House parties have nothing to do with the Princeton Review ranking.
You're half right on the second quoted sentence; these lists don't indicate how much partying is going on. That's because the methodology for Princeton Review is very subject to response bias, as it is based on student surveys.
For example, at Ohio University, students have actively been trying to top the rankings for over a decade now (in my opinion, because there isn't much else that school does well). As a result, their Princeton Review statistics for binge drinking, Greek popularity, and study hours are greatly skewed compared to data collected by Institutional Research. OU tops the list because students believe they are the top party school and have a twisted sense of pride in that, not because they actually are (which, yes, would be difficult to quantify, but could be more accurately divined through IR data).