how to compare factors?

[von_olaf]

Consider this simple example

> tibble(id = c(1,2,3,4),
+        score1 = c('good', 'bad', 'bad', 'ugly'),
+        score2 = c('good', 'ugly', 'good', 'bad'))
# A tibble: 4 x 3
     id score1 score2
  <dbl> <chr>  <chr> 
1     1 good   good  
2     2 bad    ugly  
3     3 bad    good  
4     4 ugly   bad  

Essentially, my scoring variable states that good > bad > ugly. Is there a way in R (maybe using ordered factors?) so that

mutate(mycomp = score1 >= score2)

will return the correct answer? That is TRUE, TRUE, FALSE, FALSE?

Thanks!

[tonasantiago]

Hi Olaf:

If you want to solve this by mutate() the best option, I think, is:

%>% mutate(mycomp = score1 == score2)

If you are trying to make an equal or higer (lower) test, the most elegant solution I can picture it's recode your factors to numeric values.

Regards,
Tona

[andresrcs]

@von_olaf

Is there a way in R (maybe using ordered factors?) so that

Yes, have you tried?

library(tidyverse)

df <- tibble(id = c(1,2,3,4),
       score1 = c('good', 'bad', 'bad', 'ugly'),
       score2 = c('good', 'ugly', 'good', 'bad'))

df %>% 
    mutate_at(vars(starts_with("score")),
              factor,
              levels = c("ugly", "bad", "good"),
              ordered = TRUE) %>% 
    mutate(mycomp = score1 >= score2)
#> # A tibble: 4 x 4
#>      id score1 score2 mycomp
#>   <dbl> <ord>  <ord>  <lgl> 
#> 1     1 good   good   TRUE  
#> 2     2 bad    ugly   TRUE  
#> 3     3 bad    good   FALSE 
#> 4     4 ugly   bad    FALSE

^{Created on 2020-04-20 by the reprex package (v0.3.0.9001)}