Choosing and Being Admitted to A Graduate Program in Physics

Should I Go to Graduate School?

If you want to spend your life working as a professional physicist (as opposed to using your physics as a springboard to another profession), graduate school is a necessity. The largest number of professional physicists are employed in academe, usually as college professors, but others find positions in industrial or national laboratories. The academic life has many attractions. To a larger extent than in almost any other job, you set not only your own hours, but your own goals and priorities. There are usually ample opportunities for travel. Most importantly, you are likely to actually enjoy what you do for a living.

The typical physics Ph.D. program lasts from 5 to 7 years, with no fixed time for completion of the degree. The first two years are primarily spent taking classes. Unlike undergraduate study, these classes are usually totally focused on physics, with no time to take courses in other fields (or, sometimes, no time to do anything at all except study). After the first or second year, it is usually necessary to pass a fairly tough qualifying examination. By the third year, most students are making the transition to research. Senior graduate students work closely with a professor on a theoretical or experimental project. The Ph.D. thesis is based on a serious piece of original work that typically takes one to two years to complete.

After receiving their Ph.D., most scientists spend 2-4 years as postdoctoral associates before landing a position as assistant professor (or equivalent positions in industrial or government laboratories). The next major hurdle is tenure: after 5-6 years the assistant professor is usually considered for promotion to associate professor with tenure. (Tenure in practice means that it is quite hard for a university to fire a faculty member, except in cases of serious malfeasance). It's a make-or-break situation; if you don't make tenure you have to leave and find another job, perhaps as an assistant professor at a different university.

The good news is that graduate school in physics is still a bargain. Almost all students are completely supported by a combination of teaching assistantships, research assistantships, and fellowships, while they complete the requirements for their degree. Nobody gets rich as a graduate student, but you usually don't go deep into debt or have to work an outside job, either. (The same cannot necessarily be said for graduate work in other fields.)

Many scientists look back on their graduate school years with pleasure, as a time of very hard work but also a time of relatively few responsibilities, when one was allowed to concentrate totally on research and learning.

Choosing and Applying to Graduate School

If you are planning on graduate school, you should start thinking about it seriously in the spring of your junior year, and really get into high gear in the fall of your senior year. You will need to take the Graduate Record Exam, fill out applications, and arrange to have letters of recommendation sent.

There is an important strategic decision to make regarding when to take the GRE. One school of thought says that you should register and take the GRE during the summer before your senior year. Although you can take it in the fall, the scores are not reported until the winter, and some departments will already have made decisions about financial aid by then. On the other hand, the GRE Physics Subject test is really a test of knowledge more than ability. You might therefore do better by postponing the exam as long as possible while you learn more; this might be particularly appropriate for students who have pushed a lot of their major courses (such as quantum mechanics and statistical mechanics) to their senior year. This is a place where it would be helpful to have done some research on the graduate schools you are applying to the previous spring, so that you know their requirements.

The completed applications are typically due in January of your senior year, and acceptances are usually sent out in March or April. Since there is generally a hefty application fee, and the applications take time to fill out, it is not appropriate to apply to dozens of schools. Most students take a "spreading the risk" approach: they apply to one or two schools that they think they have a reasonable chance of getting into, one or two extremely good "long shot" schools, and at least one "safety" that they feel confident of getting into.

When choosing among graduate schools, several factors need to be taken into account. First of all, try to make a realistic assessment of your own abilities and record. You want to go to the best graduate school that will admit you and in which you think you can do well, but you don't want to shoot so high that you fail to get in anywhere. Second, decide which field or fields interest you. Most physics department specialize in just two or three different areas, and it makes no sense to go to a very good university that has no faculty in the areas that excite you. Third, think about whether the geographical region and environment are important to you. Universities range from the highly urban (e.g., Penn, Columbia, Yale) to the fairly isolated and rural (Cornell, Illinois). This shouldn't be the most important criterion, but if you are going to spend up to 7 or 8 years in a place it would be nice to be reasonably happy during the times when you're not in the lab.

A useful reference is Graduate Programs in Physics, Astronomy, and Related Fields, published annually by the American Institute of Physics. This book provides detailed, quantitative information about most physics departments in the country, including research specialties of the faculty and some information on admissions standards. Most departments also have Web pages with varying amounts of information. However, your most useful resource is likely to be the physics faculty at Penn. A physics professor, particularly one in the field that you are interested in, can tell you which schools are presently "hot" and which are not, who you should try to work with in different departments, and in general can give you information about the reputations of different programs that you will not find in any book.

Finally, you should make a serious effort to visit the colleges that interest you. Some schools (such as Penn) have formal visiting programs for potential graduate students (usually in mid-spring, around the time of spring break); all of them should welcome you and be willing to show you around a bit. You can get a very good idea about whether a place attracts you by wandering around, talking to professors and graduate students, and generally absorbing the atmosphere of the place. If travel expenses are a problem, ask the university you are visiting if they can pay part of your expenses; the answer might be either yes or no, but they won't be offended.

Incidentally, although Penn has a first-rate graduate program and we are proud of our undergraduate program, we do not in general encourage undergraduates at Penn to enter our Ph.D. program. There are real benefits to moving to a different institution where the culture and academic priorities are different, and where you will be exposed to a different set of professors. Exceptions are made for first-class undergraduates who have good personal reasons for needing to stay in the Philadelphia area.

What Does it Take to Get Into Graduate School?

A common question among junior and senior physics majors is something like, "My grades are so-so but I did fairly well on the GRE. Can I get into a good graduate school?" Unfortunately, the literature provided by most graduate schools tends to be somewhat evasive on the subject of exactly what the requirements are. Accordingly, in the fall of 1995 we polled a number of graduate schools and asked them what they looked at and what scores were required. 52 schools were polled, and responses were received from 18. That isn't very many, but the responses were sufficiently uniform to believe that they give a reasonable guide to expectations nationwide. The schools responding ranged from elite private institutions to lesser-known (but still respectable) state schools.

