Undergraduate Research: How to Get Started in Theoretical Physics

[nonequilibrium]

Hello. It seems that a demand of renowned international graduate colleges is having research experience and even publications in your undergraduate years.

How does one actually get to this? Does one simply contact a professor and offer their services? I've only had the first year of bachelor in Physics (total of 3 years), so I just can't imagine I could be of any use to a Theoretical Physicist (as theoretical physics is my own interest). I'm thrilled by the sound of doing my own research in the field (under a professor), but at this point it seems like offering my services would seem like nothing but sucking-up, as there's nothing subtantial to offer. I hope I'm wrong on this, cause that's not something I'm willing to do: I do physics because I love it and making it all look like a competition would ruin it, and I do pass for that.

Thanks for all insights on the matter!

 

[gretun]

I live in Canada, I got it worst than you. I also agree with going for it, but it may seemed to them that you are just sucking up. I am not even sure if freshman research matters.

 

[eri]

If you're interested enough in the field to want to get a PhD, you should want to get research experience as early as possible since that will give you a much better idea of what you'd be doing for a living than your classes will. Ask around - you should be able to find someone willing to take you on in some capacity even with your limited background. However, keep in mind that taking on a research assistant can be a lot of work for a professor - they have to put in the time to train you and meet with you, so don't do this unless you're wiling to devote serious time to the project. I've had several undergrad researchers work with me over the last few years, and most were a severe disappointment. Don't be that person. I did an REU the summer after my first year of college, and I had a freshman REU student this summer (who's actually making progress). It's not impossible. Sure, he wanted to do theoretical physics, but it's not capable of it yet - so he's doing an observational project instead, and maybe in a few years he'll be ready for theory.

 

[gigabyte3000]

Definitely ask your physics professors if they know of any research opportunities. It's the most direct and reliable way of finding out what there is. But I definitely think that being able to work with a theoretical physicist would be fairly challenging. But, as I always find myself say, why is it so important that you get a theoretical physics research experience? I would say that you should definitely at least try out experimental first. I think a lot of people don't realize how interesting experimental is, and automatically assume that theoretical physics is the "coolest".
Oh, and I've heard that professors can sense bull from a mile away. So no, sucking up probably wouldn't be good. But you do want to develop a good relationship with them.

 

[Knochera]

Don't be worried about asking a professor if you can participate in their research with them. You're correct in thinking that research is a huge part of the grad school process, so the earlier start you have the better chance you will have at getting into the program you want. Even if you don't have much to offer a research group right now, you'll eventually acquire the skills to contribute. Most people starting out in a research group don't have a clue as to what's going on for the first week or two. You'll learn quickly, so don't worry about that factor. Pick a field you would want to go to graduate school for, pick a professor who works in that field, and ask them if they know of any research you could participate in. Don't make the decision based on what you feel you could contribute to best, or which professors seem the least intimidating -- that's what I did my freshman year, and I regret that I didn't just go straight to work in a field I was more interested in.

 

[eXorikos]

You can ask professors during the coming year (I would say a first year hasn't got sufficient knowledge of physics) if they have any projects you can do during the summer. You will in a few cases even get paid. I know that I could do it after my bachelors, but maybe if you're grades are good you could get something earlier.

 

[Annonymous111]

I strongly disagree that grad. schools look for research publications. I mean, if you can publish before going to grad. school, why go to grad. school! Just research in the comfort of your own bedroom. Grad. school is for people who need help to do research. Besides, plenty of applicants get into top grad. schools with no research. GPA and letters of recommendation and hard courses are most important. Research is more of a secondary factor. To elaborate, if you haven't got good grades, you can make it up with research, but if you have got good grades, research is unnecessary.

 

[negru]

What grad schools look for is irrelevant. What matters is what the other students have been doing. If everyone else has been doing research, your lack of experience isn't going to impress anyone.

Grad schools don't exist to "help" anyone. They're there to produce research. They get funding only if they produce research. It's not some sort of charity.

 

[DR13]

I strongly disagree that grad. schools look for research publications. I mean, if you can publish before going to grad. school, why go to grad. school! Just research in the comfort of your own bedroom. Grad. school is for people who need help to do research. Besides, plenty of applicants get into top grad. schools with no research. GPA and letters of recommendation and hard courses are most important. Research is more of a secondary factor. To elaborate, if you haven't got good grades, you can make it up with research, but if you have got good grades, research is unnecessary.

This is wrong in a few different ways. If you are applying to a top tier grad school, lots of applicants have great grades. So if one has research experience (or if they have been published) it is a great way to separate oneself from the pack. Also, the comment of research in one's bedroom is asinine. One needs to be able to collaborate with others and use equipment that only universities have. Furthermore, grad school is not for people who "need help to do research". It is for people who want to do research and further their education (and get a PhD).

 

[Annonymous111]

THe most important way to "separate from the pack" is to take highly advanced grad. classes in breadth and depth and get A's in them. If every applicant has taken at the most 6 or 7 grad. classes, and you've taken 20 grad. classes in high level math (by high level, I mean surpassing all the grad. classes at the school you're applying to) then surely that'll separate yourself from the pack. Besides, no one expects you to publish anything serious before grad. school. And few people do. What's more important is the experience of research. But the thing is, if you experienced research for god knows how many years but never published anything, there's no way to prove it.

