Doing research for a graduate student?

[anonymity]

Hello everyone,

This fall I am entering my sophomore year of my undergraduate, and am finally getting into some relatively advanced courses in my math and mechanical engineering curriculum. Math is where I am really ahead (compared to the offered curriculum). I will be taking two proof based classes (set theory and number theory), as well as partial differential equations and an applied linear algebra course. Math what I am most interested in (applied math, to be precise) -- far more so than ME.

My math classes are basically all junior level courses, and feel that I can finally be of some utility to a researcher (I also know java [and therefore C], python, and fortran -- as well as mathematica, and the very most basic things in matlab). The problem is that my university has a pretty small math department, and of the professors in the math department, only a handful of them are publishing new work currently. The school has about 15,000 students, but the math department is comprised of less than 15 professors, and only 5 or 6 of them are publishing -- one of whom is an applied mathematician.

My question is: a) is helping a grad student working on their PhD something that would be allowed as an alternative to researching for a proff (it does seem like it could be disallowed, because the PhD student is supposed to be creating something of their own as a prerequisite to entering academia...) and, b) would this be worth my time -- if i did find something to do for a grad student, could they even write a rec for grad school or an REU, would I be able to publish if I found something (for them) worth publishing, would I get as much out of it as from a prof, would their rec be as "valuable", etc

I want to get a taste for applied math research, but my options, it seems, are very limited..

 

[eri]

Yes, you could potentially work with a grad student instead of a professor. While you wouldn't be helping them with their PhD, sufficiently advanced grad students would help you come up with an original project and mentor you through it. However, they aren't getting paid for doing that, and it can take up a lot of their time, so you might only be able to interest one in helping you if they're interested in becoming a professor eventually. But if you've only got a few publishing professors, chances are good you don't have a very good math grad program, so there might not be any grad students in a position to mentor you. If you did find one, yes, they could write you a letter for an REU, or include you in a publication if the project got that far. I had 6 undergrads working for me at various times while I was in grad school; two ended up on one of my publications, and three more wrote papers for an undergraduate research journal.

 

[anonymity]

What constitutes a grad student being "sufficiently advanced" to take on such a mentoring project?

Also, how would you recommend approaching a grad student (who I very likely will not know)?

If it does come to this, I will likely just ask my adviser if there are any grad students pursuing anything in applied math, and then take it from there.

How did your 6 undergrads approach you (and also, though it appears that they were physics students, how advanced were they themselves in their studies?)?

 

[RandomGuy88]

Undergrads doing research typically do work with grad students. You don't really work with the professor. Meaning the professor is unlikely to ever be in lab with you or where ever you do your work. You typically talk to professors you are interested in and they will put you in contact with their grad students. You then work regularly with the grad student, usually by helping out on a small portion of their project. You might actually see the professor once a week or so, sometimes less than that.

It would be unwise to have a graduate student write you a letter of recommendation. Again, typically the professor will write the letter (maybe with input from the grad student). They are not in a position to judge your research potential.

So you should talk to professors that are doing work that interests you.

 

[eri]

When I was an undergrad, I did four internships with professors. My graduate university had a program where undergrads could sign up for research with professors, and my dissertation adviser agreed to do this only if I was actually the one in charge of the students. Four of my students came through this program, and another two when I worked as an adviser for our REU program (the same deal with my supervisor). I'd suggest talking to the professors; they'd know better than the grad students who might be ready for this and who might not be. While most of my students were juniors or just started senior year, one was a freshman. I taught them the skills they needed to make a contribution (basic programming, data analysis and reduction). Most of the work was computational and observational (astrophysics); I didn't work in a lab.