Report

WINETASTER ON 12/03/12 WITH 8 JUDGES AND 8 WINES BASED ON RANKS, IDENT=N
Copyright (c) 1995-2012 Richard E. Quandt, V. 1.65

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

Identification of the Wine: The judges' overall ranking:

Wine A is Torre Muga 2005 ........ 1st place (Rioja)
Wine B is Bodegas Alion (Vega Sicilia) 2003 ........ 2nd place (Ribera del Duero)
Wine C is Contino Vina del Olivo 2007 ........ 6th place (Rioja)
Wine D is Remirez de Ganuza Reserva 2005 ........ 3rd place (Rioja)
Wine E is Hacienda Monasterio 2007 ........ 4th place (Ribera del Duero)
Wine F is Vega Sicilia "Unico" 2002 ........ 5th place (Ribera del Duero)
Wine G is Prado Enea Gran Reserva 2004 ........ 7th place (Rioja)
Wine H is Marques de Riscal Reserva 2006 ........ 8th place (Rioja)

The Judges's Rankings

Judge Wine -> A B C D E F G H

Bob 4. 2. 3. 6. 8. 1. 5. 7.
Mike 7. 4. 6. 8. 1. 3. 2. 5.
Ed 3. 6. 8. 4. 1. 7. 5. 2.
Frank 2. 5. 8. 1. 7. 3. 6. 4.
Burt 1. 2. 3. 4. 7. 5. 6. 8.
Zachy 5. 6. 4. 3. 1. 7. 2. 8.
Orley 4. 1. 3. 5. 6. 2. 8. 7.
Dick 1. 5. 2. 3. 4. 8. 6. 7.

Table of Votes Against

Wine -> A B C D E F G H

Group Ranking -> 1 2 6 3 4 5 7 8
Votes Against -> 27 31 37 34 35 36 40 48

( 8 is the best possible, 64 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 = 0.1012

The probability that random chance could be responsible for this correlation
is rather large, 0.5792. 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 Correlation Price

Burt 0.5389 -0.3374
Dick 0.2048 -0.3374
Orley 0.0482 -0.3012
Frank -0.0240 0.2169
Bob -0.1437 -0.0241
Zachy -0.1905 0.3374
Ed -0.4192 0.0120
Mike -0.6667 0.3615

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 A is Torre Muga 2005
2. ........ 2nd place Wine B is Bodegas Alion (Vega Sicilia) 2003
3. ........ 3rd place Wine D is Remirez de Ganuza Reserva 2005
4. ........ 4th place Wine E is Hacienda Monasterio 2007
5. ........ 5th place Wine F is Vega Sicilia "Unico" 2002
6. ........ 6th place Wine C is Contino Vina del Olivo 2007
7. ........ 7th place Wine G is Prado Enea Gran Reserva 2004
---------------------------------------------------
8. ........ 8th place Wine H is Marques de Riscal Reserva 2006

We now test whether the ranksums AS A WHOLE provide a significant ordering.
The Friedman Chi-square value is 5.6667. The probability that this could
happen by chance is 0.5792

We now test whether the group ranking of wines is correlated with the
prices of the wines. The rank correlation between them is -0.1566. At the
10% level of significance this would have to exceed the critical value of
0.5240 to be significant.

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

Bob Mike Ed

Bob 1.000 -0.095 -0.857
Mike -0.095 1.000 0.119
Ed -0.857 0.119 1.000
Frank 0.071 -0.571 0.190
Burt 0.619 -0.548 -0.452
Zachy -0.452 0.262 0.214
Orley 0.810 -0.214 -0.619
Dick -0.119 -0.571 0.071

Frank Burt Zachy

Bob 0.071 0.619 -0.452
Mike -0.571 -0.548 0.262
Ed 0.190 -0.452 0.214
Frank 1.000 0.238 -0.357
Burt 0.238 1.000 -0.071
Zachy -0.357 -0.071 1.000
Orley 0.095 0.690 -0.405
Dick 0.000 0.643 0.429

Orley Dick

Bob 0.810 -0.119
Mike -0.214 -0.571
Ed -0.619 0.071
Frank 0.095 0.000
Burt 0.690 0.643
Zachy -0.405 0.429
Orley 1.000 0.119
Dick 0.119 1.000

Pairwise correlations in descending order

0.810 Bob and Orley Significantly positive
0.690 Burt and Orley Significantly positive
0.643 Burt and Dick Not significant
0.619 Bob and Burt Not significant
0.429 Zachy and Dick Not significant
0.262 Mike and Zachy Not significant
0.238 Frank and Burt Not significant
0.214 Ed and Zachy Not significant
0.190 Ed and Frank Not significant
0.119 Mike and Ed Not significant
0.119 Orley and Dick Not significant
0.095 Frank and Orley Not significant
0.071 Bob and Frank Not significant
0.071 Ed and Dick Not significant
0.000 Frank and Dick Not significant
-0.071 Burt and Zachy Not significant
-0.095 Bob and Mike Not significant
-0.119 Bob and Dick Not significant
-0.214 Mike and Orfley Not significant
-0.357 Frank and Zachy Not significant
-0.405 Zachy and Orley Not significant
-0.452 Bob and Zachy Not significant
-0.452 Ed and Burt Not significant
-0.548 Mike and Burt Not significant
-0.571 Mike and Dick Not significant
-0.571 Mike and Frank Not significant
-0.619 Ed and Orley Not significant
-0.857 Bob and Ed Significantly negative

COMMENT:

These were fantastic wines, that were very difficult to distinguish,
and all had distinctive tempranillo character. Our host gave us a very
difficult challenge, which we relish. The prices of the wines were
enormously different, ranging from $15/bottle to $250/bottle, but the
quality of the winemakers was uniformly very
high. Only one distinguished member of this group correctly identified the
Vega Sicilia Unico. These were all very well made wines, but produced in
somewhat different styles. So it is understandable that some members of the
group preferred wines quite different from the wines preferred by others. In fact,
the overall agreement among the members was low, as can be seen from the fact that
some wines were ranked first as well as last by some members of the group.
In the aggregate, the lowest price wine was rated lowest, but four members of the
group had negative correlations with the prices of the wines. These are relatively
young wines but they are all drinking very well. While the number 1 wine was a Rioja,
on average the Riberas ranked slightly higher than the Riojas. However, the statistical
significance test on the rank sums of the 5 Riojas versus the 3 Ribera del Dueros yields
a test statistic of 0.914, not significantly different from 1.0. (See Journal of Wine
Economics, Vol. 2, No.1, 2007.)

Return to previous page