We examine the degree of consensus in quality ratings of prominent U.S. wine publications. For the purposes of wine consumption and research, are ratings on the ubiquitous 100- point scale reliable measures of quality? The value of expert judgment has been called into question by a number of studies, especially in the context of wine competitions and tasting events. Using data on 853 wines, we find a moderately high level of consensus, measured by the correlation coefficient, between most pairs of publications, similar to the level found by Ashton (2013). Rank and intraclass correlations are similar. Consensus is not found to be related to the blinding policies (or lack thereof) of the critical publications.
© 2001-2024 Fundación Dialnet · Todos los derechos reservados