Sampling for Web Surveys

This chapter provides an overview of sampling issues that are relevant to web surveys. The first step in setting up a web survey is to define the target population. This is the population to be investigated, and to which the conclusions refer. In the survey design phase, the objectives of the survey must be translated into concrete operational procedures. This involves defining the target variables of the survey. These variables measure the various aspects of the phenomena to be investigated. To select a sample from a target population in a scientifically sound way, two ingredients are required: a sampling design based on some form of probability sampling and a sampling frame. The chapter discusses several sampling designs and sampling frames. Samples can be selected with replacement or without replacement. The best known and probably most often used sampling design is a simple random sample without replacement. Controlled Vocabulary Terms probability sampling; random sampling; sampling; Sampling frame; Web surveys