STEM Expenditures: min 6 million, max 2.2 billion. Mean 412 million, Median 320 million.
Non-STEM Expenditures: min 725k, max 124 million. Mean 22 million, Median 15 million.
Research Staff (including post-docs): min 32, max 7,300. Mean 600, Median 390.
Doctorates (Humanities/Social Sciences/STEM/Other): Min 0/1/27/0, Max 179/129/589/229. Mean 51/44/202/87, Median 45/37/152/76.
Then they use PCA to determine best fits for schools among the data. Being really high in any one area gets you R1 status, from what I can tell. Some interesting plots in the linked data.
I'm having a hard time seeing how a place with only $6 million STEM expenditures could be an R1, even if they were the place with the maximun non-STEM expenditures of $124 million.
A lot of it is dependent on size, and both raw expenditures and per-capita expenditures are reflected in the final report. But yeah, 6 million seems exceptionally low.
That said, there are 105 schools that are R1s, and some of the smallest R1s are quite small. If you're that school only giving out ~30 doctorates a year, my guess is your overall funding is pretty low.
There's always going to be a line between the largest R2 and the smallest R1 though, and the difference is always likely to be pretty small. If you look at the actual data, you can see the radius of distribution of each of the rankings has some pretty distinct groupings.