What Should I Do - Research

Post Reply
PhysRiches
Posts: 6
Joined: Sun Mar 22, 2020 3:56 am

What Should I Do - Research

Post by PhysRiches » Sun Mar 22, 2020 4:17 am

Context
Physics and Math major (3.9 GPA)
2 years physics research with organic polymers - experiment (home uni)
1 REU - Diffraction CME type of stuff
Will graduate May 2021

I will continue my organic polymer research for another year most likely - results unknown


My issue is that I don’t know what kind of research I want to do in the future. Or what I can do. I have been enjoying my computational physics classes and I always enjoy learning the theory developed in the fields I research - so I think I would prefer a computational/mathematical/ theory based physics, however that’s a lot of topics.

Condensed Matter, Astrophysics, High energy, mathematical physics, biophysics all have very interesting questions and subject matter - I can’t do all of them but I want to find out which one interests me most before I dedicate a whole career to it

The question- should I try to get into a PhD program without a solid understanding of what I want to research or should I go into a masters and try a different field to find out what I want, then go to a PhD?

If I choose masters how will it affect my chances at a PhD?
If Choose PhD, there is generally a requirement to choose your mentor 2 years in or sooner, I will I figure out what I want in that time if I don’t get to do any actual research until I join a group??

I’ll take suggestions about what I can do in the next few months to narrow down my possibilities.or maybe I’m focused on the wrong question?

User avatar
Nishikata
Posts: 212
Joined: Sun Sep 23, 2018 6:37 am

Re: What Should I Do - Research

Post by Nishikata » Sun Mar 22, 2020 11:50 am

The only way to know what you want to do, is by trying as many options as you can.

A warning from a CMT/computational guy.
It is nice and fun to learn all the established theories.
It is totally different game to try and develop a theory out of nothing. It is not easy at all.

If you have tried polymer for 2 years, maybe you should try something else.

You cannot find out what you want to do without actually trying things.
Some subfields are very shiny from outside, but are actually horrible to do as a topic.

Remember, you are supposed to produce new stuff in a phd.
The shiny subfield may already be saturated so that finding new stuff is very difficult.

So take opportunities you have now. If you see a project and the professor is good, just do it. At worst, you can cross out that one field if you ultimately do not like it.

jabennett2194
Posts: 118
Joined: Tue Jun 12, 2018 6:54 pm

Re: What Should I Do - Research

Post by jabennett2194 » Sun Mar 22, 2020 11:58 am

PhysRiches wrote:
Sun Mar 22, 2020 4:17 am

The question- should I try to get into a PhD program without a solid understanding of what I want to research or should I go into a masters and try a different field to find out what I want, then go to a PhD?
Never do a masters (in the US) unless:
+ you want a certification like for business, accounting, teaching, etc. or
+ your employer is paying for it.

In the US you don't pay tuition in STEM PhDs and you get paid to be a teaching assistant.

PhD: free education and paid roughly 20,000 to 35,000 a year
Masters: you pay upwards of 50,000 a year and will not be guaranteed a teaching assistantship unless it is a very well-funded department

*This is different in Canada or the EU (if you're an EU citizen). I dont have knowledge of how masters funding working in Asia.*

geekusprimus
Posts: 62
Joined: Sat Jun 08, 2019 1:10 pm

Re: What Should I Do - Research

Post by geekusprimus » Sun Mar 22, 2020 4:15 pm

PhysRiches wrote:
Sun Mar 22, 2020 4:17 am
If Choose PhD, there is generally a requirement to choose your mentor 2 years in or sooner, I will I figure out what I want in that time if I don’t get to do any actual research until I join a group??
This depends on the school. In my personal research while looking for schools, I've seen and heard of three kinds:
  • Some schools have an early publishing requirement. If you don't immediately know who you want to work with and get started in the first year or so, you'll leave after two or three years, probably with a master's degree
  • Some schools intentionally admit more students than they have space for. Funding becomes extremely competitive, so if you don't jump on the research train fast enough, there won't be anyone to fund you and you'll leave after two or three years, probably with a master's degree. Sometimes even really prestigious schools are famous for this. In particular, I've heard this isn't terribly uncommon at MIT.
  • Some schools make you fulfill a certain level of research-oriented coursework before you're allowed to pick a research group. For some schools, this is just a bunch of seminars about the different groups. Others, like Princeton, require students to do a "pre-thesis" with a prospective advisor before you can start working with them. If you don't know what sort of research you're interested in, these are probably the safest schools to look for, because many of them expect you to take some time to explore before you start working.

PhysRiches
Posts: 6
Joined: Sun Mar 22, 2020 3:56 am

Re: What Should I Do - Research

Post by PhysRiches » Sun Mar 22, 2020 10:34 pm

jabennett2194 wrote:
Sun Mar 22, 2020 11:58 am
PhysRiches wrote:
Sun Mar 22, 2020 4:17 am
In the US you don't pay tuition in STEM PhDs and you get paid to be a teaching assistant.

