ESRA 2027 sessions by theme

Back to Overview of Sessions

Session Details

Fresh probability samples have long been regarded as the gold standard of survey data collection. However, declining response rates, rising fieldwork costs, and the growing demand for timely and cost-efficient data are accelerating the transition towards probability-based online panels. As more surveys adopt this approach, a key methodological question is no longer whether this transition is possible, but what, if anything, is gained or lost in the process.
To address this question, the session brings together researchers from official statistics, survey infrastructures, academia, and commercial survey organisations to share empirical and methodological evidence from different survey programmes and methodological perspectives. By synthesising findings across different national contexts, the session seeks to identify where fresh probability samples continue to provide unique methodological value, where probability-based online panels offer comparable evidence, and which methodological innovations can further narrow the remaining differences between these survey designs.
We particularly welcome studies that compare fresh probability samples and probability-based online panels, as well as studies that inform such comparisons by investigating one or more of the following dimensions:
• Population coverage, representativeness, and recruitment, including strategies for reaching offline and hard-to-reach populations;
• Attitudinal estimates and substantive findings, including behavioural, attitudinal, and topic-specific outcomes;
• Data quality, respondent behaviour, and engagement, including satisficing, speeding, item nonresponse, paradata, open-ended responses, and other components of Total Survey Error;
• Methodological innovations, including mixed-mode and adaptive survey designs, as well as AI- or LLM-based tools, that improve survey quality, representativeness, inclusiveness, or respondent engagement and thereby help reduce differences between survey designs, for example through conversational interviewing, adaptive probing, automated coding, automated quality monitoring, or supporting the participation of hard-to-reach populations.