CODEWITHSUNDEEP
rcrosstable
Misha
Misha

Reputation: 3126

Set NA to "missing" in crosstable

library(crosstable)
ct1 = crosstable(mtcars2, c(disp, vs), by=am, total="both", 
                 percent_pattern="{n} ({p_row}/{p_col})", percent_digits=0) %>%
    as_flextable()
ct1

How can I easily change the value label from "NA" into "missing" instead?

Upvotes: 1

Views: 71

Answers (2)

Adriano Mello
Adriano Mello

Reputation: 2132

ct1 is a "well behaved" df and if it's just a chr label, why not str_replace_all?

ct1 <- crosstable(
  mtcars2, c(disp, vs), by = am, 
  total="both", percent_pattern = "{n} ({p_row}/{p_col})", percent_digits = 0) 

new_ct1 <- ct1 %>% 
  mutate(across(
    where(is.character), 
    \(x) str_replace_all(x, "\\(NA\\)", "(missing)")))
  
as_flextable(new_ct1)

enter image description here

Upvotes: 3

stefan
stefan

Reputation: 125268

From my understanding the labels for the summary function are hard-coded in the default summary function cross_summary. But you can simply use copy &. paste to write your own version of the summary function or use a wrapper function to set your desired label:

library(crosstable)

cross_summary2 <- function(x, dig = 1, ...) {
  c(
    `Min / Max` = minmax(x, dig = dig, ...),
    `Med [IQR]` = mediqr(x,
      dig = dig, ...
    ), `Mean (std)` = meansd(x,
      dig = dig,
      ...
    ), `N (missing)` = nna(x)
  )
}

cross_summary3 <- function(x, dig = 1, ...) {
  xx <- cross_summary(x, dig = dig, ...)
  names(xx)[names(xx) == "N (NA)"] <- "N (missing)"
  xx
}

crosstable(mtcars2, c(disp, vs),
  by = am, total = "both",
  funs = c(` ` = cross_summary2),
  percent_pattern = "{n} ({p_row}/{p_col})",
  percent_digits = 0
) |> 
  as_flextable()


crosstable(mtcars2, c(disp, vs),
  by = am, total = "both",
  funs = c(` ` = cross_summary3),
  percent_pattern = "{n} ({p_row}/{p_col})",
  percent_digits = 0
) |> 
  as_flextable()

Upvotes: 4

Related Questions