Hello Roger.
First, you should only do this for you, not for me. I think you know that I am not likely to do anything with this. But, I think this work is of interest to this page, and I am interested, and think that you are. Who knows, you may find something unexpected that will get John's attention. Also, we can run this against your global file of >M5 quakes, the one with near neighbors (aftershocks, or smaller quakes of swarms) removed. So, you can do the Kp for the years you have complete data on these quakes. But, I think you should also do this for smaller quakes in certain regions: maybe M5+ for Mid Atlantic Ridge, M3+ for California, M3+ for high latitude North America intraplate, M3+ for intraplate period (e.g., Oklahoma-Kansas, New Madrid, Tennessee, northeast, Quebec, and points between, M3+ in Coso-Long Valley. Maybe Don in Hollister has some suitable catalogues if you do not. I'd do the Kp for as many years as these catalogues are relatively complete for M3+ (maybe 1980 or so?).
Chris