Bolstad (version 0.1-8)

sscsample: Simple, Stratified and Cluster Sampling

Description

Samples from a fixed population using either simple random sampling, stratitified sampling or cluster sampling.

Usage

sscsample(size, n.samples, sample.type="simple", x = NULL, strata = NULL, cluster = NULL, cl.size = NULL, ret=FALSE)

Arguments

size

the desired size of the sample

n.samples

the number of repeat samples to take

sample.type

the sampling method. Can be one of "simple", "stratified", "cluser" or 1, 2, 3 where 1 corresponds to "simple", 2 to "stratified" and 3 to "cluster"

x

a vector of measurements for each unit in the population. By default x is not used, and the builtin data set sscsample.data is used

strata

a corresponding vector for each unit in the population indicating membership to a stratum

cluster

a corresponding vector for each unit in the population indicating membership to a cluster

cl.size

the number of clusters to sample when using cluster sampling

ret

if true then the samples, their means, and the numbers of observations from each stratum are returned as a list.

Value

• If ret is true, then a list will be returned with the following components:
• samples
  a matrix with the number of rows equal to size and the number of columns equal to n.samples. Each column corresponds to a sample drawn from the population
• s.strata
  a matrix showing how many units from each stratum were included in the sample
• means
  a vector containing the mean of each sample drawn

See Also

sscsample.data

Examples

## Draw 200 samples of size 20 using simple random sampling
sscsample(20,200)

## Draw 200 samples of size 20 using simple random sampling and store the
## results. Extract the means of all 200 samples, and the 50th sample 
res<-sscsample(20,200,ret=TRUE)
res$means
res$samples[,50]