Hey everyone!
I am working on a project and encountered an issue in my workflow for preparing data before running hot spot analysis, suitability modeling, and spatial dependence tests (Moran’s I / LISA). Before moving forward, I want to make sure my event counts are accurate.
For background:
My project involves point-based data that must be assigned to the portion of each census block group that lies within a specific neighborhood zone. Since block group polygons do not align neatly with the drainage zone boundaries, my workflow has been the following:
- Intersect block groups with neighborhood zones to create “block group × drainage zone” polygons
- Calculate area weights for each polygon fragment
- Merge all neighborhood-zone layers into a single feature class to streamline later steps
Now, I attempted a Spatial Join (Analysis tool) to count how many event points fall within each polygon fragment. However, when verifying the results, the Join_Count values are incorrect. Some polygons contain only one point when viewed on the map, yet the Spatial Join reports multiple events. The event data are simple points, and the drainage zones are boundary polygons, so I expected a straightforward point-in-polygon assignment.
Has anyone experienced similar issues where Spatial Join produces inflated point counts? Any guidance on how to correctly assign point data to subdivided polygons would be greatly appreciated before I proceed with the modeling portion.
Thanks in advance!
