Question

I have large dataset like the following:

myd <- data.frame (var1 = c("A", "A", "B1-1", "B1-1", "C", "C", "D1", "D1"),
                   var2 = c(1,    2,    2,       2,     3,  1,   1,    4))
 var1 var2
1    A    1
2    A    2
3 B1-1    2
4 B1-1    2
5    C    3
6    C    1
7   D1    1
8   D1    4

Please note that each level in var1 is repeated twice and has two values in var2.

For each levels of var1, I want to create a new dataset where var2 values are now in row wise (shorted) and combined into one column with following type of output:

   var1    var2.h1 var2.h2      h1.h2  
1    A      1         2           12
2   B1-1    2         2           22
3    C      1         3           13
4   D1      1         4           14

Answer 1

library(plyr)
ddply(myd, .(var1), function(x) {
     foo <- data.frame(rbind(x$var2), sprintf("%s%s", x$var2[1], x$var2[2]))
     names(foo) <- c("var2.h1", "var2.h2", "h1.h2")
     foo
    })

var1 var2.h1 var2.h2 h1.h2
1    A       1       2    12
2 B1-1       2       2    22
3    C       3       1    31
4   D1       1       4    14