Methods for Achieving Equivalence of Samples in Cross-National Surveys: The European Social Survey Experience
Peter Lynn, Sabine Häder, Siegfried Gabler, Seppo Laaksonen
Most surveys carried out at national or subnational level involve a single sample design and sampling frame. Where multiple frames are used, they are typically used to access different subpopulations. For cross-national surveys, however, it is usually necessary to use a different design in each nation to sample from an analogous national population. Cross-national sampling frames are rare. In this article, we describe procedures used to obtain equivalence of sample designs in 22 nations in Round 1 of the European Social Survey (ESS). We evaluate the implementation of the procedures and we summarise lessons for the design of cross-national surveys. We focus particularly on novel aspects of the procedures. These include specification of national sample sizes in terms of “effective sample size” and provision of guidelines on how to predict design effect components and how to use the predictions to determine the necessary sample size. We also discuss procedures for interaction between the various parties involved: the ESS central co-ordinating team, the ESS sampling panel, national co-ordinators and field work organisations.
Design effects, effective sample size, intra-cluster correlation, sample design, sampling frames