Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
Recent research distinguishes an individual's decision utility, inferred from her observed choices, from her experienced utility, which more closely matches the notion of happiness. Using various estimation techniques, we test whether post-choice satisfaction (experienced utility), like decision utility in an experimental data set, is S-shaped with loss aversion around a given reference point. We also present a model which estimates the satisfaction function and reference point simultaneously. When pooling the data across individuals, we find an S-shaped satisfaction function in which the reference point depends on past payments, social comparisons, and subjective expectations. There is mixed evidence of loss aversion. At the individual level, there is substantial variation in satisfaction function shapes, although the S-shape is common. Though the two notions of utility are distinct, our findings imply that the two are related at a fundamental level.