On the coefficient of variation as a criterion for decision under risk

The coefficient of variation (CV) has been advocated as a measure of risk sensitivity in meta-analyses of data for risky choices by animals and humans. We examine properties of the CV measure. We demonstrate that the CV has unsatisfactory normative properties because it is inconsistent with preferen...

Full description

Saved in:
Bibliographic Details
Published inJournal of mathematical psychology Vol. 54; no. 4; pp. 387 - 394
Main Authors Cox, James C., Sadiraj, Vjollca
Format Journal Article
LanguageEnglish
Published Elsevier Inc 01.08.2010
Subjects
Decision theory
Experiments
Risk
Decision theory
Risk
Experiments
Online AccessGet full text
ISSN0022-2496
1096-0880
DOI10.1016/j.jmp.2010.01.002

Cover

More Information
Summary:The coefficient of variation (CV) has been advocated as a measure of risk sensitivity in meta-analyses of data for risky choices by animals and humans. We examine properties of the CV measure. We demonstrate that the CV has unsatisfactory normative properties because it is inconsistent with preference for more than for less of the reward medium. Whether or not the CV succeeds as a descriptive criterion depends on its consistency or inconsistency with data. We report two experiments designed to test the distinctive properties of the CV. The experiments involve human subjects motivated by salient monetary payoffs. Data from the experiments do not support the CV as a descriptive criterion for decision under risk. We also use data from an experiment designed to test for the incidence of patterns of small-stakes risk aversion used in concavity calibration. These data also do not support the CV as a good descriptive criterion.
ISSN:0022-2496
1096-0880
DOI:10.1016/j.jmp.2010.01.002