R stratified random sampling from a data frame

Stratified.R

stratified = function(df, group, size) {
  #  USE: * Specify your data frame and grouping variable (as column 
  #         number) as the first two arguments.
  #       * Decide on your sample size. For a sample proportional to the
  #         population, enter "size" as a decimal. For an equal number 
  #         of samples from each group, enter "size" as a whole number.
  #
  #  Example 1: Sample 10% of each group from a data frame named "z",
  #             where the grouping variable is the fourth variable, use:
  # 
  #                 > stratified(z, 4, .1)
  #
  #  Example 2: Sample 5 observations from each group from a data frame
  #             named "z"; grouping variable is the third variable:
  #
  #                 > stratified(z, 3, 5)
  #
  require(sampling)
  temp = df[order(df[group]),]
  if (size < 1) {
    size = ceiling(table(temp[group]) * size)
  } else if (size >= 1) {
    size = rep(size, times=length(table(temp[group])))
  }  
  strat = strata(temp, stratanames = names(temp[group]), 
                 size = size, method = "srswor")
  (dsample = getdata(temp, strat))
}

If you want to use this, you can copy and paste the function above, or you can use the following:

require(RCurl)
temp = getURL("https://raw.github.com/gist/984691/fb8e0483b093caa871444db162ed11210a1bac5b/Stratified.R")
source(textConnection(temp))

Better options are:

• https://gist.github.com/mrdwab/933ffeaa7a1d718bd10a
• https://gist.github.com/mrdwab/6424112