This paper, 1) outlines some of the challenges in obtaining participation from older sample members in a survey that is not specifically tailored to older people, 2) provides evidence of the relative response propensity of older people in such a survey and, 3) provides experimental evidence of...

This paper describes the extent and correlates of non-response at waves 1 and 2 of Understanding Society. We examine both household-level and individual-level non-response at wave 1. For wave 2, we examine attrition relative to wave 1 both in terms of enumerated persons and in terms of adults...

Previous evidence suggests that a wave with mixed mode data collection in an otherwise face-to-face panel survey will achieve a lower response rate than other waves. But until now there has been no evidence as to whether the response rate can be expected to recover subsequently. In other words,...

Researchers often assume that respondent burden influences survey participation propensity and that interview length is a good indicator of burden. However, there is little evidence of the effect of interview length on subsequent participation propensity, particularly for face-to-face surveys....

In order to lower costs, the idea has been raised to use a mixed-mode design for the UK Household Longitudinal Study (UKHLS) using Computer-Assisted Personal Interviewing (CAPI) and Web Self-Interviewing (WSI). However, a mixed-mode CAPI-WSI design may damage data-quality because of measurement...

This paper describes the problem of maintaining cross-sectional representativeness in a longitudinal survey of a changing population. The extent and nature of the problem is outlined and potential solutions are described. The procedures adopted on Understanding Society are described. The main...

This paper describes the design of the sample for “Understanding Societyâ€. The sample consists of five components. The largest component is a newly-selected general population sample. The other four components are an ethnic minority ‘boost’ sample, a general population...

This paper outlines the strategy for the development and provision of analysis weights for Understanding Society. The strategy is placed within the context of a number of practical and statistical issues in the context of a survey with such a complex design, involving multiple waves, samples,...

The overall sample design for Understanding Society has been described in an earlier working paper in this series (Lynn 2009). This paper describes the special measures taken to boost the sample of members of five key minority ethnic groups in Great Britain. A new method was developed to...

This article describes a randomised experiment with mixed-mode survey designs in the context of a household panel survey. The experiment was designed to allow comparisons between two alternative mixed-mode designs (telephone interviewing plus face-to-face interviewing) and a unimode design...