This website is the companion of the book "Applied Survey Methods - A Statistical perspective" by Jelke Bethlehem. This is a book about the theoretical and practical aspects of surveys.
The website is a quick reference guide to some of the most important aspects of surveys. It explains what surveys are and provides a short overview of the theory of simple random sampling. Using simulations it is shown that survey based estimates are reliable if probability sampling is used.
It is explained why the term "representative sample" should not be used. It also describes the potential dangers of conducting web surveys that suffer from under-coverage and self-selection
Non-response is presented one of the most serious problems for sample survey. Non-response may lead to estimates that are seriously biased. This is shown by means of a simulation experiment. Non-response problems may be reduced by carrying out some form of adjustment weighting.A simple example of weighting is given.
The website contains a number of tools. They may assist you in doing some of the computations, There are tools for generating random numbers, selecting a random sample, computing the sample size and computing estimtes and conficence intervals.
There are two software packages that can be downloaden. One is the program SimSam for simulating random samples from finite populations. The other is a demo-versions of the Blaise Sustem for computer assisted interviewing.
|