Suppress missing values in Tabyl xtabs in R

Question

However, I can't figure out how to suppress the NA from the denominator!

Please see here for the data:

df <- data.frame(col1 = c(1,1,2,2,1, NA,NA),
                 col2 = c("this", NA,"is", "text",NA,NA,'yes'), 
                 col3 = c('TRUE', 'FALSE', NA, 'TRUE','TRUE', 'TRUE', 'FALSE'), 
                 col4 = c(2.5, 4.2, 3.2, NA, 4.2, 3.2,3)) %>% 
                  mutate_if(is.numeric, as.factor) %>% 
                  mutate_if(is.character, as.factor) 

str(df)  
df %>% 
  tabyl(col1, col3, show_missing_levels = F) %>% 
  adorn_percentages("row")%>% 
  adorn_pct_formatting(digits = 2) %>%
  adorn_ns()

Note how the percent of the NA is still showing up in the denominator. I don't want to see the NA % at all in the cross tab:

 col1      FALSE       TRUE        NA_
    1 33.33% (1) 66.67% (2)  0.00% (0)
    2  0.00% (0) 50.00% (1) 50.00% (1)
 <NA> 50.00% (1) 50.00% (1)  0.00% (0)

What I want to see:

 col1      FALSE       TRUE       
    1 33.33% (1) 66.67% (2)  
    2  0.00% (0) 100.00% (1)

Any idea how I can achieve this?

akrun · Accepted Answer · 2021-04-16T17:48:40.557

4

By default, show_na = TRUE in tabyl. If we change it to FALSE, the OP's code should work

library(dplyr)
library(janitor)
 df %>% 
   tabyl(col1, col3, show_missing_levels = FALSE, show_na = FALSE) %>%
   adorn_percentages("row")%>% 
   adorn_pct_formatting(digits = 2) %>%
   adorn_ns()

-output

#  col1      FALSE        TRUE
#    1 33.33% (1)  66.67% (2)
#    2  0.00% (0) 100.00% (1)

Earlier, was thinking about injecting na_omit to remove the NA rows and select to select the columns of interest. But, this will also changes/remove the attribute, thus making the adorn_ns to not work

edited Apr 16 '21 at 17:48

answered Apr 16 '21 at 17:35

akrun

874,273
37
540
662

I want the NA removed from the denominator, this just hides the NA columns-- for example second column, second row should be 100%, not 50% – NewBee Apr 16 '21 at 17:37
1

@akrun should it not be `na.omit`? – TarJae Apr 16 '21 at 17:42
was just on the way with the same solution. but you were faster! – TarJae Apr 16 '21 at 17:43
1

@TarJae it was from `collapse`. but you could use `na.omit` from `base R` – akrun Apr 16 '21 at 17:51
Thank you akrun. maybe `digits = 0`? – TarJae Apr 16 '21 at 17:53
1

@TarJae i think the OP wanted 2 digits rounded as in his expeced – akrun Apr 16 '21 at 17:54

score 1 · Answer 2 · answered Apr 16 '21 at 17:46

1

It is almost the same as akrun's

library(dplyr)
library(janitor)
df %>% 
  tabyl(col1, col3, show_missing_levels = FALSE) %>% 
  na.omit() %>% 
  select(-NA_) %>%
  adorn_percentages("row")%>% 
  adorn_pct_formatting(digits = 0)

Output:

 col1 FALSE TRUE
    1   33%  67%
    2    0% 100%

answered Apr 16 '21 at 17:46

TarJae

72,363
6
19
66

This is close,but I also need to see the N's (please see my desired output)! adorn_ns() %>% does not work with this – NewBee Apr 16 '21 at 18:02

Suppress missing values in Tabyl xtabs in R

2 Answers2