WINETASTER ON 05/07/12 WITH 7 JUDGES AND 7 WINES BASED ON RANKS, IDENT=N Copyright (c) 1995-2012 Richard E. Quandt, V. 1.65
FLIGHT 1: Number of Judges = 7 Number of Wines = 7
Identification of the Wine: The judges' overall ranking:
Wine A is Chave Hermitage 1989 ........ 2nd place Wine B is Chave Hermitage 1985 ........ 4th place Wine C is Chave Hermitage 1990 ........ 1st place Wine D is Chave Hermitage 1988 ........ 3rd place Wine E is Chave Hermitage 1998 tied for 5th place Wine F is Chave Hermitage 2000 ........ 7th place Wine G is Chave Hermitage 1994 tied for 5th place
The Judges's Rankings
Judge Wine -> A B C D E F G Mike 3. 5. 2. 1. 6. 7. 4. Orley 6. 7. 1. 3. 2. 4. 5. Ed 3. 5. 7. 1. 6. 2. 4. Bob 2. 3. 7. 5. 4. 6. 1. Zaki 3. 1. 2. 4. 6. 7. 5. Robby 3. 7. 1. 5. 4. 2. 6. Dick 3. 2. 1. 6. 4. 5. 7.
Table of Votes Against Wine -> A B C D E F G
Group Ranking -> 2 4 1 3 5 7 5 Votes Against -> 23 30 21 25 32 33 32
( 7 is the best possible, 49 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 = 0.1050
The probability that random chance could be responsible for this correlation is rather large, 0.6216. 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 0.6667 Zaki 0.1261 Robby 0.0741 Dick -0.0748 Orley -0.1429 Ed -0.2857 Bob -0.6307
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 Chave Hermitage 1990 2. ........ 2nd place Wine A is Chave Hermitage 1989 3. ........ 3rd place Wine D is Chave Hermitage 1988 4. ........ 4th place Wine B is Chave Hermitage 1985 5. tied for 5th place Wine E is Chave Hermitage 1998 6. tied for 5th place Wine G is Chave Hermitage 1994 7. ........ 7th place Wine F is Chave Hermitage 2000 We now test whether the ranksums AS A WHOLE provide a significant ordering. The Friedman Chi-square value is 4.4082. The probability that this could happen by chance is 0.6216 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 0.79 for significance at the 0.05 level and must exceed 0.71 for significance at the 0.1 level Mike Orley Ed Mike 1.000 0.214 0.107 Orley 0.214 1.000 -0.321 Ed 0.107 -0.321 1.000 Bob -0.071 -0.714 0.107 Zaki 0.536 -0.286 -0.357 Robby 0.036 0.607 -0.143 Dick 0.071 0.071 -0.643 Bob Zaki Robby Mike -0.071 0.536 0.036 Orley -0.714 -0.286 0.607 Ed 0.107 -0.357 -0.143 Bob 1.000 0.071 -0.679 Zaki 0.071 1.000 -0.214 Robby -0.679 -0.214 1.000 Dick -0.357 0.679 0.357 Dick Mike 0.071 Orley 0.071 Ed -0.643 Bob -0.357 Zaki 0.679 Robby 0.357 Dick 1.000 Pairwise correlations in descending order 0.679 Zaki and Dick Not significant 0.607 Orley and Robby Not significant 0.536 Mike and Zaki Not significant 0.357 Robby and Dick Not significant 0.214 Mike and Orley Not significant 0.107 Ed and Bob Not significant 0.107 Mike and Ed Not significant 0.071 Mike and Dick Not significant 0.071 Orley and Dick Not significant 0.071 Bob and Zaki Not significant 0.036 Mike and Robby Not significant -0.071 Mike and Bob Not significant -0.143 Ed and Robby Not significant -0.214 Zaki and Robby Not significant -0.286 Orley and Zaki Not significant -0.321 Orley and Ed Not significant -0.357 Bob and Dick Not significant -0.357 Ed and Zaki Not significant -0.643 Ed and Dick Not significant -0.679 Bob and Robby Not significant -0.714 Orley and Bob Significantly negative
COMMENT: Much to our delight, our host poured the mature, delicious 1997 and 1999 Mayacamas Chardonnay's as we gathered before this wine tasting. Overall, this was a fantastic tasting of consistent and typical wines of the great Hermitage Appelation. There is a great consistency of style across the vintages that is typical of the appelation and the house. Several tasters noted that as the tasting was going on that they changed their ranking an unusual number of times. There was wide agrement that these wines have noticeable acidity and would be excellent with a meal. These wines are all very expensive, averaging about $225-$250/bottle. We conclude that while there may be preference differences among the judges, these were such fantastic wines that they were all judged to be excellent. One taster, Bob, noted that he was rather negative in his evaluation of wine C, the 1990, but that might be explained by the fact that his glass was filled from the bottom of the bottle and thus had unusual amount of dregs in it. We there- fore reran the computer programwithou Bob. The new Kendall W turned out to be 0.2063, overall still not significant (the probability that it could have occurred by chance now being 0.2830, quite a bit lower than the 0.6216 we had before). However, wine C now turns out to be significantly high in quality. Of course, we don't mean this literally, since once you have computed statistics from the data, you really can't just omit some of the data and recompute, but it is an interesting exercise nevertheless.
Return to previous page