Binding and ploting SF objects - r

Question

I have a data set (Data_Base) with the number of murders in one specific year for the 32 Mexican states. I download with the help of the gadm_sf_loadCountries the spatial data for this 32 Mexican states.
Now, I want to merge this data frame with the SF object in order to plot it with filled with the murders per state. The code is the following:
#Export spatial data from Mexico
mex.sf = gadm_sf_loadCountries("MEX", level=1, basefile="./")
mex.sf <- mex.sf$sf
#Merge the SF object and the Data frame
mex.sf <- left_join(Data_Base,mex.sf, by = c('State'='shapeName'))
#Run the graph code, I want to fill the map with the murders per state
ggplot(mex.sf)+geom_sf(aes(fill=murders)) +
scale_fill_manual(values=c("#d7301f","#fc8d59","#fdcc8a","#fef0d9"))
However, at the moment I do the graph I got the following error:
Error: stat_sf requires the following missing aesthetics: geometry
Run `rlang::last_error()` to see where the error occurred.
I honestly don´t know why is showing me this error because the geometry rows have been pasted at the moment I "left joined"it.
Does anyone knows why is this error?

Answer 1

TLDR: change the order of your objects in left_join(Data_Base,mex.sf, ...)
Long version: When joining data items to a {sf} spatial object the order of your objects matters.
The class output of a dplyr::*_join() family of functions (inner, left, whatever...) will be determined by the first argument. The presence of geometry column is not sufficient for the resulting object to behave as expected; class matters.
To illustrate my argument consider this piece of code (I don't have your data on hand, so I make do with the North Carolina shapefile that ships with {sf}):
library(sf)
library(dplyr)
shape <- st_read(system.file("shape/nc.shp", package="sf"))
data <- data.frame(CNTY_ID = 1825,
label = "this be county Ashe")
data_first <- inner_join(data, shape, by = "CNTY_ID") # joining shape to data = this will break later on
class(data_first)
# [1] "data.frame"
plot(data_first["label"]) # this will be a mess...
# Error in plot.window(...) : need finite 'ylim' values
# In addition: Warning messages:
# 1: In xy.coords(x, y, xlabel, ylabel, log) : NAs introduced by coercion
# 2: In min(x) : no non-missing arguments to min; returning Inf
# 3: In max(x) : no non-missing arguments to max; returning -Inf
shape_first <- inner_join(shape, data, by = "CNTY_ID") # joining data to shape = this is the way to go!
class(shape_first)
# [1] "sf" "data.frame"
plot(shape_first["label"]) # works like a charm...