current community

your communities

more stack exchange communities

Stack Exchange
tour help
careers 2.0
Take the 2-minute tour ×
Stack Overflow is a question and answer site for professional and enthusiast programmers. It's 100% free, no registration required.
up vote 3 down vote favorite
2

Hello stackoverflow community!

Could someone kindly please assist me as I am having some difficulties with creating a choropleth map in R. As of now, I have assigned LL information to my points of interest, and I now want to create a choropleth map using the "cans" variable in the dataset (data.csv) by high school district area (highdist_n83.shp.zip). What I would like to please know is how to properly fill the map with the sum of cans per district area. I have provided code, which pulls a sample datafile from dropbox and the shape file I would like to use.

EDIT Sorry, I forgot to add that when I plot just the shape file, I am able to see it rendered via ggplot. However, when I try to "fill" the districts using the number of "cans" variable, R hangs for awhile before rendering what appears to be a mass of lines over the original shape. I wonder if the error is due to the following possible reasons

  1. the shape file is not good
  2. there may be an issue with how I am merging the data frame and the shape file as I've noticed additional rows are added to the merged file
  3. there are multiple schools in an area, which I haven't combined when using ddply.

Thank you for your time!

###load R scripts from dropbox
dropbox.eval <- function(x, noeval=F) {
require(RCurl)
intext <- getURL(paste0("https://dl.dropboxusercontent.com/",x), ssl.verifypeer = FALSE)
intext <- gsub("\r","", intext)
if (!noeval) eval(parse(text = intext), envir= .GlobalEnv)
return(intext)
}

##pull scripts from dropbox 
dropbox.eval("s/wgb3vtd9qfc9br9/pkg.load.r")    
dropbox.eval("s/tf4ni48hf6oh2ou/dropbox.r")

##load packages
pkg.load(c(ggplot2,plyr,gdata,sp,maptools,rgdal,reshape2))

###setup data frames
dl_from_dropbox("data.csv","dx3qrcexmi9kagx")
    data<-read.csv(file='data.csv',header=TRUE)

###prepare GIS shape and data for plotting
dropbox.eval("s/y2jsx3dditjucxu/dlshape.r")     
temp <- tempfile()
dlshape(shploc="http://files.hawaii.gov/dbedt/op/gis/data/highdist_n83.shp.zip", temp)
shape<- readOGR(".","highdist_n83") #HDOE high school districts  
shape@proj4string 

shape2<- spTransform(shape, CRS("+proj=longlat +datum=NAD83"))

data.2<-ddply(data, .(year, schoolcode, longitude, latitude,NAME,HDist,SDist), summarise,
        total = sum(total),
        cans= sum(cans))

coordinates(data.2) <-~longitude + latitude
shape2.df<-fortify(shape2)
mshape2.df<- merge(shape2.df,shape2@data, by.x="id", by.y="ID",all=TRUE)

newdata <- merge(mshape2.df,data.2, by.x="NAME", by.y="NAME", all=TRUE)
newdata  <- newdata [order(newdata $order),]

###choropleth map: 
mapPlot <- ggplot(newdata,aes(x=long, y=lat,drop=FALSE)) +
geom_polygon(aes(fill=cans, drop=FALSE), colour = "black", lwd = 1/9,na.rm=FALSE)
+ ggtitle("Total of Cans Across State")
print(mapPlot)
share|improve this question
3  
Thank you for the code. Before we start, can you tell us specifically what the problem is and/or include a screenshot. Do you get a warning message, or does it not plot or is the plot wrong? etc etc –  SlowLearner Jun 12 '13 at 8:38
    
Hi SlowLearner. I've edited my original message to describe the issue I am experiencing. It is quite long, but I hope it might clear up what I need help with. After running ggplot code, R basically hangs for awhile before rendering what looks to be a massive number of lines over the shape file. I've listed some possible issues, but then again I am a complete novice at this stuff... Thank you again for your time! :) –  user2434624 Jun 12 '13 at 9:10

1 Answer 1

up vote 6 down vote accepted

When you fortify your shapefile you end up with the wrong IDs. Instead, explicitly add an ID column based on the row names of the shapefile and use this to merge. You also do not tell geom_polygon how to group the rows in your data.frame so they are all plotted as one continuous overlapping, self-intersecting polygon. Add a group argument to geom_polygon. I also like to use RColorBrewer to pick nice colours for maps (using the function brewer.pal).

Try this:

require(RColorBrewer)
shape2@data$id <- rownames(shape2@data)
sh.df <- as.data.frame(shape2)
sh.fort <- fortify(shape2 , region = "id" )
sh.line<- join(sh.fort, sh.df , by = "id" )


mapdf <- merge( sh.line , data.2 , by.x= "NAME", by.y="NAME" , all=TRUE)
mapdf <- mapdf[ order( mapdf$order ) , ]

ggplot( mapdf , aes( long , lat ) )+
  geom_polygon( aes( fill = cans , group = id ) , colour = "black" )+
  scale_fill_gradientn( colours = brewer.pal( 9 , "Reds" ) )+
  coord_equal()

enter image description here

share|improve this answer
    
Thank you! I would have never figured it out without your help! It worked perfectly, and the RColorBrewer package is a huge improvement visually. –  user2434624 Jun 12 '13 at 18:37
1  
@user2434624 I'm glad this helped you. Yes the ColorBrewer package is great! –  Simon O'Hanlon Jun 12 '13 at 18:37
    
Sorry, this just popped up in my head... I'm not sure if this requires asking a new question, but how can I turn the fortified/merged file back into a shape file? Specifically, I would like to write it out as GeoJSON data using the writeOGR() function... –  user2434624 Jun 14 '13 at 0:18
    
This would be a new question. –  SlowLearner Jun 15 '13 at 16:47
    
Thank for your solution. It works well but the use of the option region in the sh.fort <- fortify(shape2 , region = "id" ) can generate the following error : Error: isTRUE(gpclibPermitStatus()) is not TRUE. Look at this question to solve this problem : link –  Giuliano yesterday

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.