I am essentially trying to merge information from two columns in the following manner: if column hf is 0 and hf1 is 1 then create new column combining both information.
df <- structure (list(subject_id = c("191-5467", "191-6784", "191-3457", "191-0987", "191-1245", "191-2365"), hf = c("1","0","0","0","1","0"), hf1 = c("NA","1","1","1","NA","0")), class = "data.frame", row.names = c (NA, -6L))
desired output:
| subject_id | hf | hf1 | hf2 |
|---|---|---|---|
| 191-5467 | 1 | NA | 1 |
| 191-6784 | 0 | 1 | 1 |
| 191-3457 | 0 | 1 | 1 |
| 191-0987 | 0 | 1 | 1 |
| 191-1245 | 1 | NA | 1 |
| 191-2365 | 0 | 0 | 0 |
What I've tried:
df$hf2 <- with(df, (!(df$hf1 == 1 & df$hf1 == 1)))
Output
| subject_id | hf | hf1 | hf2 |
|---|---|---|---|
| 191-5467 | 1 | NA | 1 |
| 191-6784 | 0 | 1 | 1 |
| 191-3457 | 0 | 1 | 1 |
| 191-0987 | 0 | 1 | 1 |
| 191-1245 | 1 | NA | 1 |
| 191-2365 | 0 | 0 | 1 |
Which is close but incorrect because 191-2365 hf2 should have remained 0.
CodePudding user response：
The structure was created with columns as character, It should be numeric. We may use pmax on the numeric columns
df <- type.convert(df, as.is = TRUE)
df$hf2 <- with(df, pmax(hf, hf1, na.rm = TRUE))
-output
> df
subject_id hf hf1 hf2
1 191-5467 1 NA 1
2 191-6784 0 1 1
3 191-3457 0 1 1
4 191-0987 0 1 1
5 191-1245 1 NA 1
6 191-2365 0 0 0