WINETASTER ON 01/26/99 WITH 5 JUDGES AND 9 WINES BASED ON RANKS, IDENT=N
Copyright (c) 1995-98 Richard E. Quandt
FLIGHT 1: All wines are 1995
Number of Judges = 5
Number of Wines = 9
Identification of the Wine: The judges' overall ranking:
Wine A is Sainte Agnes, Ermitage de Pic St L ........ 8th place
Wine B is La Roque, Cupa Numismae tied for 2nd place
Wine C is L'Aigueliere, Cote Doree ........ 1st place
Wine D is Lascaux, Nobles Pierres ........ 5th place
Wine E is Hortus, Grande Cuvee tied for 2nd place
Wine F is Prieure de Ch de Segure, Vielles V ........ 9th place
Wine G is Mansenoble, Corbiere ........ 6th place
Wine H is Mortes, Grande Cuvee ........ 7th place
Wine I is L'Aigueliere, Fut de Chene ........ 4th place
The Judges's Rankings
Judge Wine -> A B C D E F G H I
Andre 6. 7. 2. 4. 3. 9. 8. 5. 1.
Roland 9. 3. 1. 8. 5. 6. 7. 4. 2.
Kai 7. 4. 9. 2. 3. 5. 1. 6. 8.
Wolfgang 4. 2. 1. 6. 3. 8. 5. 9. 7.
Karl 8. 3. 1. 4. 5. 7. 6. 9. 2.
Table of Votes Against
Wine -> A B C D E F G H I
Group Ranking -> 8 2 1 5 2 9 6 7 4
Votes Against -> 34 19 14 24 19 35 27 33 20
( 5 is the best possible, 45 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 = 0.3120
The probability that random chance could be responsible for this correlation
is rather large, 0.1310. 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
Karl 0.8167
Wolfgang 0.4519
Roland 0.4118
Andre 0.3123
Kai -0.2929
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 L'Aigueliere, Cote Doree
---------------------------------------------------
2. tied for 2nd place Wine B is La Roque, Cupa Numismae
3. tied for 2nd place Wine E is Hortus, Grande Cuvee
4. ........ 4th place Wine I is L'Aigueliere, Fut de Chene
5. ........ 5th place Wine D is Lascaux, Nobles Pierres
6. ........ 6th place Wine G is Mansenoble, Corbiere
7. ........ 7th place Wine H is Mortes, Grande Cuvee
8. ........ 8th place Wine A is Sainte Agnes, Ermitage de Pic St L
---------------------------------------------------
9. ........ 9th place Wine F is Prieure de Ch de Segure, Vielles V
We now test whether the ranksums AS A WHOLE provide a significant ordering.
The Friedman Chi-square value is 12.4800. The probability that this could
happen by chance is 0.1310
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
Andre Roland Kai
Andre 1.000 0.517 -0.483
Roland 0.517 1.000 -0.550
Kai -0.483 -0.550 1.000
Wolfgang 0.200 0.233 -0.067
Karl 0.583 0.633 -0.233
Wolfgang Karl
Andre 0.200 0.583
Roland 0.233 0.633
Kai -0.067 -0.233
Wolfgang 1.000 0.567
Karl 0.567 1.000
Pairwise correlations in descending order
0.633 Roland and Karl Significantly positive
0.583 Andre and Karl Not significant
0.567 Wolfgang and Karl Not significant
0.517 Andre and Roland Not significant
0.233 Roland and Wolfgang Not significant
0.200 Andre and Wolfgang Not significant
-0.067 Kai and Wolfgang Not significant
-0.233 Kai and Karl Not significant
-0.483 Andre and Kai Not significant
-0.550 Roland and Kai Not significant
COMMENT:
Kai: The L'Aigueliere, Cote Doree seems to be by far to green; the Mortes
has a rough structure; Andre: There are two different kinds of wine: the spicy
and the fruity. The result of such a tasting depends on the preferences.
Return to previous page