1 / 18

Risk Attitudes of Children and Adults: Choices Over Small and Large Probability Gains and Losses

Risk Attitudes of Children and Adults: Choices Over Small and Large Probability Gains and Losses. WILLIAM T. HARBAUGH University of Oregon KATE KRAUSE University of New Mexico LISE VESTERLUND University of Pittsburgh. We look at decisions under risk in people ranging in age from 5 to 64.

quanda
Télécharger la présentation

Risk Attitudes of Children and Adults: Choices Over Small and Large Probability Gains and Losses

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Risk Attitudes of Children and Adults: Choices Over Small and Large Probability Gains and Losses WILLIAM T. HARBAUGH University of Oregon KATE KRAUSE University of New Mexico LISE VESTERLUND University of Pittsburgh

  2. We look at decisions under risk in people ranging in age from 5 to 64. • Our primary objective is descriptive: see how risk preferences change with age. • In particular, we ask how the tendency to behave in ways that are consistent w/ EUT changes. • There’s lots of evidence that college age subjects violate EUT in a variety of ways. • If these violations increase with age and experience, this suggests these violations are “real” and important, • they might result from the application of learned rules of thumb or habits. • If they decrease, it suggests that the violations are more in the nature of mistakes – which is not to say they are not important.

  3. We use the SEU model to organize our empirical results. Under SEU, people weight outcomes using subjective, rather than objective weights. This model can accommodate a wide range of deviations from EUT. One familiar example is Cumulative Prospect Theory: (Tversky and Kahneman, 1992) CPT Behavior: 1. Risk-seeking over small-probability gains 2. Risk-aversion over high-probability gains 3. Risk-seeking over high-probability losses 4. Risk-aversion over small-probability losses

  4. Subjective Probability Weighting for CPT behavior:

  5. Straightforward Protocol: • People were offered a choice between a gamble and a sure payoff equal to the gamble’s expected value. (Except for 4 checks for monotonicity.) • 14 different choice pairs, 7 over gains, 7 over losses. • Choices differed in the probability of a gain (or loss.) • Some variation in the specific gambles used.

  6. For gains, proportion choosing the gamble, by age.

  7. For losses, proportion choosing the gamble, by age.

  8. indicates the choice pattern consistent with a reflection point at 1/3 and a regressive weighting function; • + indicates the choice pattern least consistent with a reflection point at 1/3 and a regressive weighting function. • Bold indicates the modal pattern for each age group.

  9. Results: Children tend to underweight low-probability events and overweight high-probability events. Adults are closer to w(p) = p than are children, and more so over gains than over losses. Differences between choices over losses and choices over gains diminish with age. Within-subject analysis: CPT pattern is modal choice among adults; rare among younger subjects.

  10. Conclusion: We examined choices between a simple gamble and a certain outcome for children and adults and we found large age-related differences in choices. About 70% of the youngest children chose a fair gamble when the chance of the gain was 0.8, while only 43% of the oldest adults did. Over losses, about 75% of these children took a fair gamble when the chance of the loss was 0.1, compared to 53% of the adults. These differences are even more dramatic if we limit the sample to those who pass the rationality tests.

  11. Conclusions: Most experimental work on adults finds that they tend to overweight low-probability outcomes and underweight high-probability ones, a pattern of behavior that can be explained by a regressive probability weighting function. In children, we find the opposite of regressive weighting: on average they choose as if they underweight low-probability events and overweight high-probability ones. This result is statistically significant and it is large.

  12. Conclusions: In situations with a small probability of a large loss they will be more likely to take the risky than the safe option. While our data come from financial decisions, this pattern seems to fit the stylized facts of adolescent decision- making in such areas as risky sex, smoking, and impaired driving. If children respond differently than adults do to a risk, then policies that aim at informing children of the risks that they face are unlikely to achieve a response that adults will find satisfactory. The fact that our experiments do not find this same tendency to underweight low probability outcomes among adults suggests a justification for constraints on adolescent decisions in situations that present low probability risks.

More Related