Big Data Versus The SAT

Rick Bookstaber:

You would think that in the emerging world of big data, where Amazon has gone from recommending books to predicting what your next purchase will be, we should be able to find ways to predict how well a student will do in college, and more than that, predict the colleges where he will thrive and reach his potential. Colleges have a rich database at their disposal: high school transcripts, socio-economic data such as household income and family educational background, recommendations and the extra-curricular activities of every applicant, and data on performance ex post for those who have attended. For many universities, this is a database that encompasses hundreds of thousands of students.

Last year, I had to make the decision on which college to attend. Thankfully, I had managed to get fairly high scores on my SATs, but all that did was give me the possibility of having access to most colleges, I still had to pick which ones to apply for. The only easily accessible data points I could use to make these decisions were the locations of the colleges and the strengths of their computer science programs, the latter of which varied wildly depending on which study you looked at. Basically, I applied for and chose the U of A because it was close to home and the computer science program was supposed to be pretty decent. I wasn’t set on staying in Tucson, but the convenience and the lower in-state tuition was simply easier since I didn’t have any explicit reasons to choose any other universities. While I’m not unhappy with my decision, I would not be at all surprised if I’m missing out on some college out there that would have been a much better fit for me. I have no idea if that’s true though, because the way the system is right now, the job to find which college is right is left completely up to the student. This makes no sense whatsoever. We are forcing the student to pick from hundreds of colleges, the cultures and student bodies of which they know very little about, when we could instead shift the hard work onto the colleges. Colleges know very much about their own cultures and student bodies, and they also get tons of data on every student that is applying. Why not analyze all of this data and come up with educated guesses as to which schools could be right for individual students? My decision could have been completely different if I had been presented with a list of colleges telling me that I am statistically more likely to be a good match there than I am at the U of A, or any other college. I could always ignore the list and choose what I wanted anyway, but the existence of such a tool would completely change the way I, and all students, go about picking universities.

As Bookstaber discusses in his article, this isn’t an impossible goal. While setting up a new system based on big data would certainly not be easy, and would likely take years before it reached a fully optimal level, the actual tools with which to analyze the data already exist in companies like Google, Facebook and Amazon. Wouldn’t it be nice if we could use the methods these companies have already developed for analyzing big data from millions of individuals as a way to improve our education system instead of just to serve targeted ads?