Expert: Daniel Mazur - 3/24/2009

Question
Hello Daniel,

I am 26, and live in New Delhi, India. I have a Bachelor degree (4 years) in Computer Science. I have been interested in Physics and Astronomy for quite some time now and wish to have a PhD in it from the US. In the mean while I have been studying Physics on my own (Feynman lectures and Resnick Halliday.) I have asked a few graduate schools in the US about my eligibility to apply and they say I can get in as long as I have a good score in GRE and TOEFL and can get recommendation letters from Physics scholars/professors/researchers to send along with my application.

The trouble is that I don’t know any Physics professors as I never took a course in Physics (though I studied Physics upto high school). My questions from you are these.

1) Is it possible for me to get into a graduate school without having any letters of recommendation or is it a standard that the schools simply won’t compromise upon?

2) And, is it okay to get into just any graduate school in US, whether or not it scores a good rank in a list. Will I be able to find a job/post doc position afterwards? I am very confused which schools to apply to, as I am not sure if the big ones will admit me (since I lack a degree in Physics) and if with the small ones I will have decent career opportunities afterwards.

I just wish to be able to do Physics for the rest of my life. I don’t much care about the money but doing what makes me really-really happy (just like yourself, as you said in some of your answers on the site.) But I need a promise of a job even if it pays little. And I am not too sure because of this economy gloom at the moment. For me there is nothing nobler a profession but being a researcher. But need to feed a family too. You understand the troubles of this aspiring physicist, don’t you? :-)

Thanks for reading my letter and helping me out. I appreciate it very much. Please take your time.

Best regards,
Siddharth
New Delhi,
India  

Answer
Hello Siddharth,

thank you for asking, I appreciate your writing a little explanatory part about your background and motivations. You summarized well what I have been writing here about our profession: We're not getting really rich (although if you are successful even riches arrive), but we're earning enough to raise a family in fair conditions, doing what we love for a job, and having a fairly steady chance of finding a job...

Well, yes, your chances of getting a job are somewhat affected by economical climate, but not as dramatically as e.g. in car industry :-). In the West there are steadily more scientific instruments available than there are people to use them. This fact alone gives chance to a creative scientist to come with a proposal to make use of the facility, while providing him and other people with jobs. Governments of many countries (USA prominent among them) realize that world-class science is necessary for the country to have a political impact on the world as a whole, so the funds never quite dry out. One has to learn how to take benefit from it - I am still learning in that department.

If you have a family, it is essential that your family accepts that for you to have not only "a job", but an interesting and adequately paid job, they will need to move around the world with you every few years until you are hired for a permanent position. While time-limited contracts of 2-5 years are fairly well available to us (it is the category of post-docs, assistant professors, visiting scientists etc.), permanent positions are at high demand and low abundance. It is customary for a physicists to have a couple of work experiences after Ph.D. in very different parts of the globe before one is considered "matured" to be a professor, permanent staff member of a research institute or similar.

Now to your questions.
1) It is virtually IMPOSSIBLE to get accepted to graduate school in the USA without any letters of recommendation. They are a standard part of the application and, although the target departments rarely pay much attention to WHAT is written in the letters, they need to just have someone to act as a "warrant" (albeit on an honor basis) for the abilities you claim to have and your academic title. I was, in my time, in a different situation from yours as I got a physics Masters and most of my teachers were physicists. Still, when it came to the letters of recommendation, I had to choose the right "warrants" and nudge them a bit about it. I suggest that you ask your advisor(s) about their contacts to physics departments, simply to find somebody, anybody in physics, who could write that letter (those letters) for you. Your advisor will then warrant your qualities to the person writing the letter of recommendation, the willing person will warrant you to the target University. I dare say that even one single letter from a physics faculty will suffice as you are a computer science master; plus two-three from your CS faculty. Yes, try to overcome the "quality" with "quantity" :-)), but one physics faculty's favor will be needed at the very least.

2) Before I "really" answer, let me write about other related matters. You have a handicap as you intend to switch from CS to physics. You need to ask each and every department you apply to, what they (not just me) think about your situation before applying. It is because the faculty are simply people (intelligent and all that, but people) and there is no "standard" answer to your question. You need to show your genuine enthusiasm for physics. You need to tell them you accept that it may take more years for you to get the PhD degree; you accept that you will need to take a load of undergraduate courses to equalize. You need to make it clear that you are willing to do whatever it takes - not being overzealous nor hesitate too much at the same time. It's a bit of psychology - just like later, when you are looking for a job. They might suggest to you that you start from scratch - with freshman physics year to go for a Bc. You definitely do not want to do that... They may then suggest that they will accept as much as possible from your CS curriculum, but that you enroll in Physics Master's program. Not what you want, but I think you should consider this option, even if you really do not want to "downgrade". It may be that from a CS Master to a Physics Ph.D. it will be too big of a jump and taking "officially" 2-3 years to "catch up" withthe physics will make things somewhat less stressful for you and less tight in terms of the exams. Let me explain.

