US County Chropleths

You can create a choropleth of US Counties with the function county_choropleth:

library(choroplethr)

?df_pop_county
data(df_pop_county)

?county_choropleth
county_choropleth(df_pop_county)

plot of chunk unnamed-chunk-1

As demonstrated above, the only required parameter to county_choropleth is a data.frame. You can see the optional parameters by typing ?county_choropleth.

Data Requirements

The data.frame that you provide to county_choropleth must have one column named “region” and one column named “value”. Your entries for “region” must exactly match how regions are named in the map which choroplethr uses. These names are defined in the object county.regions:

library(choroplethrMaps)

?county.regions
data(county.regions)
head(county.regions)
##    region county.fips.character county.name state.name
## 1    1001                 01001     autauga    alabama
## 36   1003                 01003     baldwin    alabama
## 55   1005                 01005     barbour    alabama
## 15   1007                 01007        bibb    alabama
## 2    1009                 01009      blount    alabama
## 16   1011                 01011     bullock    alabama
##    state.fips.character state.abb
## 1                    01        AL
## 36                   01        AL
## 55                   01        AL
## 15                   01        AL
## 2                    01        AL
## 16                   01        AL

In order to use choroplethr, you must use the naming convention in the “region” column of county.regions. That is, you must use the numeric version of the county FIPS code - i.e. you must drop any leading zeroes.

Exploring Data

The county_choropleth function provides three parameters to facilitate exploring data: num_colors, state_zoom and county_zoom. num_colors defaults to 7, which means that there are 7 colors on the map. An equal number of regions is assigned to each color; a value of 1 uses a continuous scale. Both state_zoom and county_zoom default to NULL, which means that all counties are shown.

As an example of zooming by state with a continuous scale, here is code to create a map of the population of all US Counties on the West Coast. The outlier is Los Angeles County.

county_choropleth(df_pop_county,
                 title      = "2012 Population Estimates",
                 legend     = "Population",
                 num_colors = 1,
                 state_zoom = c("california", "washington", "oregon"))

plot of chunk unnamed-chunk-3

As an example of zooming by county, this code maps the population of the 9 Counties in the San Francisco Bay Area:

# FIPS codes for Alameda, Contra Costa, Marin, Napa, San Francisco, San Mateo, Santa Clara, 
# Solano, and Sonoma counties
bay_area_counties = c(6001, 6013, 6041, 6055, 6075, 6081, 6085, 6095, 6097)
county_choropleth(df_pop_county,
                 title       = "2012 Population Estimates",
                 legend      = "Population",
                 num_colors  = 1,
                 county_zoom = bay_area_counties)

plot of chunk unnamed-chunk-4

Advanced Options

Any customization outside the optional parameters presented above will require you to create a CountyChoropleth object. choroplethr uses R6 to take advantage of object-oriented programming. Here is an example of using the ggplot2_scale variable on the base Choropleth object to customize the palette used.

library(ggplot2)

choro = CountyChoropleth$new(df_pop_county)
choro$title = "2012 Population Estimates"
choro$ggplot_scale = scale_fill_brewer(name="Population", palette=2, drop=FALSE)
choro$render()

plot of chunk unnamed-chunk-5

Note: Care must be taken when manually setting the scale on CountyChoropleth objects. In particular, choroplethr uses ggplot2 custom annotations to render Alaska and Hawaii as insets. This means that the scales of the insets and the main map will only be the same if you do the following

  1. for discrete scales, pass drop=FALSE to the scale (as above).
  2. for continuous scales, pass limits which encompass the minimum and maximum values for the entire dataset.