In general, most schools require applicants to submit the following: Graduate Record Examination (GRE), letters of recommendation, transcripts, a personal statement, and for foreign students the Test of English as a Foreign Language (TOEFL) and possible the Test of Spoken English (TSE). Different universities varied widely in which factors they considered most important. However, most said that they looked at the application package as a whole, so that a deficiency in one area could be rectified by superlative performance in another area. For example, a student with poor physics grades might make it into a good grad school if the GRE scores were exceptionally high, or if the letters of recommendation were really good.

It is helpful to know a little about how the admissions process actually works. In some institutions, there is a central admissions office that screens applications. For example, they might throw out all the applications that were incomplete, or for which the GRE scores were below a certain level. However, in almost all universities the final decisions on graduate admissions are made by small committees of professors (and sometimes students) in the departments in question. These committees are in general not bound by any rigid quantitative requirements, and as mentioned above they generally take the attitude that they are evaluating applications as a whole. What this means is that the selection criteria can be quite idiosyncratic, and can vary wildly not only from institution to institution but even from year to year within the same institution.

A summary of typical results in different categories follows:

  1. Graduate Record Exam:

    This is broken down into the Verbal, Quantitative, and Analytical sections, which are taken by everyone who takes the GRE, and the Subject section (which should be Physics for students applying to Physics graduate programs). The conventional wisdom is that it is not really possible to study for the Verbal, Quantitative, and Analytical sections, which are supposed to measure ability rather than knowledge. It definitely is possible to study for the Physics exam, which measures knowledge. In the following discussion it should be noted that some schools apply different criteria for foreign and U.S. students, with the cutoff or average set about 10 percentile points higher for foreign students than U.S. students.

    a. GRE Verbal:This is probably the least important part of the exam for potential science graduate students. Typical average scores for students admitted to graduate school are in the range of 500-700 (out of a maximum of 800), corresponding to the 60'th to 95'th percentile range. Some schools will accept scores as low as 400 (25'th percentile) while others may have a minimum requirement as high as 640 (90'th percentile). Some allowance is also often made for students who are not native English speakers.

    b. GRE Quantitative:Most schools require scores in the range of 700-800 out of 800, i.e., greater than 80'th percentile. (Don't forget that these percentiles include students applying to non-scientific graduate programs such as Literature; such students might be expected to be somewhat weaker in quantitative skills than Physics majors). There is a weak correlation between the quality of the school and the level of performance required, with the "elite" schools generally admitting students with average scores on the level of 760 or above (94'th percentile) and smaller schools admitting students with scores as low as 650 (70'th percentile).

    c. GRE Analytical: Effective Fall 2002, this section of the GRE has been replaced with a writing score based on a 5 point scale.

    d. GRE Physics Subject Test:This is the area in which the averages vary the most, with no particular correlation between the score required and the quality of the school. The typical school expects scores between 600 and 900 out of 1000, corresponding to the 40'th and 90'th percentile. (Keep in mind that almost nobody besides applicants to physics graduate programs takes this test). However, some universities have a minimum requirement as high as 900, while other quite good schools will accept students with scores as low as 500 (15'th percentile), apparently on the theory that the test isn't all that useful.

  2. Test of English as a Foreign Language (TOEFL):

    This test is required by essentially all graduate programs for students who are not native English speakers. Almost all programs have a minimum requirement, which ranges from 550 to 620.

  3. Test of Spoken English (TSE):

    This test is fairly new, and is not required or even used by most programs. Furthermore, the grading scale has just been changed. Not enough information is available to evaluate average requirements for the schools that do expect it.

  4. Grade Point Average:

    Most schools surveyed reported that the average GPA of admitted students was in the range 3.20-3.40. Although most schools did not have minimum requirements, some required a minimum GPA of 3.0-3.5. In addition, the reputation of your undergraduate institution makes a difference; a GPA of 3.5 from a college believed to have lax grading standards might not be as good as a GPA of 3.0 from a school widely perceived as "tough."

  5. Quality of the Undergraduate Institution:

    A few schools mentioned this as important, while others didn't say anything about it. Still, it probably plays a role even if nobody admits it. If you got reasonable grades, etc., from a university famous for its physics program, you have a good chance of getting into a first-class graduate program. If you are coming from a tiny liberal arts college that nobody has ever heard of, you had better have great GRE scores and strong letters of recommendation.

  6. Undergraduate Research and Letters of Recommendation:

    These two really go together. A letter from a professor who taught a course you took is nice, but not nearly as helpful as a glowing letter from someone who supervised an undergraduate research project. The relative importance of research, letters, and other factors varied widely from school to school, with some schools putting them at the top of the list (or even saying that undergraduate research is an absolute requirement for admission) and others ranking them far below "objective" measures such as GRE and grades.

  7. Personal Statement:

    Most applications have a place for you to write a short essay about your personal goals. Some schools ignore what you write, while others pay a great deal of attention. The schools that do pay attention want to have an idea of what your goals are, why you want to go to graduate school and why you want to go to their school in particular. Hint: "I want to go to graduate school because I can't think of anything else to do, and I had heard of your school," doesn't cut it!

  8. Discrimination and Affirmative Action:

    All U.S. universities are bound by law not to discriminate on the basis of race, gender, religion, ethnic origin, or other such factors. (It is legal to make English language proficiency a factor in admissions).

  9. Other Factors:

    Extra-curricular activities (drama, community service, sports, etc.) are wonderful things to do, but they don't seem to be the least bit useful in getting into a good physics graduate program.