WINETASTER ON 03/27/00 WITH 8 JUDGES AND 7 WINES BASED ON RANKS, IDENT=N
Copyright (c) 1995-2000 Richard E. Quandt
FLIGHT 1:
Number of Judges = 8
Number of Wines = 7
Identification of the Wine: The judges' overall ranking:
Wine A is Heitz Cellar, Cabernet 96 ........ 7th place
Wine B is Swanson, Cabernet 97 ........ 6th place
Wine C is Frogs Leap, Cabernet 97 tied for 4th place
Wine D is Gallo Sonoma, Cabernet 96 tied for 4th place
Wine E is L Escrime, Cabernet 97 ........ 2nd place
Wine F is Sterling, Cabernet 97 ........ 1st place
Wine G is Glen Ellen, Cabernet 97 ........ 3rd place
The Judges's Rankings
Judge Wine -> A B C D E F G
Nahomi 7. 3. 6. 5. 1. 4. 2.
Raul 6. 7. 4. 3. 5. 1. 2.
Lisa 6. 2. 3. 7. 5. 1. 4.
Gaby 5. 7. 4. 1. 3. 6. 2.
Stefan 7. 4. 5. 1. 3. 2. 6.
Hans 7. 5. 2. 4. 1. 3. 6.
Anita 6. 7. 5. 4. 3. 2. 1.
Karl 7. 3. 1. 5. 4. 2. 6.
Table of Votes Against
Wine -> A B C D E F G
Group Ranking -> 7 6 4 4 2 1 3
Votes Against -> 51 38 30 30 25 21 29
( 8 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 = 0.3259
The probability that random chance could be responsible for this correlation
is quite small, 0.0158. 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
Anita 0.5714
Raul 0.5637
Stefan 0.4286
Hans 0.3929
Karl 0.2857
Nahomi 0.2342
Lisa 0.1081
Gaby 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 F is Sterling, Cabernet 97
---------------------------------------------------
2. ........ 2nd place Wine E is L Escrime, Cabernet 97
3. ........ 3rd place Wine G is Glen Ellen, Cabernet 97
4. tied for 4th place Wine D is Gallo Sonoma, Cabernet 96
5. tied for 4th place Wine C is Frogs Leap, Cabernet 97
6. ........ 6th place Wine B is Swanson, Cabernet 97
---------------------------------------------------
7. ........ 7th place Wine A is Heitz Cellar, Cabernet 96
We now test whether the ranksums AS A WHOLE provide a significant ordering.
The Friedman Chi-square value is 15.6429. The probability that this could
happen by chance is 0.0158
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
Nahomi Raul Lisa
Nahomi 1.000 0.107 0.214
Raul 0.107 1.000 0.179
Lisa 0.214 0.179 1.000
Gaby 0.143 0.393 -0.714
Stefan 0.250 0.357 0.036
Hans 0.321 0.179 0.214
Anita 0.500 0.857 0.071
Karl 0.036 0.143 0.714
Gaby Stefan Hans
Nahomi 0.143 0.250 0.321
Raul 0.393 0.357 0.179
Lisa -0.714 0.036 0.214
Gaby 1.000 0.179 0.107
Stefan 0.179 1.000 0.571
Hans 0.107 0.571 1.000
Anita 0.500 0.214 0.214
Karl -0.393 0.393 0.714
Anita Karl
Nahomi 0.500 0.036
Raul 0.857 0.143
Lisa 0.071 0.714
Gaby 0.500 -0.393
Stefan 0.214 0.393
Hans 0.214 0.714
Anita 1.000 -0.071
Karl -0.071 1.000
Pairwise correlations in descending order
0.857 Raul and Anita Significantly positive
0.714 Lisa and Karl Significantly positive
0.714 Hans and Karl Significantly positive
0.571 Stefan and Hans Not significant
0.500 Gaby and Anita Not significant
0.500 Nahomi and Anita Not significant
0.393 Raul and Gaby Not significant
0.393 Stefan and Karl Not significant
0.357 Raul and Stefan Not significant
0.321 Nahomi and Hans Not significant
0.250 Nahomi and Stefan Not significant
0.214 Stefan and Anita Not significant
0.214 Lisa and Hans Not significant
0.214 Nahomi and Lisa Not significant
0.214 Hans and Anita Not significant
0.179 Gaby and Stefan Not significant
0.179 Raul and Lisa Not significant
0.179 Raul and Hans Not significant
0.143 Nahomi and Gaby Not significant
0.143 Raul and Karl Not significant
0.107 Gaby and Hans Not significant
0.107 Nahomi and Raul Not significant
0.071 Lisa and Anita Not significant
0.036 Lisa and Stefan Not significant
0.036 Nahomi and Karl Not significant
-0.071 Anita and Karl Not significant
-0.393 Gaby and Karl Not significant
-0.714 Lisa and Gaby Significantly negative
COMMENT:
Like the tasting 3 months ago this was a tasting of wines of the lower
and medium price range. Like the tasting before the judges were mostly
inexpierenced at tasting wine.
The cheapest wines were the Glen Ellen (4$) and
Gallo Sonoma (7$). Sterling Vineyards Cabernet was 14$, L Escrime
20$, Heitz Cellar 23$, Frogs Leap 27$ and Swanson 31$. Everybody agreed
that the poor quality of Heitz was astounding. However, this was almost the only
consensus. Except Heitz and Swanson every wine was ranked at least once on the
first place. It is remarkable that one of the cheapest wines turned out
as clear winner (Sterling, 14 $).
Accordingly, a simple regression between points (dependent variable) and
bottle price (independent varible) indicates almost no relationship:
points = 0.4*price + 24.6
The R-square is 0.17.
Return to previous page