Search code examples
rdictionarydplyrusmap

How do I map us-states just by names? In R

In R I have a dataframe that looks like this:

head(map)

mapstates mapdelay 
1    Alabama      457        
2     Alaska      374       
3    Arizona     2852       
4   Arkansas      339        
5 California    11609       
6   Colorado     3298

I would like to plot the frequencies on a map.

Searching the www i have found the usmap package f.e. . But all the solutions that I have found require lattitude and longitude data which I do not have.

Using the usmap package and dplyr I tried connecting lattitude and longitude data from usmap to my dataframe:

library(dplyr)

newdata<- left_join(mapstates, delaymap, by= "state")

and this error message popped up:

"Fehler: by can't contain join column state which is missing from LHS Run rlang::last_error() to see where the error occurred."

Is there any way to plot my data just with the name of the states?

Solution

  • It looks like your data.frame “map” doesn’t have a column named “states” and therefore the join command is failing. I can’t test this, since I don’t have the exact datasets you are using, but it might just work, to change your command as follows:

    newdata <- left_join(mapstates, delaymap, by= c("mapstates" = "state"))

    I executed the following code, to reproduce your idea in my environment and it does work. I am using the “statepop” data.frame from the usmap package to join your data to, as it has the fips code needed by the package, to identify states (or counties). Also see the documentation on the data argument of ?plot_usmap().

    library(dplyr)
library(usmap)
library(ggplot2)

map_df <- tibble::tribble(
      ~"mapstates",~"mapdelay",
  "Alabama"  ,   457,
  "Alaska"  ,   374,
  "Arizona"  ,  2852,
  "Arkansas"  ,   339,
  "California"  , 11609,
  "Colorado"  ,  3298
  )

new_df <- left_join(statepop, map_df, by = c("full" = "mapstates"))

plot_usmap(data = new_df, values = "mapdelay", color = "white") + 
  scale_fill_continuous(
    low = "white", high = "red", name = "Delay", label = scales::comma
  ) + theme(legend.position = "right")

    Note: Some of the states are not colored, because I do not have your full dataset.