WINETASTER ON 04/02/12 WITH 8 JUDGES AND 8 WINES BASED ON RANKS, IDENT=N Copyright (c) 1995-2012 Richard E. Quandt, V. 1.65
FLIGHT 1: Number of Judges = 8 Number of Wines = 8 A Tasting of 2005 Burgundies
Identification of the Wine: The judges' overall ranking:
Wine A is Chambertin Clos de Bèze, Rousseau ........ 6th place Wine B is Bonnes Mares, Comte de Vogüé ........ 5th place Wine C is Clos de la Roche Vielle Vignes, Ponsot ........ 7th place Wine D is Clos de Tart ........ 8th place Wine E is Musigny, Vielles Vignes, Vogüé ........ 1st place Wine F is Chambertin, Rousseau ........ 2nd place Wine G is La Grande Rue, Monopole ........ 4th place Wine H is Nuits St Georges, Leroy ........ 3rd place
The Judges's Rankings
Judge Wine -> A B C D E F G H Frank 6. 7. 8. 5. 1. 2. 4. 3. Orley 5. 6. 3. 8. 4. 2. 1. 7. Mike 7. 4. 2. 8. 1. 5. 6. 3. Jerry 5. 4. 1. 8. 7. 2. 3. 6. Ed 7. 5. 8. 3. 2. 4. 6. 1. Bob 3. 7. 8. 6. 1. 4. 5. 2. Tom 4. 3. 8. 2. 7. 6. 5. 1. Dick 4. 3. 5. 8. 2. 1. 6. 7.
Table of Votes Against Wine -> A B C D E F G H
Group Ranking -> 6 5 7 8 1 2 4 3 Votes Against -> 41 39 43 48 25 26 36 30
( 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 = 0.1801
The probability that random chance could be responsible for this correlation is rather large, 0.1839. 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 Frank 0.7759 Bob 0.5238 Ed 0.4551 Dick 0.3952 Mike 0.2143 Orley 0.0838 Jerry -0.2156 Tom -0.4286
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 E is Musigny, Vielles Vignes, Vogüé 2. ........ 2nd place Wine F is Chambertin, Rousseau 3. ........ 3rd place Wine H is Nuits St Georges, Leroy 4. ........ 4th place Wine G is La Grande Rue, Monopole 5. ........ 5th place Wine B is Bonnes Mares, Comte de Vogüé 6. ........ 6th place Wine A is Chambertin Clos de Bèze, Rousseau 7. ........ 7th place Wine C is Clos de la Roche Vielle Vignes, Po --------------------------------------------------- 8. ........ 8th place Wine D is Clos de Tart We now test whether the ranksums AS A WHOLE provide a significant ordering. The Friedman Chi-square value is 10.0833. The probability that this could happen by chance is 0.1839 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 Frank Orley Mike Frank 1.000 0.167 0.190 Orley 0.167 1.000 0.190 Mike 0.190 0.190 1.000 Jerry -0.357 0.738 0.190 Ed 0.738 -0.476 0.190 Bob 0.810 -0.048 0.190 Tom -0.024 -0.762 -0.476 Dick 0.286 0.476 0.381 Jerry Ed Bob Frank -0.357 0.738 0.810 Orley 0.738 -0.476 -0.048 Mike 0.190 0.190 0.190 Jerry 1.000 -0.690 -0.500 Ed -0.690 1.000 0.619 Bob -0.500 0.619 1.000 Tom -0.571 0.476 0.119 Dick 0.357 -0.095 0.214 Tom Dick Frank -0.024 0.286 Orley -0.762 0.476 Mike -0.476 0.381 Jerry -0.571 0.357 Ed 0.476 -0.095 Bob 0.119 0.214 Tom 1.000 -0.571 Dick -0.571 1.000 Pairwise correlations in descending order 0.810 Frank and Bob Significantly positive 0.738 Frank and Ed Significantly positive 0.738 Orley and Jerry Significantly positive 0.619 Ed and Bob Not significant 0.476 Ed and Tom Not significant 0.476 Orley and Dick Not significant 0.381 Mike and Dick Not significant 0.357 Jerry and Dick Not significant 0.286 Frank and Dick Not significant 0.214 Bob and Dick Not significant 0.190 Mike and Ed Not significant 0.190 Orley and Mike Not significant 0.190 Frank and Mike Not significant 0.190 Mike and Jerry Not significant 0.190 Mike and Bob Not significant 0.167 Frank and Orley Not significant 0.119 Bob and Tom Not significant -0.024 Frank and Tom Not significant -0.048 Orley and Bob Not significant -0.095 Ed and Dick Not significant -0.357 Frank and Jerry Not significant -0.476 Orley and Ed Not significant -0.476 Mike and Tom Not significant -0.500 Jerry and Bob Not significant -0.571 Tom and Dick Not significant -0.571 Jerry and Tom Not significant -0.690 Jerry and Ed Significantly negative -0.762 Orley and Tom Significantly negative
COMMENT: Today we were privileged to taste a fantastic array of top drawer 2005 Burgundies. With the only exception of the Clos de Tart, which a number of us felt was a slightly flawed bottle, all the wines lived up to expectations Basically, the wines were very similar in many ways, and we will be tasting these every other year for the next 10 years. Not only will we see how they change, but we will be able to test for intertemporal con- sistency, provided that we are still all able to participate. Some of these wines could handle a few more years in the bottle, in that they had noticeable tannins. It is also interesting to observe that the Clos de la Roche and the Musigny provoked some extreme opinions, in that the former garnered four votes for being the worst and two votes of being either first or second, while the latter garnered five votes of being either first or second and two votes of second to bottom.
Return to previous page