[R] Coloring counties on a full US map based on a certain criterion

[Dimitri Liakhovitski]

Sarah, this is amazing, thank you so much.
One question: I am trying to do for the whole US (on one map) what
you've helped me do for Iowa.
In other words, I would like to create a file of inputs like you
"countycol " with 1,000+ lines - for all US counties (probably without
Hawaii and Alaska, right?) and then somehow feed it into the map
command but so that the output is the whole map of the US, and not one
state. Is it at all possible?
Or maybe it's possible to create 48 colored state maps one by one -
the way you showed me - save them, and then somehow "paste" those
states onto the whole US map?
Thanks a lot for your help!
Dimitri

On Fri, Jan 13, 2012 at 12:05 PM, Sarah Goslee <sarah.goslee at gmail.com> wrote:
> Hi,
>
> You've just about got it. See below.
>
> On Fri, Jan 13, 2012 at 11:52 AM, Dimitri Liakhovitski
> <dimitri.liakhovitski at gmail.com> wrote:
>> Dear Rers,
>>
>> is there a way to color counties on a full US map based on a criterion
>> one establishes (i.e., all counties I assign the same number should be
>> the same color)?
>> I explored a bit and looks like the package "maps" might be of help.
>> library(maps)
>> One could get a map of the US: map('usa')
>> One could get countries within a US state: map('county', 'iowa', fill
>> = TRUE, col = palette())
>
> Using a random sampling to give you the basic idea.
> There are 99 counties in Iowa, so to construct the criterion:
> countycol <- sample(1:5, 99, replace=TRUE)
> And to invent a set of colors (RColorBrewer is a better choice for
> final maps):
> classcolors <- rainbow(5)
>
> then you can use them in your map just as you would for any other
> plotting command:
>
> map('county', 'iowa', fill= TRUE, col = classcolors[countycol])
>
>> Would it be possible to read in a file with counties and their
>> assignments (some counties have a 1, some counties have a 2, etc.) and
>> then have one map of the US with counties colored based on their
>> assignment?
>
> Absolutely. The only thing you have to watch out for is that you put your
> values in the same order as:
> map('county', 'iowa', plot=FALSE)$names
>
> Sarah
>
> --
> Sarah Goslee
> http://www.functionaldiversity.org



-- 
Dimitri Liakhovitski
marketfusionanalytics.com