R dplyr删除“ <NA> 表格中的“列”

Question

Given data

d <- data.frame(g = sample(c(1:5,NA),100, replace = T)
                ,o = sample(c("yes","no",NA),100, replace = T))

Is there a one step approach to the following:

s <- d %>% group_by(g, o) %>% 
  summarise(n = n()) %>%
  ungroup() %>%
  spread(o, n, fill=0) 
s %>% select(which(names(s)!="<NA>"))

If I do this:

s <- d %>% group_by(g, o) %>% 
  summarise(n = n()) %>%
  ungroup() %>%
  spread(o, n, fill=0) %>% select(-c("<NA>"))

I get the error:

Error in -c("<NA>") : invalid argument to unary operator

Answer 1

names(s) gives

"g"    "no"   "yes"  "<NA>"

As @Psidom says in the comment, the following works. It matters if you use quotes vs back ticks.

s <- d %>% group_by(g, o) %>% 
summarise(n = n()) %>%
ungroup() %>%
spread(o, n, fill=0) %>% select(-`<NA>`)