Estimating risky behavior with multiple-item risk measures

Abstract

We compare seven established risk elicitation methods and investigate how robustly they explain eleven kinds of risky behavior with 760 individuals. Risk measures are positively correlated; however, their performance in explaining behavior is heterogeneous and, therefore, difficult to assess ex ante. Greater diversification across risk measures is conducive to closing this knowledge gap. What we find is that performance increases considerably if we combine single-item risk measures to form multiple-item risk measures. Results are improved the more single-item measures they contain, and also if these single-item risk measures use different elicitation methods. Interestingly, survey items perform just as well as incentivized experimental items in explaining risky behavior.

Risky health choices and the Balloon Economic Risk Protocol

“…Our work extends previous efforts to improve the ecological validity of economic risk tasks. Although some researchers doubt the efficacy of economic risk tasks to predict risky behaviors (Eckel, 2016;Friedman et al, 2014;Trautmann, 2016), there have been numerous attempts to improve the poor ecological validity of current risk measures (Dohmen et al, 2011;Menkhoff and Sakha, 2017;Schildberg-Hörisch, 2018). For instance, utilizing a risk elicitation task in the same decision domain (Dohmen et al, 2011) and context (Zhou and Hey, 2018) as the risky behavior of interest.…”

“…For instance, utilizing a risk elicitation task in the same decision domain (Dohmen et al, 2011) and context (Zhou and Hey, 2018) as the risky behavior of interest. Although, economics has focused on revealed preferences that are directly incentivized (Charness et al, 2013), a general question about risky behavior performs surprisingly well (Dohmen et al, 2011;Menkhoff and Sakha, 2017). Another strategy that was proposed recently, suggests combining multiple risk elicitation tasks to calculate a normalized composite risk score.…”

“…Another strategy that was proposed recently, suggests combining multiple risk elicitation tasks to calculate a normalized composite risk score. This method captures a common component of risk across multiple tasks in order to measure a generalizable aspect of risk that increases predictive performance and has the benefit of reducing measurement error (Menkhoff and Sakha, 2017;Pedroni et al, 2017;Schildberg-Hörisch, 2018).…”

“…Researchers have also expressed concerns about the stability of measured preferences across economic risk tasks (Menkhoff and Sakha, 2017;Pedroni et al, 2017;Zhou and Hey, 2018). Our study allowed us to investigate risk preferences from the BERP in relation to other economic risk tasks often used in experimental economics.…”

“…This has allowed researchers to characterize individuals' variable appetite for risk. Nonetheless, despite a wealth of experimental contributions, risk preferences stemming from a variety of elicitation tasks are, at best, weakly correlated (Pedroni et al, 2017;Menkhoff & Sakha, 2017;Eckel & Wilson, 2004) and subsequently the ecological validity of these measures is uncertain. This study primarily focuses on this latter aspect of ecological validity.…”