R语言中使用dplyr对含NA值的列做除法失败，求解决方法

Hey there! Let's sort out that division problem you're having with NA values in your data frame. The trouble with your original code df %>% mutate(c_n = c / n, na.rm = TRUE) is that na.rm = TRUE isn't a valid argument for mutate() itself — that parameter works with aggregation functions like sum() or mean(), but not directly with basic arithmetic operations like division.

Here are a few practical solutions to handle NA values correctly during the division:

Solution 1: Skip Rows with NA Using ifelse() #

This approach checks if either c or n is NA before performing the division. If either value is missing, it returns NA for the result; otherwise, it calculates the division:

library(dplyr)

df %>% 
  mutate(c_n = ifelse(is.na(c) | is.na(n), NA, c / n))

Solution 2: Clean NA Values First (Use Carefully) #

If you want to replace NA values with a default value before dividing (make sure this makes sense for your data!), you can use replace_na(). For example, if you want to treat missing c as 0 and avoid division by zero by replacing missing n with 1:

df %>% 
  mutate(
    c_clean = replace_na(c, 0),
    n_clean = replace_na(n, 1), # Adjust this default based on your data logic
    c_n = c_clean / n_clean
  ) %>%
  select(-c_clean, -n_clean) # Optional: Remove temporary columns

Solution 3: Row-Wise Check with rowwise() #

If you prefer a more explicit row-wise check, you can use rowwise() to evaluate each row individually:

df %>%
  rowwise() %>%
  mutate(c_n = if(sum(is.na(c), is.na(n)) == 0) c/n else NA) %>%
  ungroup()

All these methods will correctly handle NA values in your c and n columns, so you'll get valid division results where both values exist, and NA where either is missing.

内容的提问来源于stack exchange，提问作者Tiptop