Understanding Participation in a Long-Term Household Panel Study: Evidence from the UK

Longitudinal surveys provide valuable data to analysts and policymakers as societal topics can be studied over time. Consistent survey participation is required and encouraged. However, nonresponse or panel attrition (cumulative and permanent nonresponse) can prevent this and therefore is a concern to survey methodologists. This is particularly problematic when nonrespondents are systematically different from respondents, thus biasing survey estimates, potentially resulting in inaccurate inferences about the population. Understanding and improving survey participation has been, is and will continue to be a challenge survey methodologists face due to the dynamic nature of panel surveys (births, deaths, and migration) and how the survey landscape changes over time (e.g., mode shifts and technological advancements) so this thesis aims to contribute to the continued and changing understanding. This thesis uses data from Understanding Society: The UK Household Longitudinal Study (UKHLS), a household panel survey that follows sample members and aims to represent the UK population. Each substantive chapter uses a different component of the Study to examine aspects of survey participation. The thesis starts by introducing the relevant background of survey participation and nonresponse in longitudinal surveys. Chapter 1 uses data from the British Household Panel Survey sample (UKHLS’ predecessor study that was incorporated into the ongoing survey) to identify loyal sample members and examine response patterns over a thirty-year period. Chapter 2 considers the impact of the COVID-19 pandemic on survey dataset quality, in terms of likely nonresponse biases, by comparing survey data quality in the pre-pandemic mixed-mode UKHLS main survey with the primarily web-based UKHLS COVID-19 Study. Chapter 3 examines youth interview response behaviour to gauge whether it can predict early adult response behaviour in the UKHLS. The thesis concludes by discussing how the findings contribute to the literature on survey participation and inform survey practices in the future.