Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
A body of empirical work documents that most people believe they are above average in a variety of skills and abilities. This paper argues that such evidence does not necessarily imply that people process information in an irrational way. I build a model in which people can learn about their abilities at a cost of foregone production. Individuals in this model keep testing their abilities until their self-assessments become favorable enough, at which point they stop. This way, a disproportionately large share of the population ends up with a high opinion about their abilities. Copyright Springer-Verlag Berlin/Heidelberg 2004