WINETASTER ON 01/15/99 WITH 7 JUDGES AND 10 WINES BASED ON RANKS, IDENT=N
Copyright (c) 1995-98 Richard E. Quandt
FLIGHT 1:
Number of Judges = 7
Number of Wines = 10
Identification of the Wine: The judges' overall ranking:
Wine A is Pouilly Fume Guysaget 96 ........ 8th place
Wine B is Sancerre Roger 96 ........ 10th place
Wine C is Sancerre Bourgois 96 ........ 3rd place
Wine D is Villa Maria NZ 97 ........ 2nd place
Wine E is Pouilly Fume Bourgois 97 ........ 5th place
Wine F is Pouilly Fume Seguin 97 ........ 6th place
Wine G is Selaks 97 ........ 7th place
Wine H is TementSteirisch Klassik, AUT 97 ........ 4th place
Wine I is Pichler AUT 96 ........ 9th place
Wine J is CloudyBay 97 ........ 1st place
The Judges's Rankings
Judge Wine -> A B C D E F G H I J
Kai 9. 8. 1. 3. 4. 10. 6. 5. 7. 2.
Geli 3. 5. 8. 2. 9. 7. 4. 6. 10. 1.
Doro 2. 9. 8. 4. 3. 5. 6. 7. 10. 1.
Pascale 7. 4. 6. 8. 2. 5. 9. 3. 1. 10.
Wolfgang 10. 9. 3. 1. 5. 7. 4. 6. 8. 2.
Pit 9. 10. 4. 6. 5. 3. 7. 2. 8. 1.
Karl 6. 10. 3. 2. 8. 7. 9. 5. 4. 1.
Table of Votes Against
Wine -> A B C D E F G H I J
Group Ranking -> 8 10 3 2 5 6 7 4 9 1
Votes Against -> 46 55 33 26 36 44 45 34 48 18
( 7 is the best possible, 70 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 = 0.2782
The probability that random chance could be responsible for this correlation
is quite small, 0.0411. 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
Kai 0.6930
Wolfgang 0.6727
Karl 0.5879
Pit 0.5636
Doro 0.2918
Geli 0.1636
Pascale -0.5289
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 J is CloudyBay 97
---------------------------------------------------
2. ........ 2nd place Wine D is Villa Maria NZ 97
3. ........ 3rd place Wine C is Sancerre Bourgois 96
4. ........ 4th place Wine H is TementSteirisch Klassik, AUT 97
5. ........ 5th place Wine E is Pouilly Fume Bourgois 97
6. ........ 6th place Wine F is Pouilly Fume Seguin 97
7. ........ 7th place Wine G is Selaks 97
8. ........ 8th place Wine A is Pouilly Fume Guysaget 96
9. ........ 9th place Wine I is Pichler AUT 96
---------------------------------------------------
10. ........ 10th place Wine B is Sancerre Roger 96
We now test whether the ranksums AS A WHOLE provide a significant ordering.
The Friedman Chi-square value is 17.5247. The probability that this could
happen by chance is 0.0411
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 0.65 for significance at the 0.05
level and must exceed 0.56 for significance at the 0.1 level
Kai Geli Doro
Kai 1.000 0.127 0.152
Geli 0.127 1.000 0.600
Doro 0.152 0.600 1.000
Pascale -0.285 -0.855 -0.564
Wolfgang 0.842 0.321 0.297
Pit 0.491 0.067 0.345
Karl 0.624 0.261 0.248
Pascale Wolfgang Pit
Kai -0.285 0.842 0.491
Geli -0.855 0.321 0.067
Doro -0.564 0.297 0.345
Pascale 1.000 -0.527 -0.188
Wolfgang -0.527 1.000 0.576
Pit -0.188 0.576 1.000
Karl -0.309 0.576 0.515
Karl
Kai 0.624
Geli 0.261
Doro 0.248
Pascale -0.309
Wolfgang 0.576
Pit 0.515
Karl 1.000
Pairwise correlations in descending order
0.842 Kai and Wolfgang Significantly positive
0.624 Kai and Karl Significantly positive
0.600 Geli and Doro Significantly positive
0.576 Wolfgang and Karl Significantly positive
0.576 Wolfgang and Pit Significantly positive
0.515 Pit and Karl Not significant
0.491 Kai and Pit Not significant
0.345 Doro and Pit Not significant
0.321 Geli and Wolfgang Not significant
0.297 Doro and Wolfgang Not significant
0.261 Geli and Karl Not significant
0.248 Doro and Karl Not significant
0.152 Kai and Doro Not significant
0.127 Kai and Geli Not significant
0.067 Geli and Pit Not significant
-0.188 Pascale and Pit Not significant
-0.285 Kai and Pascale Not significant
-0.309 Pascale and Karl Not significant
-0.527 Pascale and Wolfgang Not significant
-0.564 Doro and Pascale Not significant
-0.855 Geli and Pascale Significantly negative
COMMENT:
Kai: New Zealands wines are absolutely parfumed, without any elegance
Tement's wine is considered to be almost a French wine, the real
outstanding wine is: Bourgois Sancerre, the other French wines were
nothing special
Wolfgang: Pichler is an example how to kill a wine with oak.
Amazing is the Villa Maria because of its non-over-parfumed flavour
Pascale: New Zealand are too straight
Karl: It is hard to taste these different wines agains one another.
The styles are too different.
Return to previous page