Ariana @mendible requests (she can comment if I've missed anything):
Please use the Williamstown government's own report on police activity disaggregated by call type, to cross-validate against what we do with our analysis of the dataset.
Hotlink: https://williamstownma.gov/wp-content/uploads/2022/05/ATR21.pdf
Essentially: do our numbers match theirs? If not, why not? Which aspects of the data, as we have it, have biases/patterns in missing data?
I may upload a copy of this pdf to our repository if/when that link goes defunct.