Choropleth in R: custom breaks and plotting

Question

I have a question with regard to plotting a choropleth map. What is recommended with regard to plotting a map with custom breaks? And make it look a bit attractive.

In my script I have used the plot and spplot command to visualize some data but I am not entirely happy with the result. It is easy to see why. Using spplot I get a colour ramp that repeats colours for different values which is a bit weird. Using plot with the custom breaks I have not been able to make better intervals for the values, such that for instance zero-values are white on the map. The skewed proportions between legend and map are I think probably the result of Rstudio.

So I wondered whether someone has better suggestions for plotting choropleth maps in R using custom breaks?

EDIT 20-02-2013

As suggested I have used ggplot2 for creating a choropleth map. I've used the code below. I encountered two problems. One is that after fortifying the data not all data from the shapefile is transferred to the data frame. So I have to rename the "id" variable such that I can merge the data. Minor issue.

A bigger issue is that despite the fact that the data in the dataframe is correct, the ggplot shows a map where the data is incorrect. The colour value according to the legend does not correspond with the value of the variable in the data. Am I overlooking something here?

# Plotting polygon shapefiles (try-out)
# Load the packages
require(rgdal)
require(maptools)
require(ggplot2)
gpclibPermit() # required for the fortify method
# Load the shapefile
africa=readOGR("/home/GIS",layer="africaII")
names(africa) 
# Load the data on conflict
conflict<-read.csv("africa_conflict.csv", header=TRUE)
names(conflict)
# Merge the data together
africadat<-merge(africa, conflict, by="ISO3")
# Changing the data in the shapefile
africa@data <- africadat
# Fortify so that ggplot can plot
africa.points = fortify(africa,region="ISO3")
# Change "id" to "ISO3"such that the datasets can be merged again
names(africa.points)[names(africa.points)=="id"]<-"ISO3"
africa.df=merge(africa.points,africadat,by="ISO3")
# Plot the data
ggplot(africa.df) + 
aes(long,lat,group=group) + 
geom_polygon(aes(fill =onset))+
geom_path(color="white") +
coord_equal()

Link to replication data

Update 13-11-2015: Fixed link to shape file. See answer below for solution on how to plot the choropleth.

Answer 1

I don't see a problem with the above code except that you have 6 clear levels in onset, so I would treat them as factors rather than as a continuous variable. Note the factor call in the code below.

ggplot(africa.df) +
  aes(long,lat,group=group) +
  geom_polygon(aes(fill = factor(onset)))+
  geom_path(color="white") +
  coord_equal()

That code gives the following result:

If you want custom colours you could try this:

library(RColorBrewer)
my.cols <- brewer.pal(6, "Blues")

ggplot(africa.df) +
  aes(long,lat,group=group) +
  geom_polygon(aes(fill = factor(onset)))+
  scale_fill_manual("Onset\n", values = my.cols, guide = "legend") +
  geom_path(color="white") +
  coord_equal()

Which produces this:

Answer 2

The classInt package has a number of functions for computing intervals for numeric data - install it and read the docs.

Answer 3

For the archives, the solution I used:

#### Choropleth in R example ####

## Libraries
library(classInt)    
library(lattice)
library(RColorBrewer)    
library(sp)

## Shapefile for Sub-Sahara Africa (see link in Q)
africa<-readShapeSpatial("shp/africa.shp") 

## Add data to shapefile
# In this case the number of conflict onsets between 1981-2010
africa$onset<-c(3,0,3,1,3,1,4,2,3,0,
                2,0,3,1,1,0,2,1,1,1,
                2,0,3,0,1,0,0,4,2,1,
                3,0,1,0,6,1,3,1,1,0,
                0,1,1,0,0,0,0)

## Plot 
par(mar=c(1,4,1,1),family="serif")
trellis.par.set(axis.line=list(col=NA)) # Remove borders
spplot(africa,zcol="onset",col.regions=colorRampPalette(c("white","grey10"))(20),
main=list(label="Number of civil conflict onsets 1981-2010",cex=3),
pretty=T,par.settings=list(axis.text=list(cex=3)))