Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
This paper studies the use of probabilistic expectations data to predict behavior in incomplete scenarios posed by the researcher. The information that respondents have when replying to questions posing incomplete scenarios is a subset of the information that they would have in actual choice settings. Hence such questions do not elicit pure statements of preference; they elicit preferences mixed with expectations of future events that may affect choice behavior. The analysis developed here assumes respondents recognize that their behavior may depend on information they do not have when expectations are elicited, and that they answer coherently and honestly given the information provided. The objective in imagining such ideal respondents is to place a logical upper bound on the predictive content of elicited choice expectations. Copyright 1999 by Kluwer Academic Publishers