Tags: admissions, applications, Architecture, Berkeley, California, California Institute of Technology, CalTech, Canon, Chicago, college, Color, Getty Villa, Graduate School, Greek, Harvard, HDR, High Dynamic Range, Los Angeles, Malibu, Massachusetts Institute of Technology, MIT, Photgraphy, Photo, Photomatix, physics, Rebel XSi, Roman, Southern California, Stanford, Tamron AF 28-300mm f/3.5, UC Berkeley, UC San Diego, UC Santa Barbara, UCB, UCSB, UCSD, UIUC, University of Chicago, University of Illinois Urbana Champaign, University of Pennsylvania, UPenn
Okay, I admit it. I am obsessed. I’m waiting on decisions from 11 of the 12 graduate schools I applied to (for my physics Ph.D.) and checking constantly for any news. Not just my e-mail, either. I try to limit myself to only checking the mail once a day, but I usually end up checking twice. I would like to think that the internet helps me with graduate school because of all the information I have access to, but I know it feeds my obsession. I check GradCafe many times a day to see reports from other people around the country so I know exactly when all the schools I’m applying to start to contact people. Unfortunately, schools this year are responding later that people reported the last two years, which sets me further on edge and makes me check even more.
Don’t get me started on Excel. Thanks to detailed applicant profiles PhysicsGRE.com for the last three years I have been able to assemble an immense amount of data on who gets in where. I only used data from American males who are white or Asian to control for admission biases about gender, citizenship, and race. Even with those constraints I was able to get over a hundred applicant profiles. I used their cumulative GPA’s, major GPA’s, three general GRE sub-scores and Physics GRE scores to get a multivariate linear regression correlating these variables to the success these students had in applying to grad schools. For the most part I only learned what I had already been told: that the PGRE is by far the biggest factor in admissions, but it was good to get a quantitative confirmation of this frequently repeated belief. Also, I fed my own stats into the regression to see what would come out. 😛 I’m sure this has little to no predictive power, but what can I say. I’m obsessed. I’m beyond obsessed. College and graduate school admissions have become no less than a fetish to me and millions of other students. We wait on pins and needles for admissions committees to determine our fate. So far I am proud to say that I shower every day, exercise every other day (ish, but who really sticks to their exercise schedule), and do my homework. When my obsession/fetish rises to the level of a life inhibiting pathology, I’ll blog it.
P.S. If you stumble upon this post via Google searches for physics admission things, I’ll be happy to share my data with you! I even did single variable regressions (with PGRE scores) for the ten American schools I applied to (here goes: UC San Diego, CalTech, UC Santa Barbara, Stanford, UC Berkeley, UChicago, University of Illinois – Urbana-Champaign, MIT, Harvard, and the University of Pennsylvania). That way I’ll feel all my work has some value. In the spirit of science, I will tell you that my model predicts I will be admitted to UCSD, UCSB, UIUC, and UPenn. It also predicts I have a 52% chance of admission to the physics Ph.D. program at UC Berkeley, but for various reasons I think it’s an over-estimate. Now I have made a prediction. Good thing I paid $1,000 in application fees to falsify it.
P.P.S. All the schools have a pretty good r^2 value for the linear regression admission v. PGRE, except for Harvard, where the r^2 is a measly 0.08 from the data I have access to. Maybe they look beyond the PGRE? Or maybe Harvard is so competitive nothing matters but pure chance.
Oh, and I took this photo at the Getty Villa. Tonemapped in Photomatix 4 from a single RAW. I’m posting all these pictures from that beautifully relaxing location to try to calm myself down. Think it’s working?