design: Sampling design variables

In R2BEAT: Multistage Sampling Allocation and Sample Selection

Sampling design variables

Description

Example data frame containing variables for describing the sampling design.

Usage

data(beat.example)

Format

The design data frame contains a row per each stratum with the following variables:

STRATUM

Identifier of the stratum (numeric)

STRAT_MOS

Measure of size of the stratum (numeric)

DELTA

The average size of Secondary Stage Units (SSU) in the strata. With respect to the sample on which we are interested in, it could be equal or greater than 1 (numeric). See details for a depth explanation.

MINIMUM

the minimum number of SSU to be selected in each PSU. It could be different in each stratum (numeric)

Details

Note: the names of the variables must be the ones indicated above.

The sample design can be defined through a measure of size of the stratum, the average size of each SSU (>=1) and the minimum number of SSU to be selected in each PSU. In particular, if SSU are not cluster DELTA=1 and the sample size determined will be given in term of SSU. Instead, when SSUs are, in turn, clusters (for instance, households composed by individuals), defining DELTA equal to the average size of SSUs, enables to derive a sample in term of individuals.

Furthermore, modifying the MINIMUM it is possible to tune the number of PSU in the sample (see the example in beat.1st). In fact, considering the same sample size, increasing the MINIMUM, less PSU will be involved in the sample, but worst estimates in term of expected coefficient of variations will be provided. On the contrary, decreasing the MINIMUM, more PSU will be involved in the sample and better estimates will be obtained. Instead, increasing the MINIMUM for obtaining the same expected errors, requests less PSU, but much more SSU. The contrary occurs decreasing the MINIMUM.

Examples

## Not run: 
# Load example data
data(beat.example)
design
str(design)

## End(Not run)

R2BEAT documentation built on May 31, 2023, 7:19 p.m.