NA about openpolicing

[janovergoor]

what script are you referring to?

[xnoorrr]

i'm sorry. I was trying to work out this part:

summary_stats <- function(search_conducted, contraband_found) {
n_stops = length(search_conducted)
n_searches = sum(search_conducted)
n_hits = sum(contraband_found)
search_rate = n_searches / n_stops
hit_rate = n_hits / n_searches
return(data.frame(n_stops, n_searches, n_hits, search_rate, hit_rate))
}

basic_summary_statistics_by_race = florida %>%
group_by(driver_race) %>%
do(summary_stats(.$search_conducted, .$contraband_found))
basic_summary_statistics_by_race

[xnoorrr]

A tibble: 6 x 6

Groups: driver_race [6]

driver_race n_stops n_searches n_hits search_rate hit_rate

1 "" 1345 NA NA NA NA
2 Asian 69112 NA NA NA NA
3 Black 1021743 NA NA NA NA
4 Hispanic 1173182 NA NA NA NA
5 Other 138772 NA NA NA NA
6 White 3017292 NA NA NA NA

[epierson9]

I assume this is the result of running on FL data? If so, this is happening because the FL data has NAs in the search_conducted/contraband_found columns (because some stops lack that data), so the search and hit counts/rates become NA as well. You probably want to filter those out prior to computing search and hit statistics; this is not necessary for CT, the state used in the tutorial. (Actually, FL has no contraband found data at all, so hit rates will not be computable).

[xnoorrr]

Ah I was expecting that but assumed I made a mistake with the data because i couldn't imagine there wasn't found any contraband at all. Too bad, i'm going to try some thing else. thanks for your help!

[epierson9]

Good luck! In case you haven't yet seen it, the data readme will often provide useful tips on specific states (eg, missing data). https://github.com/5harad/openpolicing/blob/master/DATA-README.md