IIRC, the methodology is based on facebook likes. This skews female and young, because if you are not a facebook user, you don't show up and if your particular school is not a heavy facebook user your school will fall behind schools that use facebook a great deal - all things being equal.
This is not as bad as some scurrilous work the Times posted several years ago that used voluntary responses from campuses to create a map by some piss poor Geography grad students. Their committee chair should have been called on the carpet for letting that piece of crap become public.
As with any social science problem - the first issue is definition. What is a fan? I would suggest that a facebook like is not necessarily a sports fan and I have firsthand proof as the wife "likes" UNC-Ch as a she has a degree from UNC-Ch, but she is most definitely an NC State's sports fan (poor girl). I don't use facebook.
Moreover, if you have had any recent procedure at UNC Hospital - you are going to be contact by email and prompted at some point to interact with them via Facebook or another platform. WF Hospital does the same, and Duke Hospital offers the same.
For me, I would attempt to define a fan as someone who would buy a ticket or actively attempt to watch a game rather than do something else with that 2-4 hours.
In NC we call the casual UNC fan a Wal-Mart fan - this is typically a person who could gain admission to any UNC institutions, let alone UNC-Ch, but likes to associate with the basketball team. (To be an NC State after 25 years in the wilderness requires you to be a real NC State with a solid connection to NC State
)
Moreover, the use of "likes" doesn't even allow for a differentiation between football and basketball. UNC-Ch and Duke have lots of basketball fans that don't begin to follow their football teams.
When I am in SC, I don't get the impression that the Gamecocks have that much more support than Clemson. I don't see that on display in the Myrtle Beach area, or in Charlotte.
"Real" fans tend to make a visual display of their allegiance with a banner, flag, sticker, etc. Accurately measuring that would be an absolute *****. That means that substitute metrics like merchandise sold or facebook likes are used.