• search hit 4 of 12
Back to Result List

At the Risk of Loss: Experimental Evidence on the Discriminant Validity of Self-Reported Measures of Risk Preferences

  • In recent years, there has been increasing use of experimentally validated self-reported items to measure individuals' risk preferences, specifically risk aversion. While previous research supports their convergent validity, we argue that self-reported risk preference measures capture a broad spectrum of additional constructs involved in risky decision-making, notably including loss aversion as a potential confound. In broader applications, such as observational studies, associations with other constructs enhance a measurement's ability to predict risk-taking behaviors across diverse natural environments, often arising from the interplay of different constructs better captured by comprehensive self-reported measures than by narrower, construct-specific incentivized measures. Conversely, in focused tests where each construct has unique behavioral implications, using broad self-reported measures as a replacement for construct-specific incentivized measurements can result in significant endogeneity issues. By analyzing three samples fromIn recent years, there has been increasing use of experimentally validated self-reported items to measure individuals' risk preferences, specifically risk aversion. While previous research supports their convergent validity, we argue that self-reported risk preference measures capture a broad spectrum of additional constructs involved in risky decision-making, notably including loss aversion as a potential confound. In broader applications, such as observational studies, associations with other constructs enhance a measurement's ability to predict risk-taking behaviors across diverse natural environments, often arising from the interplay of different constructs better captured by comprehensive self-reported measures than by narrower, construct-specific incentivized measures. Conversely, in focused tests where each construct has unique behavioral implications, using broad self-reported measures as a replacement for construct-specific incentivized measurements can result in significant endogeneity issues. By analyzing three samples from two independent studies, we demonstrate that even after accounting for incentivized risk preference measures that remove latent construct associations and focus on measurement-related confounds, self-reported risk preferences maintain a substantial association with incentivized loss aversion measures, indicating that the measurement captures a mixture of both risk and loss aversion. We also observe that the strength of associations depends on whether individuals interpret risk-taking as accepting higher variances or higher losses. Based on this finding, we propose a simple procedure that utilizes this individual heterogeneity in the confounding effect's strength to enhance the robustness of conclusions drawn from analyses of self-reported risk preference measuresshow moreshow less

Export metadata

Additional Services

Search Google Scholar
Metadaten
Author: Andrés Felipe Zambrano-Curcio, Diemo UrbigORCiD, Werner Bönte, Jana Schmutzler
URL:https://www.dropbox.com/scl/fi/4rzgu15c0g5dfpgkhq91f/2023_esa_european_meeting_11_09.pdf?rlkey=tns75pnujizp4z8iv4eapho8q&dl=0
URL:https://www.economicscience.org/page/conference/2023-european-esa-meeting-exeter-uk
DOI:https://doi.org/10.2139/ssrn.4371973
ISSN:1556-5068
Title of the source (English):SSRN eLibrary
Document Type:Conference publication peer-reviewed
Language:English
Year of publication:2023
First Page:1
Last Page:39
Comment:
2023 European ESA Meeting, Exeter, UK
Faculty/Chair:Fakultät 5 Wirtschaft, Recht und Gesellschaft / FG ABWL, insbesondere empirische Unternehmensforschung und Transformation
Einverstanden ✔
Diese Webseite verwendet technisch erforderliche Session-Cookies. Durch die weitere Nutzung der Webseite stimmen Sie diesem zu. Unsere Datenschutzerklärung finden Sie hier.