Two-stage Sampling Design and Sample Selection with the R package R2BEAT

TL;DR

R2BEAT is an R package designed for optimal sample allocation in complex, multi-domain, and multi-purpose survey designs, extending classical methods to multivariate and multi-stage sampling scenarios.

Contribution

It introduces a comprehensive R package that implements extended allocation methods for complex sampling designs, facilitating all survey planning phases.

Findings

Supports multi-domain and multi-purpose survey designs

Provides tools for optimizing and evaluating sample allocations

Extends classical allocation methods to complex, multivariate, multi-stage sampling

Abstract

R2BEAT (R "to" Bethel Extended Allocation for Two-stage sampling) is an R package for the allocation of a sample. Besides other software and packages dealing with the allocation problems, its peculiarity lies in facing properly allocation problems for complex sampling designs with multi-domain and multi-purpose aims. This is common in many official and non-official statistical surveys, therefore R2BEAT could become an essential tool for planning a sample survey. The package implements the Tschprow (1923) - Neyman (1934) method for the optimal allocation of units in stratified sampling, extending it to the multivariate (accordingly to Bethel's proposal (1989)), multi-domain and to the complex sampling designs case (Falorsi et al., 1998). The functions implemented in R2BEAT allow the use of different workflows, depending on the available information on one or more interest variables. The package covers all the phases, from the optimization of the sample to the selection of the Primary and Secondary Stage Units. Furthermore, it provides several outputs for evaluating the allocation results.