Measuring Decision and Experienced Utility
Sunday, Jan. 8, 2017 1:00 PM – 3:00 PM
- Chair: Betsey Stevenson, University of Michigan
The Sad Truth About Happiness Scales
AbstractHappiness research typically assumes happiness can be cardinalized to be distributed normally in each group studied. We show that under this assumption, we can never rank groups by average happiness. The CDFs will (almost) always cross when estimated using large samples. Moreover, there is always a cardinalization in the log-normal family that reverses the result. We summarize an extensive online appendix: many surprising results in this literature can be reversed by assuming a moderately left-skewed lognormal distribution; we can reject the joint assumption of a normal distribution and a common reporting function for the disabled and non-disabled.
Do the Less Educated Report Higher Life Satisfaction?: Response Scales, Focal Values, and Mis-Measurement of Subjective Well-Being
AbstractAn enormous “happiness” literature in economics and psychology makes use of Likert-like scales in assessing survey respondents' cognitive evaluations of their lives, overall. These “life satisfaction” data are now most standardly assessed on an eleven-point (zero to ten) scale, in accordance with recommendations by the OECD, US National Academies, and others. Typically, reduced form models of the subjective, quantitative data treat the integer responses as continuous variables and often even as interpersonally comparable; using ordinal models with weaker assumptions makes little difference in practice. However, I have noticed that the distribution of responses exhibits certain enhancements at focal values, in particular at 0, 5, and 10 on the eleven-point scale, and at integers on scales which allow for non-integer responses. I investigate the nature of these enhancements, showing that they are a result of different degrees of numeracy among the respondents, as though those who find it harder to translate a latent well-being assessment into a more refined scale simplify the scale for themselves. I propose and estimate a model to account for this behavior. I quantify the bias it introduces, showing that, as a result, it is possible for average life satisfaction scores to decrease with increasing education.
Models of Affective Decision-Making: How Do Feelings Predict Choice?
AbstractIntuitively, how you feel about potential outcomes will determine your<br /><br /><br />
decisions. Indeed, an implicit assumption in one of the most<br /><br /><br />
influential theories in psychology, prospect theory, is that feelings<br /><br /><br />
govern choice. Surprisingly, however, very little is known about the<br /><br /><br />
rules by which feelings are transformed into decisions. Here, we<br /><br /><br />
specified a computational model that used feelings to predict<br /><br /><br />
choices. We found that this model predicted choice better than<br /><br /><br />
existing value-based models, showing a unique contribution of feelings<br /><br /><br />
to decisions, over and above value. Similar to the value function in<br /><br /><br />
prospect theory, our feeling function showed diminished sensitivity to<br /><br /><br />
outcomes as value increased. However, loss aversion in choice was<br /><br /><br />
explained by an asymmetry in how feelings about losses and gains were<br /><br /><br />
weighed when making a decision, not by an asymmetry in the feelings<br /><br /><br />
themselves. The results provide new insights into how feelings are<br /><br /><br />
utilized to reach a decision.
- D0 - General
- I3 - Welfare, Well-Being, and Poverty