To download the existing datasets, reports, and maps, you must install DVC:

pip install dvc[gs]

Followed by:

dvc pull

This should automatically download all the required datasets.

To install the required python libraries, please install a fresh python environment (using anaconda or such like), then run the command:

pip install -r requirements.txt

pointing to the requirements.txt file in the NIBRS directory.

In order to produce the enforcement ratios you must either download the pre-requisite datasets from DVC, using dvc pull, or you may produce all results yourself using the following steps:

1. Download the appropriate raw NIBRS files for each state. This can be achieved by running the python script: python data_downloading/download_and_extract.py. This currently defaults to downloading years 2010-2019 and over all states. This can be changed from within the script.

2. Produce the appropriate NIBRS dataset. Using the query_nibrs.py python script python scripts/python/data_processing/query_nibrs.py. The script has a number of arguments that can be explored with: python query_nibrs.py -h.

3. Produce the enforcement ratios. Run the selection_ratio.py script scripts/python/data_processing/selection_bias.py with appropriate arguments. Run scripts/python/data_processing/selection_bias.py -h for help, or consult the image below:

There are many scripts used to create the figures and tables in the paper, please consult the table below to find which script corresponds to which figure: