On Jun 19, 2013, at 1:23 PM, Marc Schwartz <marc_schwa...@me.com> wrote:
> > On Jun 19, 2013, at 8:24 AM, Brian Perron <beper...@umich.edu> wrote: > >> Greetings: >> >> I am looking for a way to avoid using the ifelse function for >> constructing a new variable. More specifically, assume I have a set >> of variables with scores ranging from 1 to 30. >> >> set.seed(12345) >> x <- c(1:30) >> x1 <- sample(x, 15, replace = TRUE) >> x2 <- sample(x, 15, replace = TRUE) >> x3 <- sample(x, 15, replace = TRUE) >> x4 <- sample(x, 15, replace = TRUE) >> >> I want to construct a dichotomous variable that tests whether any of >> the variables contains the value 1. >> >> newVar <-ifelse(x1 == 1 | x2 == 1 | x3 == 1 | x4 == 1, 1, 0) >> >> I want to avoid the ifelse function because I have a number of large >> variable lists that will require new variables to be created. I'm >> sure there is a simple way to do this, but I haven't had any luck with >> my search! >> >> Thanks in advance. >> >> Brian > > > > If each of the vectors will be of the same length, create a matrix that > contains each one as a column: > > set.seed(12345) > x <- c(1:30) > x1 <- sample(x, 15, replace = TRUE) > x2 <- sample(x, 15, replace = TRUE) > x3 <- sample(x, 15, replace = TRUE) > x4 <- sample(x, 15, replace = TRUE) > >> cbind(x1, x2, x3, x4) > x1 x2 x3 x4 > [1,] 22 14 24 10 > [2,] 27 12 1 2 > [3,] 23 13 6 2 > [4,] 27 6 21 2 > [5,] 14 29 12 19 > [6,] 5 14 11 29 > [7,] 10 10 27 25 > [8,] 16 29 28 10 > [9,] 22 22 19 7 > [10,] 30 20 5 22 > [11,] 2 12 24 15 > [12,] 5 21 13 22 > [13,] 23 17 28 3 > [14,] 1 7 24 14 > [15,] 12 15 8 8 > > > Then you can use: > >> rowSums(cbind(x1, x2, x3, x4) == 1) > [1] 0 1 0 0 0 0 0 0 0 0 0 0 0 1 0 > > > which gets you the same result as: > >> ifelse(x1 == 1 | x2 == 1 | x3 == 1 | x4 == 1, 1, 0) > [1] 0 1 0 0 0 0 0 0 0 0 0 0 0 1 0 Actually, correction. That works in the case where only one of the values in the row is a 1, which happened to fit the example data. If there is a chance that more than one value in a row may be a 1, it returns a value equal to the number of 1's found. Thus, a more generic approach, I believe, would be: > sign(rowSums(cbind(x1, x2, x3, x4) == 1)) [1] 0 1 0 0 0 0 0 0 0 0 0 0 0 1 0 Regards, Marc ______________________________________________ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
