What is School Desirability?

What does the “desirability score” mean?

The desirability score that you see next to each university’s name is calculated on the basis of actual student decisions. For example, when presented with a head-to-head competition, 67% of students pick Harvard over MIT, with the result that Harvard ranks #2 in desirability score while MIT ranks #3.

The graphic shows the head-to-head competitions between top schools that differ by one spot in their desirability score rankings (and two that differ by more than one spot to show some interesting exceptions). Not all the rankings based on the desirability score are statistically significant. For example, Stanford is listed at #1 and Harvard at #2, but the percent of students who choose Stanford over Harvard (53%) is very close to the percent of students who choose Harvard over Stanford (47%) and thus the null hypothesis that the schools have identical desirability cannot be rejected. Although this kind of ambiguity only happens once in the top 15 schools, it happens quite often for schools that aren’t in the top 50. Even when there is ambiguity between two schools, one of them always has a slightly higher desirability score than the other. We give the school with the higher desirability score the higher rank, which is what we did with Stanford and Harvard.

Only the top 70 or so schools have enough cross-admits (students who are admitted to both universities) to create a desirability score that can be based on head-to-head competitions. For example, the US Coast Guard Academy at #68 and the US Merchant Marine Academy at #65 will have many head-to-head competitions due to students who apply and are admitted to both schools, but neither school will have much competition with the California Institue of the Arts (#84), the Rhode Island School of Design (#85), or the Fashion Institute of Technology (#106). Thus, we use a Bayesian approach to extrapolate desirability scores for schools that are not in the top tier.

Since different types of schools are hard to compare directly–for example, schools that train seafarers versus schools that train artists–the best way to use these desirability scores is to look at them when you are comparing two fairly similar schools. It’s also important to remember that just because more people choose a school doesn’t mean it is better for you. Desirability describes how students “vote with their feet” in deciding to attend one school in preference to another, not whether they made the right choice.

Finally, even though overall influence–our primary ranking metric at AcademicInfluence.com–correlates with desirability, the correlation is imperfect. In fact, that’s exactly why we decided to create the desirability score. Sometimes it’s more interesting to know where people want to go to college (desirability) than it is to know what college is home to leading faculty and alums (influence). And sometimes it’s more interesting the other way around!

We hope you enjoy exploring the desirability scores for different colleges. We spent tens of thousands of dollars acquiring and organizing the data (spanning 2012-19) needed to calculate AcademicInfluence.com’s desirability scores. As always, don’t hesitate to give us feedback, especially about things you think will help us improve our methods.