Impact of Frequentist and Bayesian Methods on Survey Sampling Practice: A Selective Appraisal

TL;DR

This paper reviews the influence of Frequentist and Bayesian methods on survey sampling, highlighting their applications in large-scale public policy surveys and small area estimation, and discusses their respective advantages.

Contribution

It provides a selective appraisal of how Frequentist and Bayesian approaches are applied in survey sampling, emphasizing recent developments and practical implications.

Findings

Frequentist methods dominate large-scale survey practice.

Bayesian methods are increasingly used for small area estimation.

Bayesian approaches offer flexible modeling options for complex surveys.

Abstract

According to Hansen, Madow and Tepping [J. Amer. Statist. Assoc. 78 (1983) 776--793], "Probability sampling designs and randomization inference are widely accepted as the standard approach in sample surveys." In this article, reasons are advanced for the wide use of this design-based approach, particularly by federal agencies and other survey organizations conducting complex large scale surveys on topics related to public policy. Impact of Bayesian methods in survey sampling is also discussed in two different directions: nonparametric calibrated Bayesian inferences from large samples and hierarchical Bayes methods for small area estimation based on parametric models.