I'm ranking universities by their R&D expenditures across a number of years. I'm interested in comparing the average performance of the top X universities (defined as most R&D expenditures) each year to different comparitors. Of course, each year, the top X changes each year. My project accounts for this through table calcs....and there is a nice partition between those schools in the Top X and those not. If a school is only sometimes in the top X, they will appear in both partitions (with nulls filling in their duplicate entry).
Is there any way to "collapse" the top X so I can see the average of this group/partition?