capm (version 0.9.1)

CalculateStratifiedSampleSize: Stratified random sample size

Description

Calculates sample size for a stratified random sampling design to estimate a total.

Usage

CalculateStratifiedSampleSize(strata = NULL, x = NULL, conf.level = 0.95, error = 0.1)

Arguments

strata

vector, matrix or data.frame. If vector, the length must be equal to the number of sampling units in the population and each element represent the strata memebership of each sampling unit. If matrix or data.frame, first column represent the size of each strata, second column represent the expected mean in each strata and third column represent the expected variance in each strata. Each row is a stra and must be named.

x

data.frame representing a pilot sample. First column has the variable to be estimated and second column has the strata membership of each observation.

conf.level

the confidence level required. It must be numeric between 0 and 1 inclusive.

error

the maximum relative difference between the estimate and the unknown population value. It must be numeric between 0 and 1 inclusive.

Value

numeric sample size rounded up to nearest integer.

References

Levy P and Lemeshow S (2008). Sampling of populations: methods and applications, Fourth edition. John Wiley and Sons, Inc.

http://oswaldosantos.github.io/capm

Examples

# Using a pilot sample from a population with 10000 sampling units.
strata <- rep(c('Rural', 'Urban'), c(100, 9900))
pilot.sample <- data.frame(c(rpois(5, 1.3), rpois(45, 0.8)),
                           rep(c('Rural', 'Urban'), c(5, 45)))
CalculateStratifiedSampleSize(strata, pilot.sample)                            
      
# Using expected mean and variance for a population with
# 10000 sampling units.
str.n <- c(Rural = 100, Urban = 9900)
str.mean <- c(Rural = 1.4, Urban = 0.98)
str.var <- c(Rural = 1.48, Urban = 1.02)
CalculateStratifiedSampleSize(cbind(str.n, str.mean, str.var))