I wonder if the following is possible in R?

I want to add a new column to my Data called task whose values are to be determined by the following statements:

• IF order is "sc", task value for rows with odd-numbered time values (1,3,...) is "simple" otherwise "complex".

• IF order is "cs", task value for rows with odd-numbered time values (1,3,...) is "complex" otherwise "simple".

Data = read.table(text="
id order  time task
1  sc     1    x
1  sc     2    x
2  cs     1    x
2  cs     2    x
3  sc     1    x
3  sc     2    x
4  cs     1    x
4  cs     2    x
", h= TRUE)

CodePudding user response：

You can use ifelse and time %% 2 to determine odd or even.

Data |>
  transform(task = ifelse(order == 'sc' & time %% 2 == 1 |
                          order == 'cs' & time %% 2 == 0,
                          'simple', 'complex'))

  id order time    task
1  1    sc    1  simple
2  1    sc    2 complex
3  2    cs    1 complex
4  2    cs    2  simple
5  3    sc    1  simple
6  3    sc    2 complex
7  4    cs    1 complex
8  4    cs    2  simple

CodePudding user response：

The other answer by Darren Tsai is great! I'm not sure what the speed of ifelse() is like, I don't have much experience using it, but here's an alternate option:

All it does is get the row numbers (via which()) that match the conditions for the "simple" case and then again for "complex" case.

index.simple <- which( (Data$order=="sc" & Data$time%%2==1) | (Data$order=="cs" & Data$time%%2==0) )
Data$task[index.simple]  <- "simple" 

index.complex <- which( (Data$order=="sc" & Data$time%%2==0) | (Data$order=="cs" & Data$time%%2==1) )                     
Data$task[index.complex]  <- "complex"