After you enroll a Masters' or Doctorate program (based on your application, undergraduate grades, GRE, TOEFL scores...) in USA you become a student, but not yet a "degree Candidate". In either of the programs you will be required to take AND PASS a Qualifying Exam (QE), after (and ONLY after passing) which you will get on the track to your degree. You will be given 3-4 attempts at the exam over 4 first semesters of your program and if you fail on your last (3rd or 4th attempt) it is over. Two years wasted... I can easily imagine you are a better student than I was (most of MY current students are, I fear:-))), but still. After an M.S. in Physics in Europe I took a shot at the QE 3 times (!) before I passed. At my university, the Illinois Institute of Technology (yes, IIT, just like the Indian Institute of Technology, which may be why I had many schoolmates of Indian descent:-)))), there was one common QE for both M.S and Ph.D. levels. This exam covered all physics and maths from the undergraduate physics curriculum, the pass mark on M.S. level was about 50% and on Ph.D. level was around 70%. Do you think you'd get past 70% mark in less than 2 years?

Here's a check: you say that you've been studying from Halliday-Resnick-Walker. In my experience 33-50% of the QE is from there, so if you can solve 100% of HRW exercises on the fly (without having to check any equations...), you may just about qualify for M.S. level. There is, however, much more to undergraduate physics than HRW textbook - and than Feynman-Leighton-Sands' series, which covers about the same topics, gives you more physical insight, but does not do better in teaching you the problem-solving attitude.

What more? First there is the Maths for physicists and I have no idea, how much of it the CS curriculum contained. My fleeing experience with CS maths was the discrete maths, which goes completely amiss for a physicist. Physicists are taught and tested on derivatives, integrals, series and sums, linear differential equations (ordinary and several special partial), calculus in complex plane, vector and matrix algebra. You must know how to solve a Laplace equation or a wave equation problem (for example) to pass the QE at the Ph.D. level. Then there is the physics unmentioned by HRW: Theoretical mechanics - teaches you to solve much more advanced mechanical (and other) problems; Electrodynamics - problems "similar" to HRW, but using the advanced maths; Thermodynamics and statistical mechanics - they become one subject, rigorous, very much disliked by students (including myself) but really useful for your scientific work in ANY field of physics; and finally: physical data acquisition and analysis. For the last part physics students have general "labs" and then (3rd and 4th year) specialized labs, which I think will be mandatory for you to take and pass. As a CS, you couldn't have taken many physics labs, could you?

If you decide for the transition to physics, it will be demanding and it will take time - no use pretending otherwise. If you phone some professors in the USA and show that you are willing to accept that, you will be accepted somewhere. Finally I get to your question: Does it make a difference, where you do your Ph.D? Yes, it does. There is a group of "high-opinionated" universities on the east coast of the States, whose degrees are the most prestigious, who earn far above average in their first post-doc (first job after Ph.D.) and about 10% more long-term than graduates of other universities. Unfortunately for you, they are flooded with applications every year and are even more so now with the world economical crisis: It is a curious but logical effect that, when the country economy goes south, people tend to extend their studies (Bc. graduates decide to take on M. rather than leave the school) and the laid-off people tend to come back to the schools to further their education to gain a competitive advantage (better salary or at least higher probability of job offers). All this is against you now and I can only sigh in relief that I myself escaped this last crisis by about 18 months...

In short, I suggest that you send out a few applications to all sorts of universities - not just the top ones, cover the spectrum well while avoiding places that you really don't find interesting (like: University of Minnesota? Utah? They are O.K., if they do research you'd be particularly interested in, but otherwise try to stick to big cities - Chicago and Dallas are O.K. - and the east or west coast). Send out not more than a dozen or so applications to well-chosen (!) universities. Take time to contact some faculty member of each school you consider and preferably call them (at a pre-arranged time) - with VoIP it's not too expensive. Ask about the assistantships (way to work for your life expenses at the department) and especially about the deadlines. Get acquainted with research subjects of the faculty of the Universities you are considering and try to find a genuine interest in at least one at each Uni that you send application to.

As a final word I should suggest again that you accept the idea of re-doing the Masters level, this time in Physics. If you do, you just need to find a department that will offer you an assistantship during the M.S. curriculum. Assistantship is money and money is life... Note that after you finish a physics M.S. at ANY university (even one in the mddle of a corn field), you can then choose again, where you would like to do your Ph.D. The economical times may become better (economy-wise) by then too... And, if you are searching for a Ph.D. in USA from WITHIN the USA, it is a very different story (i.e. easier) than doing the same from Europe or India. Well, if you are confident enough, you can try taking a Ph.D. right away and see, how it goes, but knowing what I know now, I would play it safe, if I were you.

Still, I really applaud your desire to later have a job that would not only sustain you and your family, but that you would really like doing. For me this makes 2/3 of my motivation: that I love, what I do and that I see purpose in sacrificing a little extra time and effort. I may add that a scientist also has a much more flexible schedule than e.g. a factory or office or designer's studio worker. We are evaluated upon the work done, not whether we did it in the usual working hours or at 3am. I like that part quite a lot... ;-)

Well, have a good think about what I wrote... It is not a universal truth, but I tried to convey, what I think is the best approach. As I might have been somewhat incoherent (it is late past midnight now), please feel free to ask further, if you like. I will be away for 3-4 days from now but after that I will gladly reply again.

Good luck!
Daniel