Advances in Business Statistics, Methods and Data Collection

27Introduction to Sampling and Estimation for Business Surveys

Paul A. Smith¹ and Wesley Yung²

¹S3RI & Department of Social Statistics & Demography, University of Southampton, Southampton, United Kingdom

²Statistics Canada, Ottawa, Canada

27.1 Introduction

A lot has happened in the world of establishment statistics in the more than 25 years since the first International Conference on Establishment Surveys (ICES‐I) and the seminal Business Survey Methods (BSM) book (Cox et al. 1995). Three out of the six sections of BSM covered topics of registers, sampling, and estimation, so we cannot hope to match that coverage in a single chapter. Fortunately, other chapters in this book cover some of the specific topics in detail. Here we aim to reflect the main methodologies for business surveys from the construction and use of business registers through designing and selecting samples to estimation to produce statistical outputs. We focus on the areas where there have been developments since BSM, providing enough of the general framework to set these in context.

One of those developments has been a way to describe survey processes, in the Generic Statistical Business Process Model (GSBPM; UNECE 2019). Figure 27.1 shows the top two levels of the GSBPM structure; in this chapter we focus on 2.4 (Design frame and sample), 4.1 (Create frame and select sample), 5.6 (Calculate weights), and 5.7 (Calculate aggregates). These areas contain the “traditional” elements of surveys where the use of ...