Here is another way to do it using double bangs as a prefix to the variable and converting the data type to a df symbo

mtcars %>% 
  select(var) %>% 
  group_by(!! as.name(var)) %>% 
  tally()
  #summarise(n