Disappointment-Elation Assessment and Resolution Preference

Description

Estimate how strongly a person weighs falling short of, or exceeding, an uncertain prospect's expectation, and use it to predict risk attitude and how outcomes should be revealed.

Bell's disappointment theory adds a psychological term to expected value: an outcome below the prospect's expectation causes disappointment proportional to the shortfall, and one above it causes elation proportional to the excess, with separate weights. If disappointment outweighs elation, a lottery's certainty equivalent falls below its expected value even when money is valued linearly, and risk aversion rises with the chance of winning. The model explains Allais-type choices as concern with how uncertainty resolves, separately from regret between options. It also predicts preferences between one-stage and two-stage resolution of the same lottery, which bears on how and when outcomes are revealed. Weights are estimated from certainty equivalents of simple win-or-nothing lotteries at several probabilities.

Attributes

Loss type
Unmigrated
Stage
Design, Discovery
Effort
Days

Steps

  1. Elicit certainty equivalents for win-or-nothing lotteries across several win probabilities (for example 10%, 30%, 50%, 70%, 90%).
  2. Fit the difference between disappointment and elation weights from the gap between certainty equivalents and expected values across probabilities.
  3. To separate components, ask for the lower prize that makes winning in one lottery as satisfying as winning in another with a different chance, and the parallel question for losing.
  4. Use two direct certainty equivalents to fix the trade-off between feelings and money.
  5. Check predictions: risk aversion rising with win probability and preferences between single-stage and staged reveal of the same outcome.

Sources

  • Decision Making: Descriptive, Normative, and Prescriptive Interactions (Bell et al., 1988)

Also known as

  • Disappointment theory assessment (Bell)
  • Disappointment-elation weighting
  • Staged versus single resolution preference

Categories

  • Analytics & Behaviour
  • Behavioral Science & Economics