While these sorts of ROI-based studies should be based on hard, fast numbers, they contain a lot of subjectivity in them.
For example, if you look closely, you will notice the ratio of early career earnings to one year's cost is the highest at Georgia Tech (2.30) than any other school above it. There are a few (I didn't browse far) below us that are a tick higher of a ratio, but they have a lower cost and lower salary. There is a surprising number of colleges listed as 'the best for your money' up at the top where your early career earnings are less than a year's cost (ie < 1.00). Furthermore, our REAL ratio (average cost after financial aid) is 4.76 - ie one single year of early career earnings equals 4.76 years of the average cost to attend Georgia Tech. I didn't look far and wide, but on first glance I didn't see anybody with that high of a ratio.
It looks to me that we get dinged somehow because our average student debt at the end is higher than others. But to me that's an irrelevant number - the cost of what you pay is the cost of what you pay. What you earn is what you earn. If more folks from Princeton graduate with less debt, that just means they have more rich people attend. If you ignore people's ability to pay and just look at the straight numbers of earnings versus cost to attend, we are at the top. Princeton has more students who can pay up front, we have more that pay later, and lucky for those students with rich parents - if they had to pay their own way like more Georgia Tech students do, they'd have a much harder time paying for it, based on how expensive Princeton is (they make about the same earnings, but the school costs 30% more per year).
That's my head scratcher - Princeton is ranked #1 and we are ranked #19 in colleges that are 'best for your money'. Both places send students into the workforce to make approximately $69,000/year, which is just about the top. A Princeton degree costs 30% more. But yet look at the rankings, LOL.