It looks like the underlying join that you want to work with is between the shapefile for your polygon data and the csv for your violations vs. per capital income data. That should be do-able with just a full outer join using 0 = 1 as the calculated values on the join. See Spatial file + csv on this blog post: Dual Axis Mapping - Many Ways
Since you also have several other data sources joined in or blended, you'll have to adjust to add those in as well. One method you might consider is to create two Hyper extracts, one for each set of joined data (e.g., the polygons + all additional attribute tables, and a second file with the CSV + all additional attribute tables) and then just do the join between those two (it should be the same as the spatial file + csv, since the Hyper extract based on the shapefile will have a geometry column).
It would probably be easier if you converted your CSV with point locations into a shapefile and did the join that way, however, because in 2018.3 you can use the new spatial intersection functionality.
Since I only have access to the TWBX files that you attached and not the original shapefile and csv data it's difficult for me to make an example twbx to share, but the ideas above should get you started with combining the files in Tableau.