WINETASTER ON 05/05/08 WITH 7 JUDGES AND 8 WINES BASED ON RANKS, IDENT=N Copyright (c) 1995-2003 Richard E. Quandt, V. 1.65

FLIGHT 1: Number of Judges = 7 Number of Wines = 8

Identification of the Wine: The judges' overall ranking:

Wine A is Hermitage La Chapelle 1985 ........ 7th place Wine B is Hermitage La Sizeranne 1989 ........ 6th place Wine C is Hermitage La Sizeranne 1990 ........ 1st place Wine D is Hermitage La Sizeranne 1988 ........ 2nd place Wine E is Hermitage La Chapelle 1989 ........ 3rd place Wine F is Hermitage La Chapelle 1988 ........ 8th place Wine G is Hermitage La Sizeranne 1985 ........ 5th place Wine H is Hermitage La Chapelle 1990 ........ 4th place

The Judges's Rankings

Judge Wine -> A B C D E F G H Mike 7. 4. 6. 2. 5. 8. 3. 1. Orley 3. 1. 7. 4. 2. 8. 6. 5. Ed 7. 5. 3. 1. 2. 8. 4. 6. Bob 2. 5. 1. 4. 3. 8. 7. 6. Burt 8. 7. 1. 3. 5. 6. 2. 4. John 8. 6. 2. 1. 7. 5. 3. 4. Dick 6. 5. 2. 8. 1. 3. 7. 4.

Table of Votes Against Wine -> A B C D E F G H

Group Ranking -> 7 6 1 2 3 8 5 4 Votes Against -> 41 33 22 23 25 46 32 30

( 7 is the best possible, 56 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 = 0.2478

The probability that random chance could be responsible for this correlation is quite small, 0.0960. 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 0.8503 Burt 0.4458 Bob 0.3333 Mike 0.3172 John 0.2857 Orley -0.0599 Dick -0.0714

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 Hermitage La Sizeranne 1990 2. ........ 2nd place Wine D is Hermitage La Sizeranne 1988 3. ........ 3rd place Wine E is Hermitage La Chapelle 1989 4. ........ 4th place Wine H is Hermitage La Chapelle 1990 5. ........ 5th place Wine G is Hermitage La Sizeranne 1985 6. ........ 6th place Wine B is Hermitage La Sizeranne 1989 7. ........ 7th place Wine A is Hermitage La Chapelle 1985 --------------------------------------------------- 8. ........ 8th place Wine F is Hermitage La Chapelle 1988 We now test whether the ranksums AS A WHOLE provide a significant ordering. The Friedman Chi-square value is 12.1429. The probability that this could happen by chance is 0.0960 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 Mike Orley Ed Mike 1.000 0.238 0.452 Orley 0.238 1.000 0.262 Ed 0.452 0.262 1.000 Bob -0.190 0.333 0.429 Burt 0.405 -0.524 0.595 John 0.476 -0.524 0.500 Dick -0.429 -0.119 -0.071 Bob Burt John Mike -0.190 0.405 0.476 Orley 0.333 -0.524 -0.524 Ed 0.429 0.595 0.500 Bob 1.000 0.071 -0.095 Burt 0.071 1.000 0.857 John -0.095 0.857 1.000 Dick 0.214 0.000 -0.310 Dick Mike -0.429 Orley -0.119 Ed -0.071 Bob 0.214 Burt 0.000 John -0.310 Dick 1.000 Pairwise correlations in descending order 0.857 Burt and John Significantly positive 0.595 Ed and Burt Not significant 0.500 Ed and John Not significant 0.476 Mike and John Not significant 0.452 Mike and Ed Not significant 0.429 Ed and Bob Not significant 0.405 Mike and Burt Not significant 0.333 Orley and Bob Not significant 0.262 Orley and Ed Not significant 0.238 Mike and Orley Not significant 0.214 Bob and Dick Not significant 0.071 Bob and Burt Not significant 0.000 Burt and Dick Not significant -0.071 Ed and Dick Not significant -0.095 Bob and John Not significant -0.119 Orley and Dick Not significant -0.190 Mike and Bob Not significant -0.310 John and Dick Not significant -0.429 Mike and Dick Not significant -0.524 Orley and John Not significant -0.524 Orley and Burt Not significant

COMMENT: This was an extremely memorable tasting. Fact 1 is that the 1988 La Chapelle was not in good condition. Fact 2 is that in this group the 1985 vintage performed less well than the other three. One opinion is that great Syrrah and, parenthetically, Nebbiolo-based wines provide more satisfying and complex taste experiences than most Cabernet or Merlot based wines. And a big part of that is that these are food-friendly wines, and we were served some excellent food with them. Overall, the tasting reminded us of how enjoyable the Hermitage wines are. We also performed the appropriate significance test to see whether in the aggregate the La Chapelle's better or worse than the La Sizeranne's. The test is based on the quantity R = (R_{1}/n_{1})/(R_{2}/n_{2}) where R_{i}and n_{i}are the sum of the rank-sums and the number of items in the i^{th}group (4 in this case for each group). R turns out to be 1.1356, which is not significant from the mean value, and hence neither group of wines is statistically superior to the other.

Return to previous page