| Complete Navigation
Map (revised 4/1/05)
FAQ:
Other Departments:
With help from
|
Understanding the US News Rankings
How I Interpret USNews DataDespite all its failings, I still subscribe to and read the USNews rankings religiously. But I don't just accept the data as given; I often do my own calculations, changing the relative weights assigned to each variable. For instance, in calculating placement rank, USNews assigns a weight of 30% to employment at graduation and 60% to employment 9 months later. I think that more weight should be placed on employment at graduation, since those are the folks who have jobs they want; nine months later, you'll settle for anything just to pay off the student loans. So I did my own calculation of placement rank, and shared it with you here. I also recalculated the composite rank, using the data that I could verify and changing the relative weights as I thought appropriate. There are some surprising changes -- look. Finally, I knew there was no way to cure the one fatal flaw in the USNews reputation ranking, their asking lawyers outside of a region to rank regional and local schools. What can a lawyer in Philly really know about Campbell grads, or one in Houston know about Pace? So I undertook my own survey at the Law Forums, asking knowledgeable and honest admissions officers to give me the scoop on how various law schools fare in the pertinent job markets. Here's what I learned. Beyond any statistical calculation, however, my most important advice in interpreting USNews rankings is, THINK! Berkeley and Stanford dropped in the rankings after the big quake that destroyed the Bay Bridge. Fordham's yield (a selectivity factor) skyrocketed the year they built a dorm in midtown Manhattan. And Maryland had an 80% increase in applications the year they won an NCAA championship. None of these events has anything to do with a school's quality; the teachers are the same, the library is still full of books, and the students are still arguing about trivia over lunch. But the USNews ranking will change. As I am very fond of saying, you don't know what data means unless you know what it means. So when you see anomalous data, don't just stare at it -- or worse, rely on it -- ask yourself what happened. If you learn to interpret data, it can be a valuable tool. |