So the ranking you are looking at combines 2/3rds total hires with 1/3rd undergrad %. I am looking just at undergrad %, what I called “per capita”.
The argument at that link for that measure is that author felt like per capita measures were unfairly penalizing larger public schools, and the author believes it is generally better to go to larger schools, so wanted to come up with a measure that would help larger schools do better against smaller schools. But I think the actual evidence is very much against that author’s view on that subject. Indeed, the author has internally contradictory logic. Consider this:
From an expected value perspective, a firm would rather visit Columbia than Claremont McKenna even if their Undergrad Placement % is similar. Why fight over the two kids that want to do banking at Claremont McKenna when you can get swarmed by an army of Columbia kids foaming at the mouth?
And then contrast it with this:
Even if you ask the proudest NYU graduate (which I did), they’d probably be reluctant to say that their school is the 2nd best place to get into investment banking from. NYU produces a ton of high-quality candidates, but you’d still rather be at Harvard where it’s significantly easier to impress both headhunters and your parents.
That’s essentially the same comparison, and all he really did was take the Ivy League side of it each time, using opposite logic to justify his two opposite positions. And I note he actually chose a higher per capita college (Columbia) for the CMC comparison, which loaded the question.
And it is easy to play that game a different way. For example, I think it would clearly be a real mistake to think that, say, Penn State is a better choice than Williams if you are interested in IB (Penn State being at 33). There is no way the far higher volume at Penn State somehow outweighs Williams being a highly selective college with an extremely robust alum network in finance.
So I think that author had it right when he said:
It’s not really possible to determine a denominator that is fair to all schools
I think he should have just stopped there, rather than trying to come up with a formula that got Harvard and Columbia where he thought they should be, ignoring all the other clear misguidance that produced.
In any event, since we were discussing the percentages of graduating students with different majors, the relevant IB measure to inform that discussion would be the per capita measure.