Authors
Abstract
Understanding Society has a complex sample design, and selection due to nonresponse, both of which need to be taken into account when the data are analysed. This paper provides support for users on how to take the sample design into account, why this is important, and how to identify, or derive, a weight suitable for the analysis at hand. We provide answers to many frequently asked questions collected over the years since Understanding Society data was first released in 2011. They start with basic concepts, and progress to more complex situations and their solutions. It will be most useful to Understanding Society users but will also have a wider service as a successful example of supporting data users after data release.