Creating counties spatial dataframe produces columns with NAs where data should have been, unlike the states spatial dataframe

Question

I am trying to replicate what I did for state level and works to county level. Once I have the states spatial data frame, I convert it to a simple features object because I can treat that more like a traditional data frame. Now it can be left_joined to our USArrests data set. When I am tryin to repeat the same process for county level I geta final dataframe which has NAs instead of the counties data :

library(tidyverse)
library(sf)
library(leaflet)
library(tigris)
options(tigris_use_cache = TRUE)

data("USArrests", package = "datasets")
USArrests <- USArrests %>% as_tibble(rownames = "state")
states_sf <- tigris::states() %>% 
  as("sf") %>% 
  rename(state=NAME) %>% 
  left_join(USArrests) %>% 
  na.omit()
#> Joining, by = "state"

states_sf
#> Simple feature collection with 50 features and 18 fields
#> geometry type:  MULTIPOLYGON
#> dimension:      XY
#> bbox:           xmin: -179.2311 ymin: 18.86546 xmax: 179.8597 ymax: 71.44106
#> epsg (SRID):    4269
#> proj4string:    +proj=longlat +ellps=GRS80 +towgs84=0,0,0,0,0,0,0 +no_defs
#> First 10 features:
#>    REGION DIVISION STATEFP  STATENS GEOID STUSPS          state LSAD MTFCC
#> 1       3        5      54 01779805    54     WV  West Virginia   00 G4000
#> 2       3        5      12 00294478    12     FL        Florida   00 G4000
#> 3       2        3      17 01779784    17     IL       Illinois   00 G4000
#> 4       2        4      27 00662849    27     MN      Minnesota   00 G4000
#> 5       3        5      24 01714934    24     MD       Maryland   00 G4000
#> 6       1        1      44 01219835    44     RI   Rhode Island   00 G4000
#> 7       4        8      16 01779783    16     ID          Idaho   00 G4000
#> 8       1        1      33 01779794    33     NH  New Hampshire   00 G4000
#> 9       3        5      37 01027616    37     NC North Carolina   00 G4000
#> 10      1        1      50 01779802    50     VT        Vermont   00 G4000
#>    FUNCSTAT        ALAND      AWATER    INTPTLAT     INTPTLON Murder
#> 1         A  62265597146   489902816 +38.6472854 -080.6183274    5.7
#> 2         A 138924199212 31386038155 +28.4574302 -082.4091478   15.4
#> 3         A 143788697679  6206693598 +40.1028754 -089.1526108   10.4
#> 4         A 206232257655 18929176411 +46.3158148 -094.1996628    2.7
#> 5         A  25147754905  6983312282 +38.9466584 -076.6744939   11.3
#> 6         A   2677898725  1323551636 +41.5974187 -071.5272723    3.4
#> 7         A 214042908012  2398669593 +44.3484222 -114.5588538    2.6
#> 8         A  23187396994  1028678842 +43.6726907 -071.5843145    2.1
#> 9         A 125921301190 13470062955 +35.5397100 -079.1308636   13.0
#> 10        A  23873467535  1031124865 +44.0604795 -072.6733274    2.2
#>    Assault UrbanPop Rape                       geometry
#> 1       81       39  9.3 MULTIPOLYGON (((-81.74725 3...
#> 2      335       80 31.9 MULTIPOLYGON (((-82.98624 2...
#> 3      249       83 24.0 MULTIPOLYGON (((-91.18529 4...
#> 4       72       66 14.9 MULTIPOLYGON (((-96.78438 4...
#> 5      300       67 27.8 MULTIPOLYGON (((-77.45881 3...
#> 6      174       87  8.3 MULTIPOLYGON (((-71.67264 4...
#> 7      120       54 14.2 MULTIPOLYGON (((-116.8997 4...
#> 8       57       56  9.5 MULTIPOLYGON (((-72.3299 43...
#> 9      337       45 16.1 MULTIPOLYGON (((-82.41674 3...
#> 10      48       32 11.2 MULTIPOLYGON (((-73.31328 4...

When I try to do the same with counties :

library(noncensus)
data(counties)

counties_sf <- counties %>% as_tibble()
counties_sf <- tigris::counties() %>% 
  as("sf") %>% 
  rename(county_name=NAME) %>% 
  left_join(counties)


counties_sf

Answer 1

In general it is not a good idea to merge data using names. In your case the county names are completely different in this two data frames. See head(counties$county_name) and head(counties_sf$NAME) .

To avoid the use of the names you can use the FIPS code in this case.

library(noncensus)
library(sf)
library(tigris)

data(counties)
counties <- counties %>%
  mutate(GEOID = paste0(state_fips, county_fips))

counties_sf <- tigris::counties() %>% 
  as("sf") %>% 
  left_join(counties, by = "GEOID")

head(counties_sf)

Simple feature collection with 6 features and 25 fields
geometry type:  MULTIPOLYGON
dimension:      XY
bbox:           xmin: -123.7283 ymin: 18.11774 xmax: -65.81565 ymax: 46.38562
epsg (SRID):    4269
proj4string:    +proj=longlat +ellps=GRS80 +towgs84=0,0,0,0,0,0,0 +no_defs
  STATEFP COUNTYFP COUNTYNS GEOID        NAME              NAMELSAD LSAD CLASSFP MTFCC CSAFP CBSAFP METDIVFP FUNCSTAT      ALAND   AWATER    INTPTLAT
1      31      039 00835841 31039      Cuming         Cuming County   06      H1 G4020  <NA>   <NA>     <NA>        A 1477895811 10447360 +41.9158651
2      53      069 01513275 53069   Wahkiakum      Wahkiakum County   06      H1 G4020  <NA>   <NA>     <NA>        A  680956787 61588406 +46.2946377
3      35      011 00933054 35011     De Baca        De Baca County   06      H1 G4020  <NA>   <NA>     <NA>        A 6016761713 29147306 +34.3592729
4      31      109 00835876 31109   Lancaster      Lancaster County   06      H1 G4020   339  30700     <NA>        A 2169240199 22877180 +40.7835474
5      31      129 00835886 31129    Nuckolls       Nuckolls County   06      H1 G4020  <NA>   <NA>     <NA>        A 1489645187  1718484 +40.1764918
6      72      085 01804523 72085 Las Piedras Las Piedras Municipio   13      H1 G4020   490  41980     <NA>        A   87748363    32509 +18.1871483
      INTPTLON           county_name state state_fips county_fips fips_class  CSA  CBSA population                       geometry
1 -096.7885168         Cuming County    NE         31         039         H1 <NA>  <NA>       9139 MULTIPOLYGON (((-97.01952 4...
2 -123.4244583      Wahkiakum County    WA         53         069         H1 <NA>  <NA>       3978 MULTIPOLYGON (((-123.4364 4...
3 -104.3686961        De Baca County    NM         35         011         H1 <NA>  <NA>       2022 MULTIPOLYGON (((-104.5674 3...
4 -096.6886584      Lancaster County    NE         31         109         H1  339 30700     285407 MULTIPOLYGON (((-96.9106 40...
5 -098.0468422       Nuckolls County    NE         31         129         H1 <NA>  <NA>       4500 MULTIPOLYGON (((-98.27367 4...
6 -065.8711890 Las Piedras Municipio    PR         72         085         H1 <NA>  <NA>         NA MULTIPOLYGON (((-65.91048 1...