 

[DR13]

THe most important way to "separate from the pack" is to take highly advanced grad. classes in breadth and depth and get A's in them. If every applicant has taken at the most 6 or 7 grad. classes, and you've taken 20 grad. classes in high level math (by high level, I mean surpassing all the grad. classes at the school you're applying to) then surely that'll separate yourself from the pack. Besides, no one expects you to publish anything serious before grad. school. And few people do. What's more important is the experience of research. But the thing is, if you experienced research for god knows how many years but never published anything, there's no way to prove it.

1. I am guessing that, "If every applicant has taken at the most 6 or 7 grad. classes, and you've taken 20 grad. classes in high level math " is hyperbole (just to clarify in case the OP was confused).
2. "Besides, no one expects you to publish anything serious before grad. school. And few people do." Exactly my point. If you are lucky/good enough to get published then that will be impressive.
3. "If you experienced research for god knows how many years but never published anything, there's no way to prove it." Sure there is. You get letters of recomendations from the supervising the supervising professor. Also, you put it on your application. If one has good research experience then one should let the admissions comittee know.

Plus, as a grad student you are *paid* to do research. If a university is paying you to do research then they are going to want you to know what you are doing. So they are going to want someone who can jump right in as opposed to someone who has no idea what they are doing. Furthermore, grad schools like to admit people who they think will finish their PhD. If someone with no research experience is admitted then they may find out that they hate research and drop out of the program. However, if the comittee admits someone with prior research experience then they know they have someone who will like what they will be doing.

 

[Vanadium 50]

I strongly disagree that grad. schools look for research publications. I mean, if you can publish before going to grad. school, why go to grad. school! Just research in the comfort of your own bedroom. Grad. school is for people who need help to do research. Besides, plenty of applicants get into top grad. schools with no research. GPA and letters of recommendation and hard courses are most important. Research is more of a secondary factor. To elaborate, if you haven't got good grades, you can make it up with research, but if you have got good grades, research is unnecessary.

Graduate schools train researchers. They are very interested in how well someone will do at this. Additionally, doing research means someone is developing a close relationship with someone who is writing their letters, and detailed letters are far more useful than ones that say "he took my class and got an A".

Do you have any special expertise in this area? Before you asked to have this thread deleted, you were asking if you could get into one particular grad school, which suggests you are an undergraduate - someone who hasn't even experienced grad school admissions from the point of view of the applicant, much less the department?

 

[Annonymous111]

Graduate schools train researchers. They are very interested in how well someone will do at this. Additionally, doing research means someone is developing a close relationship with someone who is writing their letters, and detailed letters are far more useful than ones that say "he took my class and got an A".

Do you have any special expertise in this area? Before you asked to have this thread deleted, you were asking if you could get into one particular grad school, which suggests you are an undergraduate - someone who hasn't even experienced grad school admissions from the point of view of the applicant, much less the department?

I didn't have special expertise in the area earlier, but I do have now after speaking to experts in the area.

 

[cristo]

I didn't have special expertise in the area earlier, but I do have now after speaking to experts in the area.

You don't gain "expertise" by talking to an expert.

 

[Fragment]

Furthermore, what looks better to grad. school committees?

Situation A: Joe the devoted student with some research at his local university. Great grades and some exposure to advanced math subjects (hopefully introductory graduate material), letters of recommendation, with perhaps some extra-curriculars.

OR

Situation B: (What Anonymous111 says is better) Superficial, at best, exposure to advanced topics from the comfort of one's basement. Self-recognized education, little or no recognized research, and no letters of recommendation.

The point here is that by going to university, employers and academics can recognize that you have taken, and passed/failed said course. Therefore they don't have to sort out the truly knowledgeable from the self-learned-perhaps-less-knowledgeable, education does that for them.

Additionally, at my university, grad. classes during undergrad are only offered to final year students with permission. Unless one is absolutely incredible (Think Daniel Kane incredible), one will not have exposure to many graduate-level topics before graduate school.

 

[Annonymous111]

I didn't claim situation B. I claimed:

Situation B: An exposure to at least 10 grad. math classes. all fast paced, which involve reading the literature and coming up with original ideas, pace equivalent to MATH55 at Harvard, and getting A's in them. Assessment in this courses involves writing survey articles on advanced aspects of the literature with highly original viewpoints. Outstanding letters of recommendation. Not necessarily orginal research but new ways of thinking on current active areas of mathematical research from a diverse selection of mathematical subjects ranging from algebra, analysis and topology.

Also I disagree that one will not have exposure to many grad. level topics before grad. school. Harvard gives permission to most of its undergraduates to take grad. clases. IN fact they encourage it from their third year onwards.

 

[Annonymous111]

Just to clarify I can understand that most people come out with superficial understanding of subjects even after getting A's in them, simply because its tough to do 4 math subjects in one go and "think outside the box" in all of them. But it's not fair to generalize that all do. If an applicant is strong enough to show that his understanding is highly advanced, then that's a different ball game altogether.

 

[DR13]

I didn't claim situation B. I claimed:

Situation B: An exposure to at least 10 grad. math classes. all fast paced, which involve reading the literature and coming up with original ideas, pace equivalent to MATH55 at Harvard, and getting A's in them. Assessment in this courses involves writing survey articles on advanced aspects of the literature with highly original viewpoints. Outstanding letters of recommendation. Not necessarily orginal research but new ways of thinking on current active areas of mathematical research from a diverse selection of mathematical subjects ranging from algebra, analysis and topology.

Also I disagree that one will not have exposure to many grad. level topics before grad. school. Harvard gives permission to most of its undergraduates to take grad. clases. IN fact they encourage it from their third year onwards.

Where are you getting these numbers from? 10 grad classes?? That seems like a rediculous amount (maybe I'm wrong but 10 seems pretty high). And I don't understand why you won't accept that grad schools like to see research? It's a fact!

As a side note: Math 55 is *one* class at *one* university.

 

[negru]

I didn't claim situation B. I claimed:

Situation B: An exposure to at least 10 grad. math classes. all fast paced, which involve reading the literature and coming up with original ideas, pace equivalent to MATH55 at Harvard, and getting A's in them. Assessment in this courses involves writing survey articles on advanced aspects of the literature with highly original viewpoints. Outstanding letters of recommendation. Not necessarily orginal research but new ways of thinking on current active areas of mathematical research from a diverse selection of mathematical subjects ranging from algebra, analysis and topology.

Also I disagree that one will not have exposure to many grad. level topics before grad. school. Harvard gives permission to most of its undergraduates to take grad. clases. IN fact they encourage it from their third year onwards.

So you're claiming that this is enough to what, exactly? Give you a strong chance? Average chance? Non-zero chance? Good chance at a particular school? Good chance at at least one out of five? I don't think anyone claimed that it's IMPOSSIBLE to get into a top school without research. We were talking about what it takes to have a strong chance, at one particular school, assuming no special hooks (like matching research interests, available funding etc.) I'm pretty sure that if your gardener applies to harvard 10^3000 times he'll get in at least once. Not all accepted students will be geniuses, true. However, for every non-genius student accepted, there will be a few hundred others not accepted.Also you should know that in general profs don't really care about teaching and courses. That's why teaching plays no role in getting tenure. Any advanced courses they teach are only meant to get students up to date with research topics, so they can start researching faster. Advanced students should know this. There's no point wasting time to find new/original proofs/ideas on things already solved. No one will care about that. People will only care if that shows you have potential to do research. But the better way of showing that you have research potential is doing research.

 

[HeLiXe]

I really appreciate this thread and would like to thank the op and everyone who replied. I have a question in regards to undergraduate research, I hope its not too much of an aberration, but I was wondering how an undergrad gets involved in research. I was originally going to attend a school that involves its upper level undergrad students in research, but I decided to stay local. Since then I've been wondering how I can get involved in research when I reach upper level, I'm a freshman now.

 

[DR13]

I really appreciate this thread and would like to thank the op and everyone who replied. I have a question in regards to undergraduate research, I hope its not too much of an aberration, but I was wondering how an undergrad gets involved in research. I was originally going to attend a school that involves its upper level undergrad students in research, but I decided to stay local. Since then I've been wondering how I can get involved in research when I reach upper level, I'm a freshman now.

There are a couple of ways to get involved. At my university we have a undergrad research program for freshman and sophomores (which I am doing). In the research group that I am in there are seven undergrads. Two of us are in the program, the other five are just doing it (they talked with the head professor). The five are actually majoring in the research area (materials science and engineering). So if your university does not have a specific program I would talk to/email some of the professors in your interest area. But before you start sending off emails you may want to talk with some upperclassmen to see which professors would be open to taking you on.

 

[Annonymous111]

There's no point wasting time to find new/original proofs/ideas on things already solved. No one will care about that. People will only care if that shows you have potential to do research. But the better way of showing that you have research potential is doing research.

I strongly disagree. Have you heard of Furstenberg's proof on the infinitude of the prime numbers? Everyone knows that there are infinitely many prime numbers, that's been solved since Euclid's time. But Furstenberg found a proof of it in his undergraduate years and got it published. And people DID care. You know why? He got into Princeton's graduate math program that's why. And the infinitude of the prime numbers is the most basic math, people learned about it when they are 10 years old or something. If you can find original proofs and new ways of looking at harder things, let's say Fermat's last theorem, or anything basic really, that's still research. So finding new proofs of known things is called research. After all if you can find proofs, you're doing research, especially if you're finding proofs of things that eluded mathematicians for a long time.

 

[Annonymous111]

Where are you getting these numbers from? 10 grad classes?? That seems like a rediculous amount (maybe I'm wrong but 10 seems pretty high). And I don't understand why you won't accept that grad schools like to see research? It's a fact!

As a side note: Math 55 is *one* class at *one* university.

Not at Harvard, no. The average admitted applicant at Harvard has 7 grad. classes. Sure grad. schools like to see research. But my point is that its so rare, that most applicants who get in don't have any real research. Even at Harvard.

 

[negru]

But my point is that its so rare, that most applicants who get in don't have any real research. Even at Harvard.

Sure, but how many of those applicants would you have called shoo-ins? That's the whole point.

