Plotting Geographic Data with Leaflet (Color by Categorical Predictor)

library(leaflet) 
library(jsonlite) 
library(tibble) 
library(plyr)
library(dplyr)
library(data.table)
library(datasets) # loading datasets package for mtcars and Iris data
library(webshot)

url <- "http://gist.githubusercontent.com/ajav17/dee0dd44357862c75ee2872038119f17/raw/0109432d22f28fd1a669a3fd113e41c4193dbb5d/USstates_avg_latLong"
statesLocation <- fromJSON(txt = url)
statesLocation <- plyr::rename(statesLocation, replace = c("state" = "State"))

USArrests2 <- USArrests
USArrests_reshape <- setDT(USArrests2, keep.rownames = TRUE)[]
#USArrests_reshape <- tibble::rownames_to_column(USArrests2) #convert states from row to first column
#USArrests_reshape <- USArrests_reshape %>% dplyr::select(rn, Murder, Assault, UrbanPop, Rape)
colnames(USArrests_reshape) <- c("State", "Murder", "Assault", "UrbanPop", "Rape")

statesFinal <- statesLocation %>% dplyr::inner_join(USArrests_reshape) #join the USArrests dataframe to the states dataset with location

## Joining, by = "State"

meanPop <- mean(statesFinal$UrbanPop)

#Creating factor labeling all states less or greater than the average urban population
statesFinal <- mutate(statesFinal, UrbanPopCutoff = ifelse(UrbanPop < round(meanPop), "<66", ">66")) 

pal <- colorFactor(topo.colors(2), statesFinal$UrbanPopCutoff) #Creating color palette of 2 colors 

#Mapping US states where states are color coded as either having urban population less than or greater than the average urban population and the radius of the circle markers represents scaled assault numbers
leaflet(statesFinal) %>% addTiles() %>% addCircleMarkers(lng = ~longitude, lat = ~latitude, radius = ~Assault/50, opacity = 0.2, color = ~pal(UrbanPopCutoff), popup = paste(statesFinal$State, "<br>", "Assaults: ", statesFinal$Assault, sep = "")) %>% addLegend(pal = pal, values = ~UrbanPopCutoff)