DATA: I have two polygon datasets that have no attributes in common other than location, so must be joined using a spatial join. One dataset is polygons of cadastral parcels (i.e. properties) and the other of hunting districts. Every part of the country is officially assigned to a hunting district, so both datasets completely cover the same area.
DESIRED OUTPUT: Each cadastral parcel should have all of the hunting districts responsible for it listed. This relationship is clearly one-to-many (i.e. one cadastral parcel to many hunting districts) -- most parcels belong to only one hunting district, but larger parcels often belong to 2+.
PROBLEM: The problem is that the boundaries between the datasets don't match completely (i.e. the hunting dataset boundaries were slightly simplified at some point) and often overlap most neighboring cadastral parcels ever so slightly, rendering a spatial join based on "INTERSECT" useless. The "HAVE CENTER IN" works perfectly for all parcels that belong to only one hunting area, which is the majority, but produces false results for parcels belonging to more than one hunting area.
This whole analysis would be easier to do in any number of other spatial databases with SQL, but I am unfortunately limited to ArcGIS for this task.
QUESTIONS: The only thing I can think of is to try a negative search radius. One can set a negative search radius for "INTERSECT" and "WITHIN A DISTANCE", but I am uncertain as to how these two queries differ in ArcGIS when giving a negative search distance. Do they not then become synonymous?