 

[Annonymous111]

Sure, but how many of those applicants would you have called shoo-ins? That's the whole point.

Ah OK. I get what you're saying now. I agree completely. Doing research definitely gets you on the good books with grad. schools. The thing is though, which was what I was trying to say, was that doing research in chem. and stuff is a lot easier than doing research in math. Often doing research in math takes a lot of luck, especially if you've only done "real" math for 3 years in undergrad. However, just to clarify, if you do research and publish a paper to a decent journal in math, does that make you a shoo-in?

 

[bogarts21]

What's "real" math?

 

[DR13]

However, just to clarify, if you do research and publish a paper to a decent journal in math, does that make you a shoo-in?

Nothing makes you a shoe-in when applying to a place like Harvard (unless you are the child of someone important :P)

 

[eri]

Not at Harvard, no. The average admitted applicant at Harvard has 7 grad. classes. Sure grad. schools like to see research. But my point is that its so rare, that most applicants who get in don't have any real research. Even at Harvard.

Are you just talking about the math department? Because I have a few friends who attended Harvard grad school for physics and astronomy; none of them had taken more than one graduate level classes as an undergrad, if that, but they all had done multiple REUs and had publications. That's what got them in.

 

[negru]

. However, just to clarify, if you do research and publish a paper to a decent journal in math, does that make you a shoo-in?

Well, if it's a good paper, and from the recs it follows that indeed you've done most of the work, and there are people there interested in that kind of work, I'd say you have a strong chance.

It's difficult to quantize these things. Like if you go physicsgre.com and check people's profiles
(eg http://www.physicsgre.com/viewtopic.php?f=3&t=2824 ), you'll see that many people with papers, good grades, etc didn't get into Harv/Pton. And that's for physics, for (pure) math I assume it's even harder.

 

[Anonymous217]

Not at Harvard, no. The average admitted applicant at Harvard has 7 grad. classes.

Do you have a link to this statistic?

 

[Annonymous111]

Well, if it's a good paper, and from the recs it follows that indeed you've done most of the work, and there are people there interested in that kind of work, I'd say you have a strong chance.

It's difficult to quantize these things. Like if you go physicsgre.com and check people's profiles
(eg http://www.physicsgre.com/viewtopic.php?f=3&t=2824 ), you'll see that many people with papers, good grades, etc didn't get into Harv/Pton. And that's for physics, for (pure) math I assume it's even harder.

Thanks for this link. Notice one thing though. Most people with a GPA of 4.0 were accepted at one highly ranked graduate program. And that's my point. If you can secure a GPA of 4.0 (with advanced classes of course!) then you should get in at least somewhere good.

Research in physics is much easier than math. You can just go to a lab and do an experiment with some prof. and that counts as research. So saying that people with lots of research in physics were not accepted doesn't speak volumes about math. It's certainly not an exaggeration to say that math research is much harder than physics research.

 

[Annonymous111]

Are you just talking about the math department? Because I have a few friends who attended Harvard grad school for physics and astronomy; none of them had taken more than one graduate level classes as an undergrad, if that, but they all had done multiple REUs and had publications. That's what got them in.

That's my point you see. If you haven't done enough grad. classes, then remember you're competing with people who have done upper division grad. classes and many of them, and secured A's. So the only way you can improve your chances is by doing research. Research has a higher weightage than grad classes but the fact is that if you can do 10-12 grad. classes, then that's something grad. schools cannot ignore, even if you haven't got research.

 

[DR13]

If you can secure a GPA of 4.0 (with advanced classes of course!) then you should get in at least somewhere good.

Talk about easier said than done!

That's my point you see. If you haven't done enough grad. classes, then remember you're competing with people who have done upper division grad. classes and many of them, and secured A's. So the only way you can improve your chances is by doing research. Research has a higher weightage than grad classes but the fact is that if you can do 10-12 grad. classes, then that's something grad. schools cannot ignore, even if you haven't got research.

10-12 grad classes... Where are you getting these numbers from?? A lot of what you are saying just seems like assumptions

 

[Anonymous217]

^That. I still want to see the link that states the 7 grad course average for Harvard students. It's not that we think you're lying. We just want to see some credible resource to these numbers you put out.

 

[Annonymous111]

Talk about easier said than done!



10-12 grad classes... Where are you getting these numbers from?? A lot of what you are saying just seems like assumptions

It's not as hard as it looks. Let's think of it this way. You do 32 courses typically in undergrad. (Let's assume this.) If you only do physics and math courses, then about 12 physics courses should be enough to cover undergrad. physics, right? Then you have 22 courses to do grad. level physics. You probably end up doing some math, so that makes at least 10 grad. level physics classes.

If someone is going to end up doing courses in some random subjects, then I have to question their hunger to become successful. If you want to become great in physics (or any subject really), you're going to make your chances a lot higher if you have a definite sense of what you will do before even starting university. If you want to get into Harvard, Princeton, Stanford etc for physics (or math), you'll need to do physics classes and only physics classes (with a little math) from freshman year onwards. No looking back. And if you do that, you'll do 10-12 grad. classes. Simple!

 

[Annonymous111]

Talk about easier said than done!

Sure. But if you want to get into Harvard, Princeton, Stanford etc., what choices do you have?

 

[ephedyn]

As far as I know, 10-12 grad courses is near impossible even if you are a very strong student in Stanford.

http://www.stanford.edu/dept/physics/academics/Sample4yearSchedules.pdf

There's not even half of that factored in here. Anonymous217 is being nice. I will call this a bluff. I believe if you've reached the point where you're doing some graduate courses, you'll have graduated and moved on the grad school rather than take 10-12 grad courses. You make two fundamental flaws in what you're saying, (1) that the number of graduate physics courses you do nearly mirror the number of upper division mathematics courses you will have to do, e.g. differential geometry, abstract algebra, PDEs/BVPs, numerical methods, (2) graduate physics courses are way too specialized, and doing 10 will stretch you very thin.

Harvard, MIT, Stanford, as far as I know, require at least 1 experimental course in the undergraduate level. Some 8 classes will have to go to humanities courses. (Let's assume the guy doesn't do both grad humanities and grad science courses in college.) The core requirements will often require at least 2 courses in calculus, and 2 courses in classical mechanics and electromagnetism. Since it's a physics major we're assuming, 2 courses in at least QM/solid state/thermo/nuclear etc. is not far-fetched. Now, I'm giving you the benefit here - I'm supposing that it only takes 6 classes to finish ALL of general requirements for undergraduate physics and calculus. I'll assume from this point on no other undergraduate physics are taken.

That's 15 classes taken out of the picture. 17 more to go.

I don't think you start college knowing all of upper division partial differential equations, linear algebra, abstract algebra, real analysis, numerical methods etc. - that, VERY surely, you will encounter in graduate classes. Even if you stretch it a little, and assume you need only take 1 of each of the above, which leaves 12, just nice to pull off your 12 graduate course trick, something is very strange about doing 12 specialized courses and 4 general physics courses (we're even assuming that the said person somehow assimilates the intermediate, upper division physics knowledge that is often prerequisite background to the specialized courses from just these 4 general courses).

More realistically, I am in that position myself doing abstract algebra, differential geometry and upper division physics in my first semester, and will be taking my first graduate physics course next semester. I am still considering double majoring, which gives me space for more classes to achieve your grand suggestion: even so, listing all of courses I will take, I will only have done 6 graduate courses. At which point, it doesn't make sense to stay on anymore because I will have satisfied all of my requirements, and I might as well be doing any further grad courses as a grad student, where it's free. It does come across as very strange to me as to why anyone would stay on and do 10-12, being in this position myself.

 

[Annonymous111]

As far as I know, 10-12 grad courses is near impossible even if you are a very strong student in Stanford.

http://www.stanford.edu/dept/physics/academics/Sample4yearSchedules.pdf

There's not even half of that factored in here. Anonymous217 is being nice. I will call this a bluff. I believe if you've reached the point where you're doing some graduate courses, you'll have graduated and moved on the grad school rather than take 10-12 grad courses. You make two fundamental flaws in what you're saying, (1) that the number of graduate physics courses you do nearly mirror the number of upper division mathematics courses you will have to do, e.g. differential geometry, abstract algebra, PDEs/BVPs, numerical methods, (2) graduate physics courses are way too specialized, and doing 10 will stretch you very thin.

Harvard, MIT, Stanford, as far as I know, require at least 1 experimental course in the undergraduate level. Some 8 classes will have to go to humanities courses. (Let's assume the guy doesn't do both grad humanities and grad science courses in college.) The core requirements will often require at least 2 courses in calculus, and 2 courses in classical mechanics and electromagnetism. Since it's a physics major we're assuming, 2 courses in at least QM/solid state/thermo/nuclear etc. is not far-fetched. Now, I'm giving you the benefit here - I'm supposing that it only takes 6 classes to finish ALL of general requirements for undergraduate physics and calculus. I'll assume from this point on no other undergraduate physics are taken.

That's 15 classes taken out of the picture. 17 more to go.

I don't think you start college knowing all of upper division partial differential equations, linear algebra, abstract algebra, real analysis, numerical methods etc. - that, VERY surely, you will encounter in graduate classes. Even if you stretch it a little, and assume you need only take 1 of each of the above, which leaves 12, just nice to pull off your 12 graduate course trick, something is very strange about doing 12 specialized courses and 4 general physics courses (we're even assuming that the said person somehow assimilates the intermediate, upper division physics knowledge that is often prerequisite background to the specialized courses from just these 4 general courses).

More realistically, I am in that position myself doing abstract algebra, differential geometry and upper division physics in my first semester, and will be taking my first graduate physics course next semester. I am still considering double majoring, which gives me space for more classes to achieve your grand suggestion: even so, listing all of courses I will take, I will only have done 6 graduate courses. At which point, it doesn't make sense to stay on anymore because I will have satisfied all of my requirements, and I might as well be doing any further grad courses as a grad student, where it's free. It does come across as very strange to me as to why anyone would stay on and do 10-12, being in this position myself.

You can surely get special exemption from humanities courses at Harvard, Stanford etc. if you prove that you're so good that they can't ignore exempting you.

