Re: [R] colMeans in a data.frame with numeric and character data

2008-03-27 Thread Gabor Grothendieck
summaryBy in the doBy package can do that. The builtin iris data set has 4 numeric columns and one factor column: > library(doBy) > summaryBy(.~1, iris, fun = mean, keep = TRUE) Sepal.Length Sepal.Width Petal.Length Petal.Width 1 5.843.0573333.7581.199333 On Thu, Mar 2

Re: [R] colMeans in a data.frame with numeric and character data

2008-03-27 Thread Douglas Bates
On Thu, Mar 27, 2008 at 3:05 PM, Dimitris Rizopoulos <[EMAIL PROTECTED]> wrote: > try this: > > dat <- data.frame(x = rnorm(10), y = rexp(10), z = letters[1:10]) > colMeans(data.matrix(dat[sapply(dat, is.numeric)])) Alternatively > sapply(dat, mean) x y z -0.5260131

Re: [R] colMeans in a data.frame with numeric and character data

2008-03-27 Thread Dimitris Rizopoulos
try this: dat <- data.frame(x = rnorm(10), y = rexp(10), z = letters[1:10]) colMeans(data.matrix(dat[sapply(dat, is.numeric)])) I hope it helps. Best, Dimitris Dimitris Rizopoulos Biostatistical Centre School of Public Health Catholic University of Leuven Address: Kapucijnenvoer 35, Leuv