February 14, 2017

· R · dataviz

I’ve previously written about how to make simple, print ready maps in R using library(maps). Unfortunately, the more recent versions the county database has some issues with relating FIPS codes to the correct county on the map. I’ve found that I had to roll back library(maps) to version 2.3-0 in order to get the county database to work as advertised. (FWIW, I haven’t had any issues with any of the other databases in the more recent versions.)

Since the original post is ancient, I’ve decided to update the code somewhat to make a better map. The code below creates a choropleth map of the Metropolitan Power Diffusion Index (MPDI) for special districts from year 2012. The base data comes from the 2012 Census of Governments. I’ll have a post on the MPDI and other measures of special district usage in a later post.

library(maps)
library(mapproj)
library(RColorBrewer)

nclr <- 5
plotclr.org <- brewer.pal(nclr, "OrRd")

mpdi <- read.table("mpdi.csv", header=TRUE, sep=",", stringsAsFactors=FALSE)

summary(mpdi)
mpdi.df <- data.frame(mpdi)
data(county.fips)
attach(mpdi.df)

mpdi$colorBuckets.mpdi <- as.numeric(cut(mpdi$mpdi, c(quantile(mpdi$mpdi, c(0.2,0.4,0.6,0.8,1)))))
mpdi.legend <- c("Q1", "Q2", "Q3", "Q4", "Q5")
mpdi.colormatched <- mpdi$colorBuckets.mpdi[match(county.fips$fips,mpdi$co_fips)]

par(mar=c(1, 1, 1, 1) + 0.1)
map(database="county", col=plotclr.org[mpdi.colormatched], fill=TRUE, lwd=0.1, resolution=0, projection="polyconic")
map("state", col = "black", fill=FALSE, add=TRUE, lwd=0.75, resolution=0, projection="polyconic")
title(main="Metropolitan Power Diffusion Index, Special Districts")
legend("bottomleft", legend=mpdi.legend, fill=plotclr.org, bty="n", cex=0.65)

Just a few lines of code and you get this.

Metropolitan Power Diffusion Index.

Metropolitan Power Diffusion Index.

All Categories


Where

subscribe

To receive updates from this site, you can subscribe to the  RSS feed of all updates to the site in an RSS feed reader

search