The data set of shapefile "Property1A" is:
df
# suburb area asst
# 0 Te Aro 14.541780 R076
# 1 Te Aro 7.655428 R076
# 2 Te Aro 3189.976134 <NA>
# 3 Te Aro 242.173386 0132
# 4 Karori 537.697290 R003
# 5 Pipitea 1116.954993 R105
# 6 Mt Cook 106.099900 R151
# 7 Te Aro 83.307421 R057
# 8 Aro Valley 544.665788 R105
# 9 Aro Valley 59.726882 R110
When I try to subset the shapefile for suburb "Te Aro", the code I use is:
TeAro <- subset(Property1A, suburb == "Te Aro")
But there is an error:
TeAro <- subset(Property1A, suburb == "Te Aro")
# Error in eval(expr, envir, enclos) : object 'suburb' not found
Why there is an error?
subset with names should work. You don't need the shapefiles package if you are reading in with readOGR:
> require(rgdal)
Loading required package: rgdal
Loading required package: sp
rgdal: version: 0.8-14, (SVN revision 496)
Geospatial Data Abstraction Library extensions to R successfully loaded
Loaded GDAL runtime: GDAL 1.9.0, released 2011/12/29
Path to GDAL shared files: /usr/share/gdal/1.9
Loaded PROJ.4 runtime: Rel. 4.8.0, 6 March 2012, [PJ_VERSION: 470]
Path to PROJ.4 shared files: (autodetected)
Get the 169 polygons of Indian states into a spatial object:
> India = readOGR(".","india_state")
OGR data source with driver: ESRI Shapefile
Source: ".", layer: "india_state"
with 169 features and 3 fields
Feature type: wkbPolygon with 2 dimensions
Now subset by name:
> Gujarat = subset(India, NAME=="Gujarat")
> dim(Gujarat)
[1] 12 3
> dim(India)
[1] 169 3
I can plot these objects, they map nicely:
> plot(Gujarat)
Without your data or knowing what version of R and the packages you have, there is no way of knowing why this fails for you.
Package: rgdal Version: 0.8-14
Package: sp Version: 1.0-14
R version 3.0.2....
来源：https://stackoverflow.com/questions/25985395/how-to-subset-a-shapefile