 

[Annonymous111]

Nothing makes you a shoe-in when applying to a place like Harvard (unless you are the child of someone important :P)

What if you solve the Riemann hypothesis tomorrow? Even if the rest of your application is full of F's and you've got nothing, you'll be accepted, in fact, offered professorship at Harvard, if you solve the Riemann hypothesis. Ditto for any long unsolved problem in a field of math, but to a lesser extent.

 

[Annonymous111]

You can surely get special exemption from humanities courses at Harvard, Stanford etc. if you prove that you're so good that they can't ignore exempting you.

Also, at Harvard, you can do whatever you like. That's because if you're good enough to get in there, you'll be good enough to make your own decisions in life. No-one needs to force you to do courses you don't want to.

Just a question: Do mathematics grad. schools care about the grades you get in non-math courses, or the breadth and depth of the non-math courses you've taken?

 

[DR13]

What if you solve the Riemann hypothesis tomorrow? Even if the rest of your application is full of F's and you've got nothing, you'll be accepted, in fact, offered professorship at Harvard, if you solve the Riemann hypothesis. Ditto for any long unsolved problem in a field of math, but to a lesser extent.

Oh come on. If you solve the riemann hypothesis (or something of the sort) you will be getting great grades.

 

[DR13]

Also, at Harvard, you can do whatever you like. That's because if you're good enough to get in there, you'll be good enough to make your own decisions in life. No-one needs to force you to do courses you don't want to.

Why are you assuming that we are only talking about Harvard undergrads?

 

[HeLiXe]

There are a couple of ways to get involved. At my university we have a undergrad research program for freshman and sophomores (which I am doing). In the research group that I am in there are seven undergrads. Two of us are in the program, the other five are just doing it (they talked with the head professor). The five are actually majoring in the research area (materials science and engineering). So if your university does not have a specific program I would talk to/email some of the professors in your interest area. But before you start sending off emails you may want to talk with some upperclassmen to see which professors would be open to taking you on.

Thanks DR13. I discovered that there are a lot of opportunities for undergrad research in the school I want to transfer to. The physics department has several professors doing research at the National High Magnetic Field Laboratory, and the chemistry department has a lot of research going on as well. I'll be looking more into everything. Thanks again for pointing me in the right direction :)

 

[Annonymous111]

Oh come on. If you solve the riemann hypothesis (or something of the sort) you will be getting great grades.

That's not necessarily true. You could be a very lazy person, or get bored of easy courses because they're too easy, and simply get bad grades not because you can't get good grades, but because you're intellectual ability is so high that university doesn't challenge you.

Just to prove that this isn't hypothetical, let's take Stephan Smale. He got F grades in university (in both undergrad. AND grad. school). But he went on to prove one of the deepest theorems in differential topology and was awarded the fields medal.

It just goes to show that grad. schools should stop being shallow and start looking at the broader picture. GPA doesn't mean all that much for everyone. Now I'm not downplaying the importance of GPA out of selfishness. I myself have great grades, I've only gotten A's. But the point is that some people don't, and they shouldn't be disregarded simply because they don't. I'll be one to be admit that I know people more talented than I who've got F grades in courses that I could get A's without virtually lifting a finger. But they're volumes more talented than me. And I'm talking about hard courses here.

For example, doing 4 courses at once means that you can get bad grades simply because you've got so much to do. It's very possible that someone could develop a deeper understanding of material if they did 2 or so courses, rather than 4.

 

[Annonymous111]

Why are you assuming that we are only talking about Harvard undergrads?

I never assumed this.

 

[negru]

Just a question: Do mathematics grad. schools care about the grades you get in non-math courses, or the breadth and depth of the non-math courses you've taken?

Non-math courses count for absolutely zero. Even if they did count for anything at all, doing virtually anything else but math-related would count more.


I think we went over this on some other thread as well, but it is indeed possible in many schools (top and non-top) to take only the courses you want, with zero general requirements outside of finishing a major. However, some math departments do require language classes. But that's just 1-2 classes. Bad/good? I don't know, undergrad is pretty much your last chance of finding other random stuff about the world (in an academic setting), but the competition for grad school is pretty fierce. You don't want to be behind. It's all about how much you're willing to sacrifice. It's always about that.


Regarding ephedyn's countdown: you can always skip the undergrad classes and take the corresponding grad ones instead. There are generally no real prereqs for classical mechanics or quantum mechanics at the grad level (or GR for that matter). Pretty much the same topics are covered (especially in QM. personally I learned what a hamiltonian is from a QM course), the only difference is you'll (possibly) need to work harder.

I don't see why a student who aces (or almost aces) undergrad QM wouldn't be able to at least decently handle grad QM instead. And after you're done with CM and QM, you're practically at the same level of any grad student.

I mean, it's not like some phase-transition occurs once you get to grad school. You don't receive 100 new skill points or smth. The only thing that changes is the mentality (both yours, your peers', and your profs'). I don't see why that change in mentality couldn't happen in one's undergrad years.

 

[Annonymous111]

Non-math courses count for absolutely zero. Even if they did count for anything at all, doing virtually anything else but math-related would count more.


