WINETASTER ON 04/01/02 WITH 8 JUDGES AND 8 WINES BASED ON RANKS, IDENT=N Copyright (c) 1995-2002 Richard E. Quandt
FLIGHT 1: Number of Judges = 8 Number of Wines = 8
Vertical tasting of Chateauneuf du Pape from the Chateau de Beaucastel
Identification of the Wine: The judges' overall ranking:
Wine A is 1985 ........ 7th place Wine B is 1989 tied for 2nd place Wine C is 1990 tied for 5th place Wine D is 1972 tied for 5th place Wine E is 1995 ........ 8th place Wine F is 1994 tied for 2nd place Wine G is 1991 ........ 1st place Wine H is 1986 ........ 4th place
The Judges's Rankings
Judge Wine -> A B C D E F G H Grant 2. 3. 5. 4. 8. 7. 6. 1. Frank 7. 8. 5. 6. 4. 2. 1. 3. Thom 4. 7. 8. 5. 6. 1. 2. 3. Burt 5. 4. 8. 2. 3. 1. 7. 6. Lesley 6. 1. 4. 8. 7. 5. 2. 3. Stephen 5. 6. 1. 8. 2. 4. 3. 7. Bob 8. 1. 5. 2. 7. 4. 3. 6. Dick 6. 1. 2. 3. 8. 7. 5. 4.
Table of Votes Against Wine -> A B C D E F G H
Group Ranking -> 7 2 5 5 8 2 1 4 Votes Against -> 43 31 38 38 45 31 29 33
( 8 is the best possible, 64 is the worst)
Here is a measure of the correlation in the preferences of the judges which ranges between 1.0 (perfect correlation) and 0.0 (no correlation):
W = 0.0915
The probability that random chance could be responsible for this correlation is rather large, 0.6447. Most analysts would say that unless this probability is less than 0.1, the judges' preferences are not strongly related. We now analyze how each taster's preferences are correlated with the group preference. A correlation of 1.0 means that the taster's preferences are a perfect predictor of the group's preferences. A 0.0 means no correlation, while a -1.0 means that the taster has the reverse ranking of the group. This is measured by the correlation R.
Correlation Between the Ranks of Each Person With the Average Ranking of Others
Name of Person Correlation R Lesley 0.4148 Bob 0.3114 Frank 0.1667 Dick 0.0964 Thom 0.0491 Grant -0.1905 Stephen -0.4311 Burt -0.4458
The wines were preferred by the judges in the following order. When the preferences of the judges are strong enough to permit meaningful differentiation among the wines, they are separated by -------------------- and are judged to be significantly different.
1. ........ 1st place Wine G is 1991 2. tied for 2nd place Wine B is 1989 3. tied for 2nd place Wine F is 1994 4. ........ 4th place Wine H is 1986 5. tied for 5th place Wine C is 1990 6. tied for 5th place Wine D is 1972 7. ........ 7th place Wine A is 1985 8. ........ 8th place Wine E is 1995 We now test whether the ranksums AS A WHOLE provide a significant ordering. The Friedman Chi-square value is 5.1250. The probability that this could happen by chance is 0.6447 We now undertake a more detailed examination of the pair-wise rank correla- tions that exist between pairs of judges. First, we present a table in which you can find the correlation for any pair of judges, by finding one of the names in the left hand margin and the other name on top of a column. A second table arranges these correlations in descending order and marks which is significantly positive significantly negative, or not significant. This may allow you to find clusters of judges whose rankings were particularly similar or particularly dissimilar. Pairwise Rank Correlations Correlations must exceed in absolute value 0.74 for significance at the 0.05 level and must exceed 0.64 for significance at the 0.1 level Grant Frank Thom Grant 1.000 -0.476 -0.071 Frank -0.476 1.000 0.690 Thom -0.071 0.690 1.000 Burt -0.310 -0.095 0.262 Lesley 0.262 0.119 0.024 Stephen -0.667 0.333 -0.214 Bob -0.048 -0.095 -0.071 Dick 0.524 -0.500 -0.548 Burt Lesley Stephen Grant -0.310 0.262 -0.667 Frank -0.095 0.119 0.333 Thom 0.262 0.024 -0.214 Burt 1.000 -0.524 -0.381 Lesley -0.524 1.000 0.071 Stephen -0.381 0.071 1.000 Bob 0.190 0.381 -0.333 Dick -0.381 0.476 -0.310 Bob Dick Grant -0.048 0.524 Frank -0.095 -0.500 Thom -0.071 -0.548 Burt 0.190 -0.381 Lesley 0.381 0.476 Stephen -0.333 -0.310 Bob 1.000 0.619 Dick 0.619 1.000 Pairwise correlations in descending order 0.690 Frank and Thom Significantly positive 0.619 Bob and Dick Not significant 0.524 Grant and Dick Not significant 0.476 Lesley and Dick Not significant 0.381 Lesley and Bob Not significant 0.333 Frank and Stephen Not significant 0.262 Grant and Lesley Not significant 0.262 Thom and Burt Not significant 0.190 Burt and Bob Not significant 0.119 Frank and Lesley Not significant 0.071 Lesley and Stephen Not significant 0.024 Thom and Lesley Not significant -0.048 Grant and Bob Not significant -0.071 Thom and Bob Not significant -0.071 Grant and Thom Not significant -0.095 Frank and Burt Not significant -0.095 Frank and Bob Not significant -0.214 Thom and Stephen Not significant -0.310 Grant and Burt Not significant -0.310 Stephen and Dick Not significant -0.333 Stephen and Bob Not significant -0.381 Burt and Stephen Not significant -0.381 Burt and Dick Not significant -0.476 Grant and Frank Not significant -0.500 Frank and Dick Not significant -0.524 Burt and Lesley Not significant -0.548 Thom and Dick Not significant -0.667 Grant and Stephen Significantly negative
COMMENT:All the wines were really lovely and Parker's distinctions were greatly exaggerated. In particular, the 1991 actually had the lowest points against, while Parker rated it worst in the group. Even the youngest of the wines are drinking well and the 1972 was not over the hill either. These are all outstanding wines.
It is also interesting to note that that the rank correlation between the group's preferences and the preferences implied by the rating of wines in the Wine Advocate as of August 31, 1997 is -0.3091 (using Spearman's Rho). While this is not statisically significantly different from zero, it is worth noting that the group's preferences are not correlated with the Wine Advocate on this occasion.
Return to previous page