- Avoid taking upper-level E&M, quantum mechanics, and thermodynamics. A 3.7 with those classes beats a 4.0 without them every time. You can also really go for broke and enroll in them and then withdraw. Admissions committees really do take the time to try and figure out not only your grades but what upper-level physics classes you have taken.
- Avoid research opportunities like the plague. A PhD is really an apprenticeship in research. If you have never done any real research, there is a chance that (a) you are bad at it or (b) you will hate it. Programs have no desire to admit students, support them for a couple years, and then lose them when they decide that they would rather be pastry chefs than researchers.
- Take the general GRE without studying at all. Quick question: in which 3 departments at CU do the incoming graduate students have the highest GRE verbal score? If you guesses classics, physics, and astrophysical and planetary sciences, you would be correct. A really bad general GRE score stands out. With the general GRE you don’t need a home-run, but striking out completely looks very bad. With a bit of studying you should be able to hit the 80th percentile in math and at least the 50th percentile in both verbal and writing without breaking a sweat. If you score in the 20th percentile or below I suggest studying and then retaking the test ASAP.
- Bomb the physics GRE. No single number gets looked at with more reverence by the admissions committee than the physics GRE. The overriding concern when evaluating an application in my department is can this person pass the comprehensive exam, which is essentially a test of a person’s knowledge in physics and mathematics. In our department some moderately formal statistics have shown that those coming into the department with a 60th percentile or better have close to a 100% pass rate on our comprehensive exam, while those below the 40th percentile have closer to a 50% pass rate. The overall pass rate is about 90%. Your physics GRE score really can tell us something about how you are likely to do on the comprehensive exam.
- Profess your undying commitment to a very narrow research topic. Sometimes undergraduates get a taste of research and fall head over heels in love with the narrow piece they’ve been working on. I think everyone in graduate school will agree that their research has changed in some way from undergrad to grad school. It’s okay to state some broad preferences (e.g. observational cosmology, numerical astrophysics, ultra-fast optics, etc.), but as a applicant you need to realize that there are many fish in the research sea. It can actually hurt you if you write an application completely focused on your desire to work on one narrow piece of research.
- Don’t look at the research going on in the department you are applying to. You should avoid sucking up in your personal statement, but it’s not bad to highlight potential areas of overlap your previous research experience has with the research going on in the department to which you are applying. Here at CU we have the PI for the Cosmic Origins Spectrograph (COS) on Hubble, so if you want to do anything in UV spectroscopy, from cosmology to exoplanet atmospheres, there is funding available. Furthermore if you look into the research interests of our faculty, you’ll find that 6 of the 22 faculty are involved in topics related to COS. So if you’re interested in UV spectroscopy, mention it. On the other hand, CU has no one doing exoplanet detection, so if you want to find exoplanets, maybe that should not be the focus of your application.
- Get your application materials in late. This should go without saying but if your transcripts, GRE scores, or personal statement are late that’s points against you. However unfairly the applicant also suffers a penalty in perception if their letters of recommendation are late. So talk with your recommenders early and follow through. If the letters are a couple days late you are probably okay but when days turn into weeks you may have some serious issues.
Wednesday, June 16, 2010
Previous Posts in This Series: