Sampling for Web Surveys

Abstract

This chapter provides an overview of sampling techniques that are relevant to web surveys. The first step in setting up a web survey is to define the target population. The chapter describes several sampling designs and discusses sampling frames. The best known and probably most often used sampling design is a simple random sample without replacement. The chapter provides an overview of four other sampling designs. These sampling designs are stratified sampling, sampling with unequal probabilities, cluster sampling, and two-stage sampling. The chapter also describes some examples of improved estimation procedures. It is assumed that simple random sampling without replacement is applied. If every element in the population has a known and positive probability of selection, it is always possible to define an unbiased estimator. This is the so-called Horvitz–Thompson estimator.

Authors 2

1. Silvia Biffignandi

   University of Bergamo

   Affiliation as printed

   University of Bergamo, Italy

2. Jelke G. Bethlehem Aachen Faculty of Social and Behavioral Sciences Institute of Political Science

   Leiden University

   Affiliation as printed

   Faculty of Social and Behavioral Sciences, Institute of Political Science, Leiden University, The Netherlands