Geoenrichment of census polygons by aggregating points

04-22-2021 07:26 AM
Occasional Contributor III

Is there a way to GeoEnrich the census block polygon feature layer with columns containing counts of points from a private point layer in such a way that the columns are publicly accessible?

Let's say I have points in a private hosted featurelayer representing locations of crime incidents.  The precise locations are sensitive, considered Personal Identifiable Information (PII) by the agency that owns the data.  

I would like to aggregate the point information to the census block level and/or tract level, and publish that for public consumption by analysts.

For example, say I have 3 types of incidents: vandalism, assault, and burglary.  How can I present a census block polygon featurelayer with these columns: Vandalisms, Assaults, Burglaries, such that each feature contains a count of the respective number of points that fall within the census polygon?

I would then like to publish it so that an analyst could then add arcade field based on expressions like : $feature.vandalisms / $feature.FAMHH00, where FAMHH00 is field from the census bureau containing number of households.

Again, I do not want the analyst to have access to the actual point locations, which are PII. I do want the analyst to be able to use the aggregated crime fields.

Tags (1)
0 Kudos
1 Reply
Esri Contributor

Hi Kirk,
I think what you might be looking for is the Aggregate Points tool, available in the Analysis tab under Summarize Data.
You can set your Crime Incidents layer as the input point layer, and then select a census block level or tracts layer (either from your Contents or added from the Living Atlas) as the polygon layer into which to aggregate the points.
Assuming you have a field called Incident_type, you'll then want to use Choose a field to group by and set Incident_type as the groupby field. Then you can provide a unique output name, uncheck Use current map extent and hit Run Analysis.
Your results will be two layers: one polygon layer that will contain the count of all points per boundary, and a groupBy summary table, which will break down the count of points in each boundary by Incident_type. This will allow you to see how many Vandalisms occurred within each polygon.

Let me know if you need any further help or clarification

- Amanda

Product Engineer - Web Analysis
0 Kudos