[R] plotting points on a map, assigning vectors to values from dataset
David L Carlson
dcarlson at tamu.edu
Sat Sep 15 20:33:08 CEST 2012
You didn't include the data so we can't really test anything, but
you can simplify things quite a bit by replacing the pch= and bg=
lines with
pch <- c(24, 23, 22, 21)
bg <- c("grey24", "black", "grey77", "white")
map$number is a factor which means your categories default to
alphabetical order so the first category (factor code 1) is
fourfive and the second category (factor code 2) is sixseven.
The above lines take this into account.
Then replace the whole for loop (you really need to learn about
vectorization in R) with a single function call:
points(map$longitude, map$latitude, pch=pch(as.numeric(map$number)),
col="black", bg=bg(as.numeric(map$number)), cex=1.5)
----------------------------------------------
David L Carlson
Associate Professor of Anthropology
Texas A&M University
College Station, TX 77843-4352
> -----Original Message-----
> From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-
> project.org] On Behalf Of Neele
> Sent: Friday, September 14, 2012 12:41 PM
> To: r-help at r-project.org
> Subject: [R] plotting points on a map, assigning vectors to values from
> dataset
>
> Dear everyone, as a newbie I fear I have made an easy mistake and am
> just too
> blind to see it. Can anyone help me seeing the (possibly) obvious?
> I have a command that creates a map of South America. In a separate
> txt-file I stored data about latitude and longitude of languages, and
> in a
> third column values that differ from language to language. The command
> should read that file and assign the values to pch vectors (with bg as
> well,
> because they are 21-24) and plot them on the map.
> Now, the map is properly displayed, and the location of the pch symbols
> is
> correct as well. But the actual symbols (and bg colors) do not match
> the
> original data in the txt-file. (e.g. a white circle should display
> languages
> with value "zeroone", and a black diamond those with "sixseven", but
> there
> are white diamonds for "zeroone". That is easy to change if I only had
> two
> values, but I have four, and whatever I change it never matches the
> actual
> data).
> I want to automatically match my values (e.g. zeroone, twothree etc.)
> with
> shapes from pch and background colors. That works fine in the legend,
> but
> not for the actual map. I guess there is something missing so that the
> command does not assign the correct vectors to the correlation rows of
> the
> matrix.
> I specified the pch and bg vectors and then tried (
> pch=pch[[map$number[i]]], bg=bg[[map$number[i]]]), because the header
> of the
> column is "number", to make it read the correct column, but I guess
> somewhere the link is missing.
> The legend is displayed correctly, the retrieval path for the txt.file
> is
> correct, and the .pdf is also created.
> I had problems because the lat/long values were read as factors and
> not as
> numericals at first; I changed that with gsub and it works. Do I
> perhaps
> have to change the values for the "number" column as well? (I put
> str(map)
> at the end below, it looks all right so far to me...)
>
> I hope you understand my problem, and like I said, it is probably
> something
> very easy and clear. I found lots of info about point, pch, bg etc.,
> but not
> of how to automatically assign values to points using a data file.
>
> I am grateful for any comment!
> Best wishes,
> Neele
>
>
> library(maps)
> library(mapdata)
> map = read.delim("C:/Users/Neele/Desktop/number_of_modals.txt",
> header=TRUE)
>
> map$longitude=gsub(",",".",map$longitude)
> map$longitude=as.numeric(as.character(map$longitude))
>
> map$latitude=gsub(",",".",map$latitude)
> map$latitude=as.numeric(as.character(map$latitude))
>
> xlim = c(-85,-35)
> ylim = c(-55,15)
>
> pch = list("zeroone"=21, "twothree"=22,"fourfive"=24,"sixseven"=23)
>
> bg =
> list("zeroone"="white","twothree"="grey77","fourfive"="grey24","sixseve
> n"="black")
>
> pdf('C:/Users/Neele/Desktop/number_of_modals.pdf', height=8,
> width=6)
>
> map("worldHires",xlim=xlim, ylim=ylim,cex.lab=0.5)
>
> map.axes()
>
> for (i in 1:nrow(map)) {
> points(map$longitude[i],map$latitude[i], *until here
> everything
> as it should be*
> pch=pch[[map$number[i]]], *but here there seems to be random
> assigning of vector to value*
> bg=bg[[map$number[i]]],
> col="black",
> cex = 1.5)
> }
> title(main="Number of marked Modals")
> legend("bottomright",
> legend=c("0-1","2-3","4-5","6-7"),
> title="Legend",
> inset=0.05,
> pch=c(21,22,24,23),
> pt.bg=c("white","grey77","grey24","black")
> )
>
> dev.off()
>
> > str(map)
> 'data.frame': 63 obs. of 6 variables:
> $ latitude : num -5 -9 1.5 -17 -13.1 ...
> $ longitude: num -78 -67 -78.2 -69 -64.2 ...
> $ family : Factor w/ 26 levels "Araucanian","Arawakan",..: 13 2 5 4
> 2 14
> 22 9 24 23 ...
> $ language : Factor w/ 63 levels "Aguaruna","Apurina",..: 1 2 3 4 5 6
> 7 8 9
> 10 ...
> $ number : Factor w/ 4 levels "fourfive","sixseven",..: 3 1 1 3 1 3
> 1 4 3
> 1 ...
> $ X : logi NA NA NA NA NA NA ...
> >
>
>
>
>
>
>
> --
> View this message in context: http://r.789695.n4.nabble.com/plotting-
> points-on-a-map-assigning-vectors-to-values-from-dataset-tp4643182.html
> Sent from the R help mailing list archive at Nabble.com.
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-
> guide.html
> and provide commented, minimal, self-contained, reproducible code.
More information about the R-help
mailing list