alternative to ifelse

On Jun 19, 2013, at 8:24 AM, Brian Perron <beperron at 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

sign(rowSums(cbind(x1, x2, x3, x4) == 1))