Stratified sampling
Stratified sampling with equal/unequal probabilities.
strata(data, stratanames=NULL, size, method=c("srswor","srswr","poisson",
"systematic"), pik,description=FALSE)data |
data frame or data matrix; its number of rows is N, the population size. |
stratanames |
vector of stratification variables. |
size |
vector of stratum sample sizes (in the order in which the strata are given in the input data set). |
method |
method to select units; the following methods are implemented: simple random sampling without replacement (srswor), simple random sampling with replacement (srswr), Poisson sampling (poisson), systematic sampling (systematic); if "method" is missing, the default method is "srswor". |
pik |
vector of inclusion probabilities or auxiliary information used to compute them; this argument is only used for unequal probability sampling (Poisson and systematic). If an auxiliary information is provided, the function uses the inclusionprobabilities function for computing these probabilities. |
description |
a message is printed if its value is TRUE; the message gives the number of selected units and the number of the units in the population. By default, the value is FALSE. |
The data should be sorted in ascending order by the columns given in the stratanames argument before applying the function. Use, for example, data[order(data$state,data$region),].
The function produces an object, which contains the following information:
ID_unit |
the identifier of the selected units. |
Stratum |
the unit stratum. |
Prob |
the unit inclusion probability. |
############
## Example 1
############
# Example from An and Watts (New SAS procedures for Analysis of Sample Survey Data)
# generates artificial data (a 235X3 matrix with 3 columns: state, region, income).
# the variable "state" has 2 categories ('nc' and 'sc').
# the variable "region" has 3 categories (1, 2 and 3).
# the sampling frame is stratified by region within state.
# the income variable is randomly generated
data=rbind(matrix(rep("nc",165),165,1,byrow=TRUE),matrix(rep("sc",70),70,1,byrow=TRUE))
data=cbind.data.frame(data,c(rep(1,100), rep(2,50), rep(3,15), rep(1,30),rep(2,40)),
1000*runif(235))
names(data)=c("state","region","income")
# computes the population stratum sizes
table(data$region,data$state)
# not run
# nc sc
# 1 100 30
# 2 50 40
# 3 15 0
# there are 5 cells with non-zero values
# one draws 5 samples (1 sample in each stratum)
# the sample stratum sizes are 10,5,10,4,6, respectively
# the method is 'srswor' (equal probability, without replacement)
s=strata(data,c("region","state"),size=c(10,5,10,4,6), method="srswor")
# extracts the observed data
getdata(data,s)
# see the result using a contigency table
table(s$region,s$state)
############
## Example 2
############
# The same data as in Example 1
# the method is 'systematic' (unequal probability, without replacement)
# the selection probabilities are computed using the variable 'income'
s=strata(data,c("region","state"),size=c(10,5,10,4,6), method="systematic",pik=data$income)
# extracts the observed data
getdata(data,s)
# see the result using a contigency table
table(s$region,s$state)
############
## Example 3
############
# Uses the 'swissmunicipalities' data as population for drawing a sample of units
data(swissmunicipalities)
# the variable 'REG' has 7 categories in the population
# it is used as stratification variable
# Computes the population stratum sizes
table(swissmunicipalities$REG)
# do not run
# 1 2 3 4 5 6 7
# 589 913 321 171 471 186 245
# sort the data to obtain the same order of the regions in the sample
data=swissmunicipalities
data=data[order(data$REG),]
# the sample stratum sizes are given by size=c(30,20,45,15,20,11,44)
# 30 units are drawn in the first stratum, 20 in the second one, etc.
# the method is simple random sampling without replacement
# (equal probability, without replacement)
st=strata(data,stratanames=c("REG"),size=c(30,20,45,15,20,11,44), method="srswor")
# extracts the observed data
getdata(data, st)
# see the result using a contingency table
table(st$REG)Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.