mydata.dt=
| Candidate.index | Score |
|---|---|
| 1 | 0: most of the time (5-7days/week) |
| 2 | 0: most of the time (5-7days/week) |
| 3 | NA |
| 4 | 3 : less likely (less than 1 per day) |
| 5 | 1: a moderate amount of time (3-4 days per week) |
| 6 | 0: most of the time (5-7days per week) |
| 7 | 2: some times (3-4 days per week) |
Output desired---
mydata.dt=
| Candidate.index | Score |
|---|---|
| 1 | 0 |
| 2 | 0 |
| 3 | NA |
| 4 | 3 |
| 5 | 1 |
| 6 | 0 |
| 7 | 2 |
Code used:
The observations were in factors so converted them to character first--
mydata.dt [, Score := as.character(Score)]
Then tried to remove the description of scores using this code---
mydata.dt$Score <- as.integer(gsub('[a-zA-Z]', '', mydata.dt$Score))
this gives a Warning message: In eval(ei, envir) : NAs introduced by coercion. Output all the observations in Score column were converted to NA
This may be a repeated question, I tried a lot using the codes available in the previous thread but it doesn't seem t work well for me. Please help.
CodePudding user response：
Just use sub here:
mydata.dt$Score <- sub("^(\\d ):.*$", "\\1", mydata.dt$Score)