I think we went over this on some other thread as well, but it is indeed possible in many schools (top and non-top) to take only the courses you want, with zero general requirements outside of finishing a major. However, some math departments do require language classes. But that's just 1-2 classes. Bad/good? I don't know, undergrad is pretty much your last chance of finding other random stuff about the world (in an academic setting), but the competition for grad school is pretty fierce. You don't want to be behind. It's all about how much you're willing to sacrifice. It's always about that.


Regarding ephedyn's countdown: you can always skip the undergrad classes and take the corresponding grad ones instead. There are generally no real prereqs for classical mechanics or quantum mechanics at the grad level (or GR for that matter). Pretty much the same topics are covered (especially in QM. personally I learned what a hamiltonian is from a QM course), the only difference is you'll (possibly) need to work harder.

I don't see why a student who aces (or almost aces) undergrad QM wouldn't be able to at least decently handle grad QM instead. And after you're done with CM and QM, you're practically at the same level of any grad student.

I mean, it's not like some phase-transition occurs once you get to grad school. You don't receive 100 new skill points or smth. The only thing that changes is the mentality (both yours, your peers', and your profs'). I don't see why that change in mentality couldn't happen in one's undergrad years.

Definitely agree that it shouldn't count toward anything. But I have heard "rumours" that it hurts your application if you get bad grades in aything, whether math related or not. Does anyone know anything about these rumours?

Yeah, I know what you mean with sacrificing. I really don't see the point in taking something other than math just for the sake of it. I mean, if someone's transcript is just math courses, that's OK. Maths in itself counts for several subjects, and it's quite possible to be good at one aspect of math but not be very nimble with another. So rather than try to learn what the smell of NH3 is in a Chem. lab, it's probably better to try to accumulate as much math as possible.

Grad. school is important, but one thing I've learned recently is that there's no point thinking about it all the time. At the end of the day, if you stick to doing the best you can and persevere for 4 years, and you have some talent, you'll be as prepared as you can be for grad. school. Trying to do things just to get grad. schools attention is pretty pointless unless you really want to do it for your own sake. Grad. school should be because you want to do research. So I hate people taking baby classes to get a high GPA just for the sake of grad. school. I mean what's the point in life if you're going to do that?

Anyway grad. school isn't the end of the world even if you don't get in a top one. For example, Dennis Gaitsgory, who is now a full professor at Harvard, did his PhD at Tel Aviv University, which isn't necessarily the best, but it isn't bad either.

Fully agree with you. There're certain top schools that don't have requirements but sometimes these have to be negotiated. You only have one chance in life to do some random course, certainly true, but on the other hand, you only live once. There's no point in being jack-of-all-trades and master-of-none.

Anyway, enough of my babbling.

 

[hadsed]

In a perfect world, Anon111, you would be right. But we don't live in a perfect world, so there will always be those people who will try to beat you out by beating the system. It just so happens now that there are so many people doing this that it's beginning to have a real effect on who gets what places in the top schools. Admissions has to do something, and the easiest thing to do is to say, 'Everyone who we think is smart enough should get amazing grades effortlessly anyway'. That way they can try and fish out the really special people from the average.

That's what I think anyway.

 

[ephedyn]

Getting back on topic, I don't think bad GPA + research vs good GPA + no research requires comparison.

Personally, I've never met a person involved in research who isn't already doing well in school. I don't think someone who doesn't have enough to get through his classes, at least by talent or focus and effort, has what it takes to spend time outside the class doing any kind of research, be it experimental "grunt" work or computational "number-crunching" work even if it is true that these require less intellectual ability.

In the case of Stephen Smale, the story is very much incomplete. He did have some bad grades, but those were the exception, not the rule. Moreover, his biography does mention that he started working hard before he reaped the results. My favorite story involvess the Nobel laureate chemist Robert Woodward, who was expelled from his school because he was too absorbed in his own endeavors in the lab after school than his classes per se, but went back, did 16-17 courses in in a semester and eventually made it to grad school.

But one has to take these with a grain of salt - Smalt was born in 1930. In his time, you could attend the first day of college, decide you didn't like it that much, pack up, and drive down to another part of the country and attend another college. It wasn't as difficult to get in undergrad/grad school, because the world population was still recovering from the Great Depression. Today, a little less than 7 out of 100 students who apply to Harvard are accepted - and the number of "valedictorians" in the college market outnumber the total applicant pool you had in those days. Similarly, there weren't 'established standards' to look at for grad school applicants - research, Putnam/IMC, scholarships (Goldwater, Churchill etc.), fellowships whatnot. You could really get in grad school nearly just because you applied, and had a close enough relationships to faculty staff who had connections.**************off-topic*************

Anon111: I agree with you that one's grad school might not matter that much. There are people who have graduated from ETH, the grand Ecoles etc. but went on to do a lot. We don't have to look very far - Geim and Novoselov, who won the 2010 Nobel prize in Physics.

negru: Surely. Most of the time we can either test out of classes, or we already have humanities credits from high school. Moreover, if you are in a top school, it's likely that you already have many credits transferred over from high school. But that doesn't happen very often - not enough for there to have enough undergrads-who-had-done-ten-grad-courses to fill whole grad school departments! The people I know who have gone to grad school in Cambridge, Oxford, MIT, Imperial etc. were every bit human: they attended introductory courses in calculus, linear algebra and differential equations like almost everyone else and by no means managed to exempt themselves from humanities (many of them hated these as much). As a matter of fact, the Putnam fellows I know at some of these top grad schools did 4 years in college like everyone else, because they had to stay back and slowly finish off their humanities courses, like everyone else.

