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!