WINETASTER ON 11/05/01 WITH 8 JUDGES AND 6 WINES BASED ON RANKS, IDENT=N Copyright (c) 1995-2001 Richard E. Quandt
FLIGHT 1: Number of Judges = 8 Number of Wines = 6
Identification of the Wine: The judges' overall ranking:
Wine A is Ch. Lynch Bages 1961 ........ 2nd place Wine B is Ch. de Marbuzet 1961 ........ 3rd place Wine C is Ch. Trotanoy 1961 ........ 1st place Wine D is Ch. Talbot 1961 ........ 4th place Wine E is Ch. Giscours 1961 ........ 6th place Wine F is Les Caves Maxim 1957 ........ 5th place
The Judges's Rankings
Judge Wine -> A B C D E F Frank 4. 6. 1. 5. 2. 3. Bob 4. 5. 1. 2. 6. 3. John 1. 2. 3. 6. 5. 4. Ed 4. 2. 1. 3. 6. 5. Grant 2. 1. 4. 3. 6. 5. Orley 2. 6. 1. 3. 5. 4. Dick 4. 1. 3. 6. 5. 2. Burt 5. 4. 1. 2. 3. 6.
Table of Votes Against Wine -> A B C D E F
Group Ranking -> 2 3 1 4 6 5 Votes Against -> 26 27 15 30 38 32
( 8 is the best possible, 48 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 = 0.2625
The probability that random chance could be responsible for this correlation is quite small, 0.0622. 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 Ed 0.7827 Grant 0.5508 Bob 0.5218 Orley 0.4286 John 0.1160 Burt 0.0857 Dick -0.0286 Frank -0.0857
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 C is Ch. Trotanoy 1961 --------------------------------------------------- 2. ........ 2nd place Wine A is Ch. Lynch Bages 1961 3. ........ 3rd place Wine B is Ch. de Marbuzet 1961 4. ........ 4th place Wine D is Ch. Talbot 1961 5. ........ 5th place Wine F is Les Caves Maxim 1957 --------------------------------------------------- 6. ........ 6th place Wine E is Ch. Giscours 1961 We now test whether the ranksums AS A WHOLE provide a significant ordering. The Friedman Chi-square value is 10.5000. The probability that this could happen by chance is 0.0622 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 0.89 for significance at the 0.05 level and must exceed 0.83 for significance at the 0.1 level Frank Bob John Frank 1.000 0.257 -0.143 Bob 0.257 1.000 -0.143 John -0.143 -0.143 1.000 Ed -0.143 0.600 0.314 Grant -0.771 0.029 0.600 Orley 0.486 0.771 0.143 Dick -0.143 -0.086 0.600 Burt 0.314 0.429 -0.371 Ed Grant Orley Frank -0.143 -0.771 0.486 Bob 0.600 0.029 0.771 John 0.314 0.600 0.143 Ed 1.000 0.600 0.371 Grant 0.600 1.000 -0.029 Orley 0.371 -0.029 1.000 Dick 0.314 0.314 -0.314 Burt 0.543 -0.086 0.371 Dick Burt Frank -0.143 0.314 Bob -0.086 0.429 John 0.600 -0.371 Ed 0.314 0.543 Grant 0.314 -0.086 Orley -0.314 0.371 Dick 1.000 -0.429 Burt -0.429 1.000 Pairwise correlations in descending order 0.771 Bob and Orley Not significant 0.600 Bob and Ed Not significant 0.600 John and Dick Not significant 0.600 John and Grant Not significant 0.600 Ed and Grant Not significant 0.543 Ed and Burt Not significant 0.486 Frank and Orley Not significant 0.429 Bob and Burt Not significant 0.371 Ed and Orley Not significant 0.371 Orley and Burt Not significant 0.314 Frank and Burt Not significant 0.314 Ed and Dick Not significant 0.314 Grant and Dick Not significant 0.314 John and Ed Not significant 0.257 Frank and Bob Not significant 0.143 John and Orley Not significant 0.029 Bob and Grant Not significant -0.029 Grant and Orley Not significant -0.086 Bob and Dick Not significant -0.086 Grant and Burt Not significant -0.143 Frank and Ed Not significant -0.143 Bob and John Not significant -0.143 Frank and John Not significant -0.143 Frank and Dick Not significant -0.314 Orley and Dick Not significant -0.371 John and Burt Not significant -0.429 Dick and Burt Not significant -0.771 Frank and Grant Not significant
COMMENT: Everyone agreed that no wine was oxidized, that the 1957 Restaurant Maxim Cuvee was a minor miracle, and that the 1961 Trotanoy was out- standing. On the whole the wines are drinking very nicely, but there is no reason to keep them any longer. The bad news is: we discarded five bottles that were badly flawed before starting the tasting. The discarded wines were Ch. Bouscaut '61, Margaux B&G '61, Ch. Rausan Segla '60, Ch. Bouscaut '64, Ch. Lafite '67.
Return to previous page