So you want to be a scientist [Physicist's perspective]

Notes provided by forum experts as FAQ and reference material. This is not a wiki, but rather a small repository for material relevant to common discussions.

So you want to be a scientist [Physicist's perspective]

Postby Lincoln on March 13th, 2008, 7:57 am 

I am often asked for information on how to become a scientist. You should realize that there are definite field-dependent differences and that this is a physics-specific post. But many of the lessons are universal. Caveat emptor and all that.

If you'd like to be a professional physics researcher, there are some important lessons I'd like to share. Note that several of these lessons were learned to late for me and to my detriment.

If you are a high school kid, I advise taking as much math as you can, preferably through calculus. If your school doesn't offer calculus, it's not a huge problem. It just helps. You obviously should take physics too. I also suggest any computer programming you can get (and not just the whole web browsing thing that substitutes for intelligent computer use nowadays.) Learning to type well is useful, so a keyboarding class is a good idea. Finally, it is very helpful to be able to write well. I would take your English composition classes rather seriously. And, of course, if you are a non-American/Brit/Aussie/what-have-you, it's extremely helpful to be fluent in English.

Once you get to the university level, things are a little different. Note that the university systems are different in different countries and what I'm saying is very US-centric.

At the university, you need to be a physics major. You should note that many people who start as a physics major do not go on to research, so it's probably smart to double major, with the other major being something that can lead to a useful career (computers, engineering, etc.)

But in college, there are a few things you need to do to keep on the research path. First is to do well in your classes. That means don't overload too much. Grades matter a lot. Do all the required physics and a few electives. Take a lot of math, but unless you intend to be a theoretical physicist, don't overdo it on the math. Mathematicians think differently than physicists and the upper-level math classes can actually hurt, unless you take great care to not get too hung up in the whole "proof" thing. Math is a tool. View it that way.

One thing that is often overlooked by undergraduates is the need to do some simple research as an undergraduate. This doesn't mean (usually) doing something ground-breaking. It means working with an established research group to start to get a sense of the techniques and culture. Plus, you can make contacts and will establish a reputation that can be mentioned in the all-important letters of recommendations to get into graduate school. This research can be done during the summer or during the school year, although during the school year is pretty hard. You can find research opportunities by talking to your profs.

In the US, there are many tiers of colleges and the differences are not as rigid as in other countries. Should you go to a college or a university? What's the difference? Most simply, colleges tend to be better places to learn in a classroom, with smaller classes and more engaged faculty. Universities tend to be bigger and more impersonal, yet usually have far more research opportunities. Plus the more famous and politically connected tend to be at universities. So I lean towards suggesting you attend a university. The famous ones are best (CalTech, Columbia, Berkeley, MIT, etc.) but your local state university (especially the leading one in your state) is probably fine too.

Once you have a bachelor's in physics, you're not very useful, except to teach high school and say "You want fries with that?" (Hence the advice to double major.) So to go on in physics requires graduate school.

My most important advice in this thread is the following. When you go to graduate school, go to the most prestigious, snotty, high-end graduate school you can get into. The academic world is very hierarchical and going to graduate school at a high end place gets you "into the club." Plus such schools tend to attract the best and most-connected faculty. A Ph.D. from such a school is VERY helpful for the rest of your career.

To get into graduate school, you need good grades, good physics GRE scores and letters of recommendation that benefit from having tales of your research experience. Another useful piece of advice. It is harder to be a theorist than an experimentalist. Thus if you say you want to be a theorist in your letters, they will be judged more harshly than if you want to do experiments. This makes sense, as there are only a handful of theory jobs in the country per year. You really do have to be a hotshot to survive. If you want to be an experimentalist, stress it.

In your application letters, you probably shouldn't mention professors by name unless you really know what you're doing. But you can say stuff like, "I am uncertain at this time about my area of specialty, but your research groups of X, Y & Z are all doing exciting work by excellent scientists and I enjoyed reading their recent publications." [But you better have then read them if they interview you.] If you do know what you are doing, contacting a research group directly is helpful.

When you are in graduate school, you will have to teach the first couple of years. Unlike college, you get paid (and pretty well...say $20k/year) to go to graduate school. After the first two years, you should get a research assistantship and not have to teach anymore. You will take classes for two or so years, and start doing research. You will be in graduate school for probably 6 years total. It starts to become important what topics you pick to study at this point. If you have the right personality to fight for recognition among other forceful people, grab a sexy research topic and run with it. If your personality isn't so competitive, it's better to pick a "tier 2" topic. The most important thing to do in graduate school is to show initiative.

After graduate school, if you want to continue the research path, the next step is a postdoctoral position, which will end up being about 6-ish years long and pays about $45k/year. If you're fed up with research, you might decide to teach at a small college. Then you probably should not take a postdoctoral position, but rather a temporary professor job. This pays more poorly, but establishes teaching credentials. Finally, if you want to give up on science completely, you can join the real world and make big bucks.

After your postdoctoral stint, you will search for a professor job. This is VERY HARD and VERY DEPRESSING to most people. In my own field, there are about 10 jobs per year in the entire US and about 100 applicants. Not all the applicants are top-shelf. But there are 5-10 applicants that get all the interviews. If you're one of them, great. If not, then you may get 1-2 interviews. Ultimately about 1/3-1/2 of the postdocs eventually become profs.

If you get the prof job, then you are an Assistant professor for about 6 years, after which you get tenure. You will be about 40 years old at this point in your life. After that, there is the Associate professor and full professor level, but if you get there, you won't need this post.
User avatar
Resident Expert
Posts: 10729
Joined: 29 Dec 2005
Location: Deep in a lab...

Return to Expert Notes

Who is online

Users browsing this forum: No registered users and 1 guest