使用 ifelse() 的 R 语法有什么问题？

Question

I am trying to create a column after merging two date frames that indicates 1=successful match, 2=row from data frame 1 that didn't match, 3=row from data frame 2 that didn't match.

I created a column equal to 1 in each data frame before merging so that I could analyze the merge.

The merged data frame looks like this:

merge_s  merge_p
1        NA   
NA       1  
1        1

Here is my code:

pretest_merged$merge_result <- ifelse(pretest_merged$merge_p == 1 & pretest_merged$merge_s==1,1,
                                      ifelse(pretest_merged$merge_p == 1 & is.na(merge_s),2,
                                             ifelse(pretest_merged$merge_s==1 & is.na(merge_p),3,"error")))

The resulting data frame looks like this:

merge_s  merge_p  merge_result
1        NA       NA         
NA       1        NA      
1        1        1

When I want it to look like this:

merge_s  merge_p  merge_result 
1        NA       3     
NA       1        2            
1        1        1            

I've tried getting rid of the is.na() but that doesn't work. I've also tried using if_else() but it tells me that condition must be a logical vector and not a data frame.

What am I doing wrong? Sorry if this question is super basic but I can't think of what else to try. Thanks.

Answer 1

Without a sample dataset it's hard to help. You should add a full reprex .

However maybe this chunk could help:

library(tidyverse)

df1 <- data.frame(id1 = c(1,2,3), val1 = c('blue','red','green'))
df2 <- data.frame(id2 = c(2,4,7,3), val2 = c('red','pink','grey','green'))


df_output <- merge(df1, df2, by.x = "id1", by.y = "id2", all.x = TRUE, all.y = TRUE)

df_output %>% mutate (chkcol = 
                        
                        case_when(
                          is.na(val1) == FALSE & is.na(val2) == FALSE ~ 1,
                          is.na(val1) == TRUE & is.na(val2) == FALSE ~ 2,
                          is.na(val1) == FALSE & is.na(val2) == TRUE ~ 3
                        )
                        
                        )

df_output

  id1  val1  val2 chkcol
1   1  blue  <NA>      3
2   2   red   red      1
3   3 green green      1
4   4  <NA>  pink      2
5   7  <NA>  grey      2