WINETASTER ON 04/05/10 WITH 8 JUDGES AND 8 WINES BASED ON RANKS, IDENT=N Copyright (c) 1995-2010 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 Côte Rôtie La Mouline 2003 tied for 6th place Wine B is Côte Rôtie La Landonne 2001 tied for 6th place Wine C is Côte Rôtie La Mouline 2004 ........ 1st place Wine D is Côte Rôtie La Turque 2001 ........ 2nd place Wine E is Côte Rôtie La Turque 2003 ........ 5th place Wine F is Côte Rôtie La Mouline 2001 tied for 3rd place Wine G is Côte Rôtie La Landonne 2004 tied for 3rd place Wine H is Côte Rôtie La Landonne 2003 tied for 6th place
The Judges's Rankings
Judge Wine -> A B C D E F G H Bob 8. 4. 2. 3. 7. 1. 6. 5. Burt 5. 7. 2. 1. 4. 8. 3. 6. Mike 4. 3. 1. 2. 7. 6. 5. 8. Ed 4. 7. 5. 8. 3. 6. 1. 2. Tom 7. 2. 8. 3. 5. 1. 4. 6. John 5. 4. 2. 8. 6. 1. 7. 3. Zaki 5. 6. 2. 1. 3. 4. 8. 7. Dick 2. 7. 5. 6. 4. 8. 1. 3.
Table of Votes Against Wine -> A B C D E F G H
Group Ranking -> 6 6 1 2 5 3 3 6 Votes Against -> 40 40 27 32 39 35 35 40
( 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 = 0.0580
The probability that random chance could be responsible for this correlation is rather large, 0.8610. 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 Mike 0.2410 Burt 0.2275 Zaki 0.0838 Bob 0.0723 John -0.4286 Tom -0.4880 Dick -0.6429 Ed -0.6826
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 Côte Rôtie La Mouline 2004 2. ........ 2nd place Wine D is Côte Rôtie La Turque 2001 3. tied for 3rd place Wine G is Côte Rôtie La Landonne 2004 4. tied for 3rd place Wine F is Côte Rôtie La Mouline 2001 5. ........ 5th place Wine E is Côte Rôtie La Turque 2003 6. tied for 6th place Wine B is Côte Rôtie La Landonne 2001 7. tied for 6th place Wine A is Côte Rôtie La Mouline 2003 8. tied for 6th place Wine H is Côte Rôtie La Landonne 2003 We now test whether the ranksums AS A WHOLE provide a significant ordering. The Friedman Chi-square value is 3.2500. The probability that this could happen by chance is 0.8610 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 Burt Mike Bob 1.000 -0.071 0.357 Burt -0.071 1.000 0.524 Mike 0.357 0.524 1.000 Ed -0.595 0.000 -0.619 Tom 0.405 -0.429 -0.119 John 0.524 -0.619 -0.119 Zaki 0.405 0.476 0.500 Dick -0.786 0.333 -0.262 Ed Tom John Bob -0.595 0.405 0.524 Burt 0.000 -0.429 -0.619 Mike -0.619 -0.119 -0.119 Ed 1.000 -0.452 -0.071 Tom -0.452 1.000 -0.048 John -0.071 -0.048 1.000 Zaki -0.643 -0.071 -0.048 Dick 0.833 -0.619 -0.429 Zaki Dick Bob 0.405 -0.786 Burt 0.476 0.333 Mike 0.500 -0.262 Ed -0.643 0.833 Tom -0.071 -0.619 John -0.048 -0.429 Zaki 1.000 -0.500 Dick -0.500 1.000 Pairwise correlations in descending order 0.833 Ed and Dick Significantly positive 0.524 Bob and John Not significant 0.524 Burt and Mike Not significant 0.500 Mike and Zaki Not significant 0.476 Burt and Zaki Not significant 0.405 Bob and Tom Not significant 0.405 Bob and Zaki Not significant 0.357 Bob and Mike Not significant 0.333 Burt and Dick Not significant 0.000 Burt and Ed Not significant -0.048 Tom and John Not significant -0.048 John and Zaki Not significant -0.071 Ed and John Not significant -0.071 Bob and Burt Not significant -0.071 Tom and Zaki Not significant -0.119 Mike and John Not significant -0.119 Mike and Tom Not significant -0.262 Mike and Dick Not significant -0.429 John and Dick Not significant -0.429 Burt and Tom Not significant -0.452 Ed and Tom Not significant -0.500 Zaki and Dick Not significant -0.595 Bob and Ed Not significant -0.619 Burt and John Not significant -0.619 Tom and Dick Not significant -0.619 Mike and Ed Not significant -0.643 Ed and Zaki Not significant -0.786 Bob and Dick Significantly negative
COMMENT: This was a most remarkable tasting in that all the wines were extra- ordinary, which may explain the complete lack of correlation among them. They all had superb bouquets, fantastic fruit and satisfactory levels of acid. As one taster remarked, they were all flawless, and the rankings which are forced (no ties are ever allowed) reflected this feature. It is our feeling that these wines will last a long time, although in spite of their relatively young age, they are all perfectly ready to drink. In spite of these comments, we would expect all these wines to get better and even more enjoyable with time. Even though the following result is not statistically significant, it is interesting to note that the 2003s did not do better than any of the others, in fact did worse on the whole. It is also clear that some of the tasters had definite differential preferences among the vintages, but the results do not mean that there were no differences among the vineyards or vintages. It was also abundantly clear that the least liked wine by each taster would nevertheless make him very happy.
Return to previous page