ESRA 2027 sessions by theme

Back to Overview of Sessions

Session Details

Survey weighting is essential for drawing inferences from sample surveys about a target population. Appropriate weighting procedures account for unequal selection probabilities, unit nonresponse, and coverage errors, thereby reducing bias and improving the representativeness of survey estimates. Increasing survey complexity, declining response rates, the growing availability of auxiliary information from multiple sources, and the expansion of longitudinal studies have accelerated the development of innovative weighting methods and raised important questions regarding their statistical properties, practical implementation, and evaluation.
This session invites contributions addressing recent advances in survey weighting for both cross-sectional and longitudinal surveys. We welcome methodological, applied, and simulation-based research on all aspects of weighting, including but not limited to
- design weighting,
- nonresponse adjustment,
- calibration,
- model-assisted approaches,
- model-based approaches,
- machine learning methods,
- weighting in panel surveys,
- treatment of temporary nonresponse,
- variance estimation, and
- diagnostics for assessing weight quality.
We particularly encourage contributions in both theoretical developments and empirical applications using official statistics, social surveys, or other survey data.
The aim of this session is to provide a forum for discussing current methodological challenges, emerging developments, and practical recommendations in survey weighting. By bringing together researchers working on different aspects of weighting methodology, the session seeks to foster exchange between theory and practice and to identify promising directions for future research in survey methodology.