Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
Economic models and experiments frequently use lotteries with only a few outcomes. We study the correlation of such lotteries and discuss its relevance for economic applications. In particular, we fully characterize the joint distribution of two binary lotteries via their first three univariate moments and their correlation coefficient. As we illustrate alongside several examples, the resulting parametrization may be useful for economic modeling and experimental design.