WINETASTER ON 10/06/03 WITH 8 JUDGES AND 8 WINES BASED ON RANKS, IDENT=N Copyright (c) 1995-2003 Richard E. Quandt, V. 1.65
FLIGHT 1: Number of Judges = 8 Number of Wines = 8
Identification of the Wine: The judges' overall ranking:
Wine A is La Lagune 1982 ........ 3rd place Wine B is Ducru Beaucaillou 1982 tied for 4th place Wine C is Leoville Barton 1982 ........ 7th place Wine D is Gruaud Larose 1982 ........ 8th place Wine E is Gruaud Larose 1983 tied for 4th place Wine F is Cos d'Estournel 1982 ........ 6th place Wine G is Petit Village 1982 ........ 2nd place Wine H is Pichon Longueville 1982 ........ 1st place
The Judges's Rankings
Judge Wine -> A B C D E F G H Bob 2. 4. 5. 6. 1. 7. 8. 3. Orley 5. 4. 3. 8. 7. 6. 2. 1. Burt 2. 1. 4. 7. 6. 3. 8. 5. Frank 6. 8. 7. 4. 3. 5. 2. 1. Grant 1. 2. 5. 6. 8. 7. 3. 4. Ed 5. 4. 7. 8. 1. 3. 2. 6. Dick 4. 3. 5. 8. 7. 6. 2. 1. John 7. 8. 6. 5. 1. 2. 3. 4.
Table of Votes Against Wine -> A B C D E F G H
Group Ranking -> 3 4 7 8 4 6 2 1 Votes Against -> 32 34 42 52 34 39 30 25
( 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 = 0.1793
The probability that random chance could be responsible for this correlation is rather large, 0.1862. 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 Dick 0.5749 Orley 0.4311 Frank 0.1916 Ed 0.1905 Grant 0.0838 Bob -0.0120 Burt -0.1905 John -0.3095
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 H is Pichon Longueville 1982 2. ........ 2nd place Wine G is Petit Village 1982 3. ........ 3rd place Wine A is La Lagune 1982 4. tied for 4th place Wine B is Ducru Beaucaillou 1982 5. tied for 4th place Wine E is Gruaud Larose 1983 6. ........ 6th place Wine F is Cos d'Estournel 1982 7. ........ 7th place Wine C is Leoville Barton 1982 --------------------------------------------------- 8. ........ 8th place Wine D is Gruaud Larose 1982 We now test whether the ranksums AS A WHOLE provide a significant ordering. The Friedman Chi-square value is 10.0417. The probability that this could happen by chance is 0.1862 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 Bob Orley Burt Bob 1.000 -0.119 0.333 Orley -0.119 1.000 0.024 Burt 0.333 0.024 1.000 Frank -0.048 0.214 -0.762 Grant 0.048 0.524 0.405 Ed 0.071 -0.024 -0.071 Dick -0.024 0.929 0.143 John -0.119 -0.190 -0.595 Frank Grant Ed Bob -0.048 0.048 0.071 Orley 0.214 0.524 -0.024 Burt -0.762 0.405 -0.071 Frank 1.000 -0.286 0.214 Grant -0.286 1.000 -0.167 Ed 0.214 -0.167 1.000 Dick 0.214 0.690 0.095 John 0.690 -0.762 0.571 Dick John Bob -0.024 -0.119 Orley 0.929 -0.190 Burt 0.143 -0.595 Frank 0.214 0.690 Grant 0.690 -0.762 Ed 0.095 0.571 Dick 1.000 -0.262 John -0.262 1.000 Pairwise correlations in descending order 0.929 Orley and Dick Significantly positive 0.690 Frank and John Significantly positive 0.690 Grant and Dick Significantly positive 0.571 Ed and John Not significant 0.524 Orley and Grant Not significant 0.405 Burt and Grant Not significant 0.333 Bob and Burt Not significant 0.214 Frank and Dick Not significant 0.214 Orley and Frank Not significant 0.214 Frank and Ed Not significant 0.143 Burt and Dick Not significant 0.095 Ed and Dick Not significant 0.071 Bob and Ed Not significant 0.048 Bob and Grant Not significant 0.024 Orley and Burt Not significant -0.024 Orley and Ed Not significant -0.024 Bob and Dick Not significant -0.048 Bob and Frank Not significant -0.071 Burt and Ed Not significant -0.119 Bob and John Not significant -0.119 Bob and Orley Not significant -0.167 Grant and Ed Not significant -0.190 Orley and John Not significant -0.262 Dick and John Not significant -0.286 Frank and Grant Not significant -0.595 Burt and John Not significant -0.762 Grant and John Significantly negative -0.762 Burt and Frank Significantly negative
COMMENT: Some people were surprised to find that the 1983 Gruaud Larose was in the middle of the pack. The other surprise for some was that the Petit Village performed so well. There was considerable disagreement about the Gruaud Larose 1982, although the score at the bottom of the wines speaks for itself. The 1982s are darker, fatter and more brooding than the 1983s which the group recently tasted. The Ducru Beacaillou was particularly distinctive and easily recognisable. It is worthy to note that Pichon Lalande which won the tasting also won our previous tasting of 1982s and is Parker's top rated 1982 wine that we tasted.
Return to previous page