Assigning values based conditional on two other variables in R

Question

I have a dataset of instances of protests in the US and want to assign certain values to a district variable based on the state and the county. Basically I want to "tell" the code "If an event occured a) in X state and b) in one of Y districts it belongs in Z district. I tried this:

#Alabama
dfprotest$district[dfprotest$ADMIN1=="Alabama" & dfprotest$ADMIN2=="Washington" | "Mobile" | "Baldwin" | 
                                                          "Escambia" | "Monroe" | "Clarke"]<- "AL-1"

It didn't work though. Anyone want to help an R-noob out?

Noob (working) way is `dfprotest$ADMIN2=="Washington" | dfprotest$ADMIN2=="Mobile" | dfprotest$ADMIN2=="Baldwin"...`. Better way is `dfprotest$ADMIN2 %in% c("Washington", "Mobile", "Baldwin", ...)` — Gregor Thomas, Nov 02 '20 at 20:08
Alternately, pro way is to make a look-up table. Create a data frame that has columns `ADMIN1`, `ADMIN2` and `district`, use a merge/join to add the `district` column to `dfprotest`: `dplyr::left_join(dfprotest, district_lookup, by = c("ADMIN1", "ADMIN2"))`. — Gregor Thomas, Nov 02 '20 at 20:11
Here's a worked example of [using a lookup table](https://stackoverflow.com/q/64649153/903061). — Gregor Thomas, Nov 02 '20 at 20:12

score 0 · Answer 1 · edited Nov 02 '20 at 20:31

0

I would use dplyr library mutate + case_when combination

library(dplyr)
dfprotest <- dfprotest %>%
    mutate(district = case_when(ADMIN2 %in% c("Washington", "Mobile", "etc") ~ "AL-1"))

edited Nov 02 '20 at 20:31

markus

25,843
5
39
58

answered Nov 02 '20 at 20:11

hachiko

671
7
20

Where does the state variable come in (ADMIN1)? – Gustaf Westin Nov 02 '20 at 20:36
i hope this works :| dfprotest <- dfprotest %>% mutate(district = case_when(ADMIN1 == "Alabama" & ADMIN2 %in% c("Washington", "Mobile", "etc") ~ "AL-1")) – hachiko Nov 02 '20 at 20:43

Assigning values based conditional on two other variables in R

1 Answers1