Using ggplot2 to Fill in Counties Based on FIPS Code

Question

I am looking to fill in counties on a ggplot2 US map from a vector of FIPS codes that I am interested in. I would like to do this using ggplot2 and its various functions such as geom_map and geom_polygon as they seem to offer quite a lot of flexibility in customisation, however I have only managed to accomplish a rudimentary version using choropleth. The rudimentary version is:

library(choroplethr)
library(ggplot2)
library(dplyr)
library(tidyverse)
library(choroplethrMaps)

onesvec <- rep(100000, 210)
county <- data.frame(region = OOTFIPS, value = onesvec)

data(county)

choro              = CountyChoropleth$new(county)
choro$ggplot_scale = scale_fill_brewer(name="Population", palette=2, drop=FALSE)
choro$render() 

which gives this. Here OOTFIPS is a vector of the FIPS codes. To make this work, I have had to create a fake "value" vector so that it will highlight the counties desired from the OOTFIPS vector.

Starting with ggplot2, when I make a first pass I cant even get an accurate map. I have started with this code

counties <- map_data("county")
ggplot() + geom_polygon(data = counties, aes(x = long, y = lat))

but it returns a map that isn't even geographically accurate. Any advice on how to make a map similar to the first image but with the techniques from the second?

Thanks!

ExampleFIPS <- data.frame("FIPS" = c(19097, 17155, 50009, 27055, 39143, 55113, 44003, 55011, 19105, 46109, 19179, 55099, 51057))

Edit: Here is as far as I have been able to manage

states <- map_data("state")
counties <- map_data("county")

us_base <- ggplot(data = states, mapping = aes(x = long, y = lat, group = group)) + coord_fixed(1.3) + 
  geom_polygon(color = "black", fill = "white")


us_base + theme_void() + geom_polygon(data = counties, fill = NA, color = "gray") +
  geom_polygon(color = "black", fill = NA)

But now the question is how to specify and highlight counties with FIPS codes in the vector?

Answer 1

I'd call this a straight forward example of data wrangling.. you need to find where all your needed data is. That is, map_data("county") is missing the fips, then you need to google where to find it - maps::county.fips, check the format, create a data frame from it and then join it in and use it.

library(tidyverse)

ExampleFIPS <- c(19097, 17155, 50009, 27055, 39143, 55113, 44003, 55011, 19105, 46109, 19179, 55099, 51057)

maps::county.fips %>%
  as.tibble %>% 
  extract(polyname, c("region", "subregion"), "^([^,]+),([^,]+)$") ->
  dfips

map_data("county") %>% 
  left_join(dfips) ->
  dall

dall %>% 
  mutate(is_example = fips %in% ExampleFIPS) %>% 
  ggplot(aes(long, lat, group = group)) +
  geom_polygon(aes(fill=is_example), color="gray70") +
  coord_map() +
  scale_fill_manual(values=c("TRUE"="red", "FALSE"="gray90"))

This produces