PhD: free education and paid roughly 20,000 to 35,000 a year
Masters: you pay upwards of 50,000 a year and will not be guaranteed a teaching assistantship unless it is a very well-funded department

*This is different in Canada or the EU (if you're an EU citizen). I dont have knowledge of how masters funding working in Asia.*
There are some master's programs that will fund the majority of your time ( similar to PhD but with less money). These are generally smaller schools where the highest degree offered is a masters and will pay for you to be a TA in the first/second year with research opportunities. If there is a good enough offer from one of these smaller schools, and they have some interesting research, would it hurt my chances at a PhD because it appears that I couldn't succeed in a PhD immediately after undergrad??
Last edited by PhysRiches on Sun Mar 22, 2020 10:47 pm, edited 1 time in total.

PhysRiches
Posts: 6
Joined: Sun Mar 22, 2020 3:56 am

Re: What Should I Do - Research

Post by PhysRiches » Sun Mar 22, 2020 10:37 pm

geekusprimus wrote:
Sun Mar 22, 2020 4:15 pm
PhysRiches wrote:
Sun Mar 22, 2020 4:17 am
If Choose PhD, there is generally a requirement to choose your mentor 2 years in or sooner, I will I figure out what I want in that time if I don’t get to do any actual research until I join a group??
This depends on the school. In my personal research while looking for schools, I've seen and heard of three kinds:
  • Some schools make you fulfill a certain level of research-oriented coursework before you're allowed to pick a research group. For some schools, this is just a bunch of seminars about the different groups. Others, like Princeton, require students to do a "pre-thesis" with a prospective advisor before you can start working with them. If you don't know what sort of research you're interested in, these are probably the safest schools to look for, because many of them expect you to take some time to explore before you start working.
Do you know any of these schools that would be worth looking into. I do not go to a top tier institution nor do I have the research experience to go to t-25 (probably). How would it look on an application if I project the attitude/thought/feeling that...
'I love to learn physics but I after 3 years of some research I have no clue what I want to do'

PhysRiches
Posts: 6
Joined: Sun Mar 22, 2020 3:56 am

Re: What Should I Do - Research

Post by PhysRiches » Sun Mar 22, 2020 10:44 pm

Nishikata wrote:
Sun Mar 22, 2020 11:50 am
The only way to know what you want to do, is by trying as many options as you can.

A warning from a CMT/computational guy.
It is nice and fun to learn all the established theories.
It is totally different game to try and develop a theory out of nothing. It is not easy at all.

If you have tried polymer for 2 years, maybe you should try something else.

You cannot find out what you want to do without actually trying things.
Some subfields are very shiny from outside, but are actually horrible to do as a topic.

Remember, you are supposed to produce new stuff in a phd.
The shiny subfield may already be saturated so that finding new stuff is very difficult.

So take opportunities you have now. If you see a project and the professor is good, just do it. At worst, you can cross out that one field if you ultimately do not like it.
1) Is it possible to pursue two unrelated research topics as and undergraduate, how will professors feel about me trying to split the time? I will not be asking to be paid. Regardless, I do not want to half-ass both.

2) You mention that CMT is challenging. For you, is it a rewarding challenge, do you enjoy 'the grind' of trying to develop a 'theory?'
How did you decide it was the the field that interested you most?
( This is a lot to ask but I'm really curious about what it's like, day to day and big picture. What I should realistically expect)

PhysRiches
Posts: 6
Joined: Sun Mar 22, 2020 3:56 am

Re: What Should I Do - Research

Post by PhysRiches » Sun Mar 22, 2020 10:45 pm

Thank you all for replying. This has been helpful (and somewhat cathartic) and I know it can take some time to formulate a response, so Thank you!

User avatar
Nishikata
Posts: 212
Joined: Sun Sep 23, 2018 6:37 am

Re: What Should I Do - Research

Post by Nishikata » Mon Mar 23, 2020 1:50 am

PhysRiches wrote:
Sun Mar 22, 2020 10:44 pm
1) Is it possible to pursue two unrelated research topics as and undergraduate, how will professors feel about me trying to split the time? I will not be asking to be paid. Regardless, I do not want to half-ass both.
You have 1 year, right? I suggest do 1 at each semesters. You can continue to rollover the first one to the second semester while also doing another, if you find it interesting.

In fact, you do not really need to have approval from one for concurrently doing another. It is your time and energy, You just need to deliver their deliverables consistently. Half-ass or not, that's for you to decide how much time you want to put in.

