A proposed approach for evaluating expert wine judge performance using descriptive statistics

Abstract

Descriptive statistics were used to evaluate the performance of five expert wine judges using a 20‐point quality scorecard in a wine competition environment. For each judge, frequency distributions were used to assess scorecard usage, and deviation font‐means were calculated as an index of consistency with the group. Differences in curve shapes were described using means, skewness and kurtosis values. Four of the five judges showed unimodal frequency distributions that were slightly skewed to lower scores. This information was believed to be useful for identifying expert judges whose use of a 20‐point quality scorecard and perspective of wine quality are consistent with the group.