ESRA 2025 Preliminary Glance Program

All time references are in CEST

Survey research and the COVID-19 pandemic
Session Organiser	Dr Daniel Seddig (KFN)
Time	Wednesday 16 July, 13:30 - 15:00
Room	Ruppert Wit - 0.52

-

Papers

Exploring survey response rates on the CSEW – pre and post pandemic

Miss Catherine Grant (ONS) - Presenting Author

The Crime Survey for England and Wales (CSEW) has held one of the highest face-to-face survey response rates in the UK with pre-pandemic response rates of around 70-75%. Like many other surveys the response rate fell significantly when face-to-face interviewing resumed at the close of 2021.
Previous research suggests that lower response rates may have minimal impact on non-response bias and the reliability of survey estimates. We explore whether this holds true with a sudden and dramatic decrease in response rates by comparing estimates from pre-pandemic (with response rates achieved in the mid-seventies) to those produced post-pandemic (with response rates in the mid-forties).
This presentation explores some of the potential reasons behind the fall in response rate, the effectiveness of measures taken to improve response and the impact of lower response rates on the respondent profile and core survey estimates.

Improved self-rated health during the COVID-19-pandemic: Collateral benefit or methodological artefact?

Dr Niels Michalski (Robert Koch Institute) - Presenting Author
Mrs Carolin Heil (Robert Koch Institute)
Mrs Angelika Schaffrath Rosario (Robert Koch Institute)
Dr Claudia Hövener (Robert Koch Institute)

Survey-based measures of self-rated health (SRH) in Germany’s general population indicated improved health during the first wave of the COVID19 pandemic in 2020. Slowed-down working life, changes in leisure activity patterns due to the lockdown, and altered reference points arising from uncertainties surrounding the novel virus serve as potential explanations. However, contact restrictions also impacted data collection routines and sampling probabilities. The German Socio-Economic Panel study (SOEP) switched its regular interview mode from CAPI to telephone during this phase.
In our analysis, we used data from the telephone survey German Health Update (GEDA 2019/2020-EHIS) and the SOEP to disentangle substantive changes from methodological issues. Both surveys cover different time spans from February to June in 2019 and 2020 respectively and provide SRH measures based on a 5-point scale ranging from 1 (very bad) to 5 (very good). Mean differences in SRH between the pre-pandemic (before 16 March 2020) and pandemic phase were estimated and adjusted for sociodemographic controls using OLS regression. For both samples, differences between the pandemic phase of 2020 and the same time span in 2019 were estimated to consider seasonal effects. For the SOEP sample, the effect was estimated applying a DiD approach comparing 2020 with 2019.
Results confirm a significant rise in mean SRH from the pre-pandemic to the pandemic phase in both samples (GEDA b=.087; SOEP b=.098) during the first wave of the COVID19 pandemic in Germany. The DiD effect estimated for the SOEP was b=.137. Results suggest that the mode switch from CAPI to telephone in the SOEP sample had a negligible impact. Insights into the volatility of the SRH measure are important for health monitoring since SRH is a core item in the Minimum Health Module and the SF12 questionnaire.

The Challenges and Countermeasures in Data Collection of The Taiwan Social Change Survey (TSCS)

Professor Chyi-in Wu (Institute of Sociology, Academia Sinica) - Presenting Author

For more than 40 years, Taiwan Social Change Survey (TSCS) has always used face-to-face interviews as the main method to collect social survey data. Basically, it has been implemented smoothly and successfully. However, since the outbreak of the Covid-19 Pandemic in 2020, as is the case around the world, this method of collecting survey data in a face-to-face manner has been severely impeded. In the past, interviewers and respondents gathered together, the scene of harmonious question and answer sessions is almost gone forever. Even in the current post-Covid-19 era, face-to-face questionnaire interviews, as the most suitable way to collect survey data, have become "mission impossible." Innovative and practical social survey methods must be rigorously designed and developed under this dilemma.
This study will, first, depict the challenges encountered by TSCS when implementing traditional face-to-face interview methods and the short-lived coping strategies (countermeasures) currently adopted. Next, this article will discuss possible breakthroughs. Finally, this study will specifically provide actual experimental test results as empirical evidences for evaluating the development of innovative social survey methods. For example, based on the actual living conditions of residents in contemporary Taiwanese society, this study conceives a "community-based" social survey method to cultivate trust and connection with community residents through longer-term and frequent contact with ordinary residents. One way to do this is to task "community liaisons" (employed on a permanent basis by the research project) to develop links with specific communities (which can be regarded as strong ties of familiarity). Once a community is selected as a sampling area for TSCS, the community liaison person can lead the interviewer into the community to conduct substantial face-to-face interviews. In other words, this study will experiment with the method of "developing

Representativity and mode effects in the Swiss Election Study (Selects): Are paper surveys still relevant in the post-pandemic era?

Dr Jan-Erik Refle (University of Lausanne / University of Geneva)
Dr Annika Lindholm (FORS)
Mr Lukas Lauener (FORS / University of Lausanne) - Presenting Author

Web-based surveys have gained popularity due to their cost-effectiveness and changing participation habits, a trend accelerated by the COVID-19 pandemic. This raises questions about the value of paper-based surveys in mixed-mode studies like the Swiss Election Study (Selects), which uses a push-to-web design, offering paper questionnaires only after several reminders. Those opting for paper surveys are typically older and less politically engaged.
Using data from the 2019 and 2023 Selects post-electoral surveys, we examine mode selection effects by sociodemographic profile, comparing web and paper samples. Findings suggest that age-related biases have decreased in 2023, but that language effects persist. Thus, the value of paper surveys in correcting selection biases currently still persists but especially age-related effects will become less relevant over the next years.