Use of principal component analysis for the evaluation of judge performance at wine competitions

• Autores: Margaret A. Cliff, Marjorie C. King
• Localización: Journal of wine research, ISSN 0957-1264, Vol. 10, Nº 1, 1999, págs. 25-32
• Idioma: inglés
• Texto completo no disponible (Saber más ...)
• Resumen

  • Principal component analysis (PCA) was used to evaluate the performance of five wine judges using a 20‐point quality scoring system. Data were collected for 97 wines as part of a wine competition. Principal component plots were used to (1) identify underlying mechanisms that judges used to differentiate wine quality, and (2) evaluate the relationships among the quality components. Judges’ ‘styles’ differed according to their use, or lack of use, of the individual quality components (appearance, aroma, flavour, acidity, finish, body, balance and overall character) in making a final quality assessment of the wines. In general, judges showed individual patterns of separating the wines according to the presence or absence of aroma and flavour and overall character. These component scores were highly correlated with each other and with the final wine quality score. A comparison of the PCA plots with eggshell plots, an established methodology, indicated that judges who used a systematic approach to quality evaluation were closer to the group consensus. This technique serves as a complement to existing methodologies to assist competition co‐ordinators in future judge selection.