Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
Research on the effects of publicly reported hospital quality report cards on patient market shares is mixed. Higher-ranking hospitals do not consistently experience increases in market share. We argue that this may be because the report cards do not always convey "news" about quality; in some cases the rankings conform with prior beliefs about quality. We develop a structural model of the "news" in report cards and estimate the model using data from New York State in 1989-1991. We show hospitals with negative news in the original 1990 report cards experienced a decrease in market share, but that a misspecified model might continue to find no report card effect.