Skip to content

Plot by FIPS Code using Shapefiles

8 messages · Anthony Damico, Shouro Dasgupta, Corey Sparks +1 more

Original
Shouro Dasgupta
# 
I am trying to plot data by FIPS code using county shapes files.

library(data.table)
I have data by FIPS code which looks like this:
I add leading zeroes by:

max_change <- as.data.table(max_change)
max_change$FIPS <- sprintf("%05d",as.numeric(max_change$FIPS))

I downloaded shapefiles from here:
ftp://ftp2.census.gov/geo/tiger/TIGER2014/COUNTY/.

I obtain the FIPS codes from the shapefiles and order them using:

shapes_fips <- shapes$GEOID
Then I merge the FIPS codes with my original dataset using:
Which looks like this:

structure(list(FIPS = c("01001", "01003", "01005", "01007", "01009",
But when I try to merged data back to the SpatialPolygonsDataFrame called
shapes, I get the following error:

shapes$change <- merged_data$change

Error in `[[<-.data.frame`(`*tmp*`, name, value = c(-1.47141054005866,  :
Apologies for the messy example, what am I doing wrong? Any help will be
greatly appreciated. Thank you!

Sincerely,

Shouro
Anthony Damico
# 
hi, after running each individual line of code above, check that the object
still has the expected number of records and unique county fips codes.  it
looks like length( shapes$GEOID ) == 3233 but nrow( merged_data ) == 3109.
the way for you to debug this is for you to go through line by line after
creating each new object  :)

i'm also not sure it's safe to work with gis objects as you're doing, there
are some well-documented examples of working with tiger files here
https://github.com/davidbrae/swmap
On Tue, May 5, 2015 at 11:00 AM, Shouro Dasgupta <shouro at gmail.com> wrote:

            

      
      
      
    

        

      
      
    

        

  
  


  


      

    

    

      
      

        


  

        

      Shouro Dasgupta
    

    

      
      #
    

  


  

      
Hello,

Thank you for your reply. My original data has 3109 FIPS codes. Is there a
way to merge only this data into the shapefiles? I hope I am clear.

Thank you also for the link, I am trying to do something like this:
https://gist.github.com/reubano/1281134.

Thanks again!

Sincerely,

Shouro

        
On Tue, May 5, 2015 at 5:21 PM, Anthony Damico <ajdamico at gmail.com> wrote:

        

            

      
      
      
    

        

      
      
    

            

      
      
      
    

        

  
    

      
      
    

  


  


      

    

    

      
      

        


  

        

      Anthony Damico
    

    

      
      #
    

  


  

      
so check the unique number of fips codes in the objects before and after

            

      
      
      
    

        
also note that all.X should be all.x and you might want to use FALSE for
one or both of those

        
On Tue, May 5, 2015 at 11:40 AM, Shouro Dasgupta <shouro at gmail.com> wrote:

        

            

      
      
      
    

        

      
      
    

            

      
      
      
    

        

  
  


  


      

    

    

      
      

        


  

        

      Corey Sparks
    

    

      
      #
    

  


  

      
Joining data the way you're doing it is dangerous, Roger Bivand and others
describes a standard way to do this process here:
http://r-sig-geo.2731867.n2.nabble.com/Merging-shapefiles-and-csv-td7586839.html

And I do an example using US Census data here, using merge():
http://spatialdemography.org/wp-content/uploads/2013/04/9.-Sparks.pdf
<http://spatialdemography.org/wp-content/uploads/2013/04/9.-Sparks.pdf>  

look at page 134 of that pdf.

Hope this helps



-----
Corey Sparks, PhD
Assistant Professor
Department of Demography
University of Texas at San Antonio
501 West C?sar E. Ch?vez  Blvd 
Monterey Building 2.270C
San Antonio, TX 78207
210-458-3166
corey.sparks 'at' utsa.edu
coreysparks.weebly.com
--
View this message in context: http://r.789695.n4.nabble.com/Plot-by-FIPS-Code-using-Shapefiles-tp4706830p4706840.html
Sent from the R help mailing list archive at Nabble.com.

  
  


  


      

    

    

      
      

        


  

        

      Roger Bivand
    

    

      
      #
    

  


  

      
Corey Sparks <corey.sparks <at> utsa.edu> writes:

            

      
      
      
    

        
http://r-sig-geo.2731867.n2.nabble.com/Merging-shapefiles-and-csv-td7586839.html


Quite right - the chunks Corey is referring to are:

Please do refer to the vignette in the maptools package, and to previous
threads which have advised that merge() should not be used, and that the
row.names of the data frames be used as ID keys. Typically using match() on
the row.names of the two objects will show which are not correctly aligned.

and

Beware that the data from the objects may be jumbled - never use merge,
always use match() on the row.names vectors of the objects to ensure that
the key-IDs agree. Jumbled data happens, it is important not to think
"shapefile" but to think DBMS with the ID key your way of staying sane. 

The maptools vignette is at:

http://cran.r-project.org/web/packages/maptools/vignettes/combine_maptools.pdf

or:

library(maptools)
vignette("combine_maptools")

Here I also suspect that you'll find that there are non-unique FIPS in the
county polygons file, so may need to go through
maptools::unionSpatialPolygons() first.

Roger

            

      
      
      
    

        
http://r.789695.n4.nabble.com/Plot-by-FIPS-Code-using-Shapefiles-tp4706830p4706840.html

            

      
      
      
    

        

      
      
    

  
  


  


      

    

    

      
      

        


  

        

      Shouro Dasgupta
    

    

      
      #
    

  


  

      
Dear Anthony,

Thanks again for your reply. The following worked for merge:

merged_data <- merge(shapes_fips,max_change,by="FIPS",all.x=T, all.y=F)


 However, I think I am doing something wrong - as I have 3109 FIPS code in
my original data but when I merge with the shapes
file SpatialPolygonsDataFrame, its not merging properly, many NA.

Is it a good idea to convert the shapefiles into data.frame/data.table for
merging and then transform it back to shapefiles? This is what I have been
doing:

shapes <- readShapePoly("F:/GCM//tl_2014_us_county/tl_2014_us_county.shp")

            

      
      
      
    

        
Thanks again!

Sincerely,

Shouro

        
On Tue, May 5, 2015 at 6:00 PM, Anthony Damico <ajdamico at gmail.com> wrote:

        

            

      
      
      
    

        

      
      
    

            

      
      
      
    

        

  
    

      
      
    

  


  


      

    

    
1 day later

    

      
      

        


  

        

      Shouro Dasgupta
    

    

      
      #
    

  


  

      
Excellent suggestions Professors! I really appreciated it. This is what I
am using:

library(data.table)

            

      
      
      
    

        
max_change is my csv file while shapes is my spatial data:

max_change <-read.csv ("F:/GCM/max_change.csv")

            

      
      
      
    

        
I am plotting the results now. I am getting different plots regarding
county boundaries (most likely because I am doing something wrong). When I
use the *plot *command:

colours = brewer.pal(6,"PuRd")

            

      
      
      
    

        
I get a "beautiful" plot with the county boundaries clearly separated but
for *spplot*;

pal = brewer.pal(6,"Reds")

            

      
      
      
    

        
The county boundaries/FIPS are not defined. What am I doing wrong? Thanks
again!

Sincerely,

Shouro

        
On Wed, May 6, 2015 at 12:23 PM, Roger Bivand <Roger.Bivand at nhh.no> wrote: