Differential functioning of the direct loneliness question across 27 EU countries
Single direct questions about feeling lonely and multi-item indirect scales give different prevalence estimates, and the gap is usually attributed to stigma-driven under-reporting. This project tests whether the direct loneliness question used in EU monitoring (“How much of the time, during the past 4 weeks, have you been feeling lonely?”) is answered differently by respondents with the same latent loneliness, measured by the three-item UCLA Loneliness Scale. We use the public EU Loneliness Survey (European Commission Joint Research Centre, 2022): 25,646 respondents from all 27 member states, with confirmation in a probability-based sample of 4,029 respondents from four countries. Hypotheses come from a completed Polish study (N = 3,376), in which current depressive and anxiety symptoms increased, and being in a relationship decreased, endorsement of a direct item at equal latent loneliness, whereas gender and stigma of loneliness showed no effect. We expect the same pattern for current distress, partnership and gender, and differential functioning across countries. Using MIMIC models, we will estimate these effects and quantify how much country rankings and group differences in directly reported loneliness change once they are removed. Descriptive and observed-score analyses of the EU data were run before registration; latent-variable models were not.
Authors
- Błażej Misiak (ORCID: https://orcid.org/0000-0002-5392-6398)
- Kacper Żełabowski (ORCID: https://orcid.org/0009-0004-1151-2332)
- Wiktor Grzegorzewicz
Publication Details
- Journal
- Open Science Framework
- Published
- 2026-11-12
- DOI
- https://doi.org/10.17605/osf.io/ybjp3
- Primary Topic
- Psychometric Methodologies and Testing
- Type
- article
- Field-Weighted Citation Impact
- 0.00