This means you need to sacrifice sleep or social time (most likely both)
2) You mention that CMT is challenging. For you, is it a rewarding challenge, do you enjoy 'the grind' of trying to develop a 'theory?'
How did you decide it was the the field that interested you most?
( This is a lot to ask but I'm really curious about what it's like, day to day and big picture. What I should realistically expect)

CMT is simply the field that got me where I wanted to go, as my previous research topics as undergraduate were in CMT/CME field. (I did three)
I have not decided that it is the one. However, I have not done enough on other fields to be confident to jump and have them convinced to accept me. I am still taking as many classes as possible in other topics to keep myself open for change

In CMT, there have been so many theories with incremental improvements.
These leave little or no room for direct improvement. The most recent theories in CMT have to incorporate some super-complicated mathematical concepts that make you wonder if these are actually more suited for mathematicians than us, or whether it is even worth doing at all.

If you look at the list of research publications of CMT professors in top universities, you either find:
  • they have somehow moved to soft matter/biophysics research
  • they are looking at black holes
  • they have somehow started machine-learning / quantum information research
  • their publications are a bunch of materials/phenomenological paper that have little in common
this one means their research is just helping the experimentalists explain the results with their current theory, hence it is not a development.

or
  • they last updated their page in 2002 :)

geekusprimus
Posts: 62
Joined: Sat Jun 08, 2019 1:10 pm

Re: What Should I Do - Research

Post by geekusprimus » Mon Mar 23, 2020 2:53 am

PhysRiches wrote:
Sun Mar 22, 2020 10:37 pm
Do you know any of these schools that would be worth looking into. I do not go to a top tier institution nor do I have the research experience to go to t-25 (probably). How would it look on an application if I project the attitude/thought/feeling that...
'I love to learn physics but I after 3 years of some research I have no clue what I want to do'
I wouldn't word your interests as "I have no clue what I want to do." Emphasize your love of physics and the kinds of problems you really like solving, then say something a little more diplomatic like, "I am undecided on my specific field of interest, but I think that Dr. X's work on topological insulators is fascinating for <insert reasons>, and I am equally intrigued by Dr. Y's research into nonlinear optics because <insert more reasons>." Find a couple professors whose work looks interesting, reach out to them to talk about it, and see if it really does sound interesting. If it does, mention a couple of them in your SOP. Since you're undecided, avoid the temptation to say you're interested in professors in HEP-theory or quantum gravity. It's not that you can't do those things, but they're extremely competitive fields that favor applicants with top-notch scores, good connections, and prior research experience. As a general rule, you'll be a lot safer if you stick closer to condensed matter or AMO experiment, as both are large fields with lots of funding. Look for big, well-funded schools with big, well-funded physics departments. The more prestigious ones (UCB, UIUC, Maryland, Michigan, etc.) will definitely favor students with prior research experience, but places like Arizona, Texas A&M, Virginia Tech, and Notre Dame are some good mid-tier physics schools that might be worth a look.

As for schools with specific courses on research offerings, a couple that I researched and applied to were Georgia Tech and Penn State. GT has a "special problems" class that is kind of like Princeton's "pre-thesis" requirement. Penn State is probably harder to get into (especially if you apply for quantum gravity), but they have a mandatory research seminar for new students to help them get exposed to all the different research areas and make some decisions. Other than that, I would look at the same kinds of schools that are (probably) going to be most favorable to undecided students: big, well-funded schools with big, well-funded physics departments.

One final note: don't be afraid to apply to some top-25 schools, especially if you have good recommendation letters and solid grades/GRE scores. I'm happy with the schools I got into, but if I could do it over again, I would apply to a couple of the super-prestigious schools in my field just to see if I had a chance.

jabennett2194
Posts: 118
Joined: Tue Jun 12, 2018 6:54 pm

Re: What Should I Do - Research

Post by jabennett2194 » Mon Mar 23, 2020 6:18 am

PhysRiches wrote:
Sun Mar 22, 2020 10:34 pm

There are some master's programs that will fund the majority of your time ( similar to PhD but with less money). These are generally smaller schools where the highest degree offered is a masters and will pay for you to be a TA in the first/second year with research opportunities. If there is a good enough offer from one of these smaller schools, and they have some interesting research, would it hurt my chances at a PhD because it appears that I couldn't succeed in a PhD immediately after undergrad??
Huh! Didn't know there were many of these! Off the top of my head I can only think of 2.
I still think it would be wise to go straight the PhD to not run the risk of the PhD program you get into not viewing your courses, quals, etc. as rigourous enough and making you take their anyway!
I don't think it would hurt your chances at all, no! I just think if your goal is a PhD, don't do it in 2 steps. (Unless you want to maybe go abroad for your PhD! Since they don't mush the Masters and PhD into one package like we do, they require a Masters for a PhD.)
Less money is non-trivial, especially if doing the masters and PhD separately may take you longer. 5-7 years of making 20k a year is a hurdle already ;)



Post Reply