilbert space commutators and stuff like that. Seems like a LOT for any class, let alone a single quarter or semester of a Freshman class.

    I remember my 2nd year mathematical physics class being all about simple harmonic oscillators, which made us solve them via Laplace transforms, Fourier transforms and series, linear algebra, gahh... some other stuff I am forgetting now. In the classical mechanics class that year, well, it was the first time it was offered for sophomores, so we were the guinea pigs. We had some Hamiltonian stuff, some Lagrangian/Calculus of Variations stuff. I also remember a lot of pain.

    Anyway, my point is that doing all that stuff in first year is incredible ridiculous, or possibly even wicked awesome. You're either a REALLY smart kid for following along, or you all barely scratched the surface of each topic (which could STILL mean you are a really smart kid. :wink: )
  18. Dec 16, 2008 #17


    User Avatar
    Homework Helper

    From physics.harvard.edu
    "The goal of Phys 16 is to make sure that no student is bored. We’re pretty successful at this."
    "Only about 10 freshmen take 153."
    From math.harvard.edu

    Math 25
    A rigorous treatment of multivariable calculus, linear algebra, and introductions to other topics in advanced mathematics. This class is a springboard to the study of advanced math; the class thoroughly covers its topics but moves very quickly, and examples tend to be theoretical instead of concrete. Class meets three hours per week and homework can take from seven to fifteen hours per week to complete. The course is taught by a professor; previous instructors have varied greatly in teaching ability and prestige. A previous knowledge of proofs, linear algebra, or multivariable calculus is helpful, but not necessary. However, being 'gung-ho' about mathematics is a definite prerequisite. You should take this class if one or more of these describes you:
    You completed BC calculus, are very interested in math, and want a class that will allow you, with a lot of hard work, to catch up to the most advanced math students in your class.
    You have completed multivariable calculus and linear algebra and want a thorough, proof-based review of the topics before moving on to other mathematics.
    You have completed bits and pieces of advanced mathematics and want to take a class that will patch up the holes in your knowledge.

    Math 55
    This is probably the most difficult undergraduate math class in the country; a variety of advanced topics in mathematics are covered, and problem sets ask students to prove many fundamental theorems of analysis and linear algebra. Class meets three hours per week, plus one hour of section, and problem sets can take anywhere from 24 to 60 hours to complete. This class is usually small and taught by a well-established and prominent member of the faculty whose teaching ability can vary from year to year. A thorough knowledge of multivariable calculus and linear algebra is almost absolutely required, and any other prior knowledge can only help. Students who benefit the most from this class have taken substantial amounts of advanced mathematics and are fairly fluent in the writing of proofs. Due to the necessity of working in groups and the extensive amount of time spent working together, students usually meet some of their best friends in this class. The difficulty of this class varies with the professor, but the class often contains former members of the International Math Olympiad teams, and in any event, it is designed for people with some years of university level mathematical experience. In order to challenge all students in the class, the professor can opt to make the class very, very difficult. You should take this class if one or more of these describes you:
    You are fairly certain that you want to be a math concentrator and want to be challenged to your limit.

    You have a solid base in advanced mathematics and are very comfortable with proofs and rigorous arguments.

    You want math to be your most important class.
    Last edited: Dec 16, 2008
  19. Dec 17, 2008 #18
    I mean this is a great place; I probably learn more from other students than in any class. I guess I'm just having a really hard time adjusting to this.

    I don't really understand how I got in; I did do a lot of Self-Study (In that I learned Bio/Chem/Physics from books and not in actual courses), but I didn't really do anything with them except take AP tests and do some small local science competitions around town. I started like two clubs in HS, and did Marching Band. Maybe like 1/3rd of the stuff everyone else here did.

    But I'm here now, so what matters is catching up. I hope that the gap will seem less mammoth eventually.
  20. Dec 17, 2008 #19
    I sometimes feel that there is a huge gap between me and where I should be in terms of understanding math, but the only thing we can do is suck it up and keep at it. Right now I'm a freshman like you, and the way I'm going about it is to just do the best I can and not worry about GPA. If you like math and physics, I would stick with it. You are in a better position than most, being in Harvard and taking classes like Math 25 :smile:
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook