WINETASTER ON 11/9/15 WITH 8 JUDGES AND 9 WINES BASED ON RANKS, IDENT=N Copyright (c) 1995-2015 Richard E. Quandt, V. 1.65
FLIGHT 1: Number of Judges = 8 Number of Wines = 9
Identification of the Wine: The judges' overall ranking:
Wine A is Ch. Léoville Poyferré 1996 ........ 7th place Wine B is Ch. Léoville las Cases 2000 tied for 8th place Wine C is Ch. Léoville Barton 1996 tied for 3rd place Wine D is Ch. Léoville las Cases 1995 ........ 5th place Wine E is Ch. Léoville Poyferré 1997 ........ 1st place Wine F is Ch. Léoville Barton 1995 ........ 2nd place Wine G is Ch. Léoville Poyferré 1995 ........ 6th place Wine H is Ch. Léoville las Cases 1997 tied for 8th place Wine I is Ch. Léoville las Cases 1996 tied for 3rd place
The Judges's Rankings
Judge Wine -> A B C D E F G H I Alexa 5. 2. 6. 9. 7. 1. 4. 8. 3. Ed 7. 3. 5. 1. 2. 6. 8. 9. 4. Burt 7. 8. 5. 6. 1. 3. 2. 9. 4. Orley 9. 8. 7. 6. 3. 1. 2. 5. 4. Mike 9. 1. 3. 2. 5. 6. 7. 4. 8. Zaki 1. 8. 3. 2. 9. 4. 5. 6. 7. Bob 3. 9. 1. 7. 2. 8. 4. 5. 6. Dick 4. 9. 7. 5. 3. 6. 8. 2. 1.
Table of Votes Against Wine -> A B C D E F G H I
Group Ranking -> 7 8 3 5 1 2 6 8 3 Votes Against -> 45 48 37 38 32 35 40 48 37
( 8 is the best possible, 72 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 = 0.0688
The probability that random chance could be responsible for this correlation is rather large, 0.8194. 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 Burt 0.6895 Orley 0.1333 Ed 0.0753 Bob -0.1345 Dick -0.2427 Alexa -0.4118 Zaki -0.4667 Mike -0.4958
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 Ch. Léoville Poyferré 1997 2. ........ 2nd place Wine F is Ch. Léoville Barton 1995 3. tied for 3rd place Wine C is Ch. Léoville Barton 1996 4. tied for 3rd place Wine I is Ch. Léoville las Cases 1996 5. ........ 5th place Wine D is Ch. Léoville las Cases 1995 6. ........ 6th place Wine G is Ch. Léoville Poyferré 1995 7. ........ 7th place Wine A is Ch. Léoville Poyferré 1996 8. tied for 8th place Wine B is Ch. Léoville las Cases 2000 9. tied for 8th place Wine H is Ch. Léoville las Cases 1997 We now test whether the ranksums AS A WHOLE provide a significant ordering. The Friedman Chi-square value is 4.4000. The probability that this could happen by chance is 0.8194 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 0.70 for significance at the 0.05 level and must exceed 0.60 for significance at the 0.1 level Alexa Ed Burt Alexa 1.000 -0.150 0.200 Ed -0.150 1.000 0.200 Burt 0.200 0.200 1.000 Orley 0.233 -0.133 0.733 Mike -0.283 0.467 -0.367 Zaki -0.200 -0.217 -0.233 Bob -0.450 -0.200 0.300 Dick -0.367 -0.033 -0.017 Orley Mike Zaki Alexa 0.233 -0.283 -0.200 Ed -0.133 0.467 -0.217 Burt 0.733 -0.367 -0.233 Orley 1.000 -0.267 -0.333 Mike -0.267 1.000 -0.183 Zaki -0.333 -0.183 1.000 Bob -0.100 -0.300 0.150 Dick 0.117 -0.433 -0.133 Bob Dick Alexa -0.450 -0.367 Ed -0.200 -0.033 Burt 0.300 -0.017 Orley -0.100 0.117 Mike -0.300 -0.433 Zaki 0.150 -0.133 Bob 1.000 0.200 Dick 0.200 1.000 Pairwise correlations in descending order 0.733 Burt and Orley Significantly positive 0.467 Ed and Mike Not significant 0.300 Burt and Bob Not significant 0.233 Alexa and Orley Not significant 0.200 Alexa and Burt Not significant 0.200 Ed and Burt Not significant 0.200 Bob and Dick Not significant 0.150 Zaki and Bob Not significant 0.117 Orley and Dick Not significant -0.017 Burt and Dick Not significant -0.033 Ed and Dick Not significant -0.100 Orley and Bob Not significant -0.133 Ed and Orley Not significant -0.133 Zaki and Dick Not significant -0.150 Alexa and Ed Not significant -0.183 Mike and Zaki Not significant -0.200 Alexa and Zaki Not significant -0.200 Ed and Bob Not significant -0.217 Ed and Zaki Not significant -0.233 Burt and Zaki Not significant -0.267 Orley and Mike Not significant -0.283 Alexa and Mike Not significant -0.300 Mike and Bob Not significant -0.333 Orley and Zaki Not significant -0.367 Burt and Mike Not significant -0.367 Alexa and Dick Not significant -0.433 Mike and Dick Not significant -0.450 Alexa and Bob Not significant
COMMENT: All the wines drank well and were pretty obviously ready to drink; no flawed bottles were found. The rank averages were 42.75 for the Léoville las Cases, 39 for the Léoville Poyferré and 36 for the Léoville Barton; if anything in reverse order of their prices. But comparing the pair with the biggest difference, the Las Cases and the Barton, shows no significant difference, and so we can say that these wines are to all intents and purposes identical in quality in the tasters' estimation. Essentially the same conclusion is reached if one compares the vintages '95, '96 and '97, with the 1997 vintage marginally ahead of the other two. On the whole, the wines were perceived to be very similar, which is underscored by the fact that the correlation in the group was essentially zero.
Return to previous page