America's best-bang-for-the-buck colleges: in this year's rankings, we show which schools get their students over the finish line at a reasonable price.

• Author: Fishman, Rachel

The main flaw in most college rankings is that they tend to measure how prestigious institutions are rather than how effectively they serve their students. Indeed, many schools have moved up the U.S. News & World Report rankings by abandoning the students they traditionally serve in favor of recruiting "a better sort" by raising their admissions standards.

The Washington Monthly has long believed that such behavior by colleges doesn't serve the broader interests of the country, and that rewarding such behavior is wrong. And so the magazine designed its own ranking system to do the opposite: to rate colleges based on how well they perform with the students they have, regardless of the students' backgrounds or SAT scores, on metrics that measure the widely shared national goals of increasing social mobility, producing research, and inspiring public service.

One goal that has long been missing in the magazine's rankings, however, is cost-effectiveness. After all, college may be a good investment, but not if you pay too much for it. Pursuing a college education still makes economic sense for most students, but that won't be true for much longer if tuitions continue to rise, as they have for years, at rates faster even than health care costs.

So this year, the Washington Monthly rankings incorporate a new measure we call the "cost-adjusted graduation rate." This involves tweaking the calculations the magazine has long used to derive a school's social mobility score. In the past, we predicted a college's graduation rate using the median SAT/ACT score of each school and the percentage of its students receiving Pell Grants and then compared it to the actual graduation rate. This year, we made two changes. First, to increase our ability to predict graduation rates, we used additional student and institutional characteristics, such as the percentage of students attending full time and the admit rate. Second, to get at cost-effectiveness, we took the gap between the predicted and actual graduation rate of a school and divided it by the net price of attending that institution. (Net price represents the average price that first-time, full-time students pay after subtracting the need-based financial aid they receive.) The aim of our new cost-adjusted graduation rate is to highlight those colleges that use their resources to effectively educate students at a relatively low cost--and to call out those that burn though tuition dollars without much to show for it.

What did we find? First, that colleges and universities that do well by this measure tend to be public institutions. That's not a surprise, given that tuition at these schools is kept relatively low by state subsidies (though per-student subsidies have been declining in many states). It also turns out that quite a few minority-serving institutions, such as the University of...