In fact even in a school like MIT where students have gone through a lot of high school preparation in university calculus and establish themselves as talented science students, most of the people actually start out with calculus or at most differential equations, introduction to classical mechanics and like everyone else among them, did their load of humanities stipulated in their university requirements.

Top schools are in fact more familiar with the genius kid who wants to get out of his humanities classes, and so have higher standards for people who want to do the same. Like Anonymous217, I would love to know who are these real people (enough to fill a whole department) who have done 10-12 courses in grad school level, not because I don't want to believe it, but because I want to believe it - it really interests me if I can have a credible source about them.

Lastly, to me, it sounds that my argument has sealed enough of a proof that most people who get in these top grad schools don't actually do 10-12 grad school courses in college, because the counterarguments that one can come up with against it require an immense suspense of belief: "you have to work very hard" and "you can become so good that you convince them to let you do whatever you want". This sounds more like Hermione from Harry Potter, or some other book than reality.

 

[Annonymous111]

Getting back on topic, I don't think bad GPA + research vs good GPA + no research requires comparison.

Personally, I've never met a person involved in research who isn't already doing well in school. I don't think someone who doesn't have enough to get through his classes, at least by talent or focus and effort, has what it takes to spend time outside the class doing any kind of research, be it experimental "grunt" work or computational "number-crunching" work even if it is true that these require less intellectual ability.

In the case of Stephen Smale, the story is very much incomplete. He did have some bad grades, but those were the exception, not the rule. Moreover, his biography does mention that he started working hard before he reaped the results. My favorite story involvess the Nobel laureate chemist Robert Woodward, who was expelled from his school because he was too absorbed in his own endeavors in the lab after school than his classes per se, but went back, did 16-17 courses in in a semester and eventually made it to grad school.

But one has to take these with a grain of salt - Smalt was born in 1930. In his time, you could attend the first day of college, decide you didn't like it that much, pack up, and drive down to another part of the country and attend another college. It wasn't as difficult to get in undergrad/grad school, because the world population was still recovering from the Great Depression. Today, a little less than 7 out of 100 students who apply to Harvard are accepted - and the number of "valedictorians" in the college market outnumber the total applicant pool you had in those days. Similarly, there weren't 'established standards' to look at for grad school applicants - research, Putnam/IMC, scholarships (Goldwater, Churchill etc.), fellowships whatnot. You could really get in grad school nearly just because you applied, and had a close enough relationships to faculty staff who had connections.


**************off-topic*************

Anon111: I agree with you that one's grad school might not matter that much. There are people who have graduated from ETH, the grand Ecoles etc. but went on to do a lot. We don't have to look very far - Geim and Novoselov, who won the 2010 Nobel prize in Physics.

negru: Surely. Most of the time we can either test out of classes, or we already have humanities credits from high school. Moreover, if you are in a top school, it's likely that you already have many credits transferred over from high school. But that doesn't happen very often - not enough for there to have enough undergrads-who-had-done-ten-grad-courses to fill whole grad school departments! The people I know who have gone to grad school in Cambridge, Oxford, MIT, Imperial etc. were every bit human: they attended introductory courses in calculus, linear algebra and differential equations like almost everyone else and by no means managed to exempt themselves from humanities (many of them hated these as much). As a matter of fact, the Putnam fellows I know at some of these top grad schools did 4 years in college like everyone else, because they had to stay back and slowly finish off their humanities courses, like everyone else.

In fact even in a school like MIT where students have gone through a lot of high school preparation in university calculus and establish themselves as talented science students, most of the people actually start out with calculus or at most differential equations, introduction to classical mechanics and like everyone else among them, did their load of humanities stipulated in their university requirements.

Top schools are in fact more familiar with the genius kid who wants to get out of his humanities classes, and so have higher standards for people who want to do the same. Like Anonymous217, I would love to know who are these real people (enough to fill a whole department) who have done 10-12 courses in grad school level, not because I don't want to believe it, but because I want to believe it - it really interests me if I can have a credible source about them.

Lastly, to me, it sounds that my argument has sealed enough of a proof that most people who get in these top grad schools don't actually do 10-12 grad school courses in college, because the counterarguments that one can come up with against it require an immense suspense of belief: "you have to work very hard" and "you can become so good that you convince them to let you do whatever you want". This sounds more like Hermione from Harry Potter, or some other book than reality.

Eric Larson and Akhil Mathew are two examples of students who are likely to do 10-12 grad. courses in math during their undergrad. I hope I am not giving personal information since I'm only naming their names, and not linking to any of their information, and naming their names seems to be OK. If it isn't would a pf mentor please delete this. This is not my intention to delibrately give out personal information.

It's simple. The people who do 10-12 grad. classes are the ones who finish all of undergrad. math during high school. It's rare, but it happens more frequently these days. Take the two examples I gave. In fact, they've published research in high school and done grad. math in high school.