More choices improve behavioral risk measures but surveys remain more reliable
Abstract It is essential for researchers and practitioners advising decision-makers to understand and measure precisely individual risk preferences. Behavioural risk- preference elicitation tasks have several advantages over self-report measures, including reducing biases and providing objective assessments. However, their low test–retest reliability limits their predictive value and overall measurement quality. This study investigates whether increasing the number of decisions within behavioural risk-elicitation tasks can improve their reliability. We hypothesise that adding multi-items to behavioural tasks can reduce decision noise and improve test–retest reliability. To test this hypothesis, we compared an enhanced multi-item behavioural risk elicitation task with standard behavioural- and self-report risk measures. Our results show that, although increasing the number of decisions improves reliability relative to traditional behavioural risk-preferences elicitation tasks, well-established multi-item self-report psychological questionnaires still demonstrate superior test–retest consistency.
Authors
- Steve Heinke (ORCID: https://orcid.org/0000-0002-9000-5897)
- Jörg Rieskamp (ORCID: https://orcid.org/0000-0003-2632-8015)
- Sandra Andraszewicz (ORCID: https://orcid.org/0000-0001-6958-4962)
- Oliver Schürmann (ORCID: https://orcid.org/0000-0003-2471-6583)
Institutions
- University of Basel (CH)
- ETH Zurich (CH)
- Singapore-ETH Centre (SG)
Publication Details
- Journal
- Scientific Reports
- Published
- 2026-11-15
- DOI
- https://doi.org/10.1038/s41598-026-79718-1
- Primary Topic
- Decision-Making and Behavioral Economics
- Type
- article
- Field-Weighted Citation Impact
- 0.00