04.2 Police shootings – extra material

A slight problem is that the two data sets from the Washington Post – one from their website, the other from GitHub – are similar but not identical. In particular, that from the website contains names of police stations but has no location data (latitude/longitude) while for the GitHub data it is the other way around. Additionally the names in both files are not identical and both files are missing a few hundred names. Those missing names in general are missing a lot of other variable data too. So I have created a new file that eliminates all data rows with a missing name, and where both the police station and latitude/longitude is available for as many entries as possible. It is OK to work from such a reduced data set so long as you acknowledge how it was obtained. Data is messy, and for analytical purposes we sometimes need to reduce our data sets.