WINETASTER ON 04/05/99 WITH 9 JUDGES AND 10 WINES BASED ON RANKS, IDENT=N
Copyright (c) 1995-99 Richard E. Quandt
FLIGHT 1:
Number of Judges = 9
Number of Wines = 10
Identification of the Wine: The judges' overall ranking:
Wine A is Federico Paternina Res.Espec. 1961 ........ 1st place
Wine B is Rioja Alta Gran Res.Cosecha 1985 ........ 2nd place
Wine C is Marques de Riscal Reserva 1995 ........ 3rd place
Wine D is Marques de Murrieta 1975 ........ 6th place
Wine E is Gran Conda Gran Reserva 1976 ........ 4th place
Wine F is Marques de Murrieta 1980 ........ 9th place
Wine G is Federico Paternina Gran Res. 1961 ........ 8th place
Wine H is Herencia Remondo Cosecha 1987 ........ 10th place
Wine I is Torre Albeniz Reserva 1994 ........ 5th place
Wine J is Marques de Murrieta 1966 ........ 7th place
The Judges's Rankings
Judge Wine -> A B C D E F G H I J
Bob 2. 8. 1. 4. 5. 9. 6. 7. 3. 10.
Frank V. 1. 2. 3. 4. 10. 9. 5. 6. 7. 8.
John 6. 1. 2. 5. 4. 3. 9. 10. 8. 7.
Frank L. 2. 1. 8. 9. 5. 4. 6. 7. 10. 3.
Ed 1. 4. 2. 9. 7. 10. 8. 6. 3. 5.
Burt 2. 1. 8. 5. 4. 10. 3. 7. 6. 9.
Orley 2. 3. 4. 1. 5. 6. 10. 8. 7. 9.
Grant 4. 5. 6. 9. 2. 7. 8. 10. 1. 3.
Dick 2. 1. 3. 5. 6. 10. 9. 8. 4. 7.
Table of Votes Against
Wine -> A B C D E F G H I J
Group Ranking -> 1 2 3 6 4 9 8 10 5 7
Votes Against -> 22 26 37 51 48 68 64 69 49 61
( 9 is the best possible, 90 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 = 0.3793
The probability that random chance could be responsible for this correlation
is quite small, 0.0003. 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
Dick 0.9152
Orley 0.6242
Ed 0.6201
Frank V. 0.5532
Bob 0.4909
Burt 0.4788
John 0.4182
Grant 0.2606
Frank L. 0.1337
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 Federico Paternina Reserva Especia
2. ........ 2nd place Wine B is La Rioja Alta Gran Reserva Cosecha
---------------------------------------------------
3. ........ 3rd place Wine C is Marques de Riscal Reserva 1995
4. ........ 4th place Wine E is Gran Conda Gran Reserva 1976
5. ........ 5th place Wine I is Torre Albeniz Reserva 1994
6. ........ 6th place Wine D is Marques de Murrieta 1975
7. ........ 7th place Wine J is Marques de Murrieta 1966
8. ........ 8th place Wine G is Fedirico Paternina Gran Reserva 19
---------------------------------------------------
9. ........ 9th place Wine F is Marques de Murrieta 1980
10. ........ 10th place Wine H is Herenic Remondo Cosecha 1987
We now test whether the ranksums AS A WHOLE provide a significant ordering.
The Friedman Chi-square value is 30.7212. The probability that this could
happen by chance is 0.0003
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
Bob Frank V. John
Bob 1.000 0.467 0.055
Frank V. 0.467 1.000 0.188
John 0.055 0.188 1.000
Frank L. -0.491 0.164 0.345
Ed 0.527 0.552 0.067
Burt 0.273 0.564 0.067
Orley 0.479 0.539 0.636
Grant 0.139 -0.248 0.164
Dick 0.539 0.697 0.455
Frank L. Ed Burt
Bob -0.491 0.527 0.273
Frank V. 0.164 0.552 0.564
John 0.345 0.067 0.067
Frank L. 1.000 0.127 0.309
Ed 0.127 1.000 0.261
Burt 0.309 0.261 1.000
Orley 0.091 0.212 0.370
Grant 0.176 0.491 0.103
Dick 0.152 0.770 0.552
Orley Grant Dick
Bob 0.479 0.139 0.539
Frank V. 0.539 -0.248 0.697
John 0.636 0.164 0.455
Frank L. 0.091 0.176 0.152
Ed 0.212 0.491 0.770
Burt 0.370 0.103 0.552
Orley 1.000 -0.006 0.685
Grant -0.006 1.000 0.394
Dick 0.685 0.394 1.000
Pairwise correlations in descending order
0.770 Ed and Dick Significantly positive
0.697 Frank V. and Dick Significantly positive
0.685 Orley and Dick Significantly positive
0.636 John and Orley Significantly positive
0.564 Frank V. and Burt Not significant
0.552 Burt and Dick Not significant
0.552 Frank V. and Ed Not significant
0.539 Bob and Dick Not significant
0.539 Frank V. and Orley Not significant
0.527 Bob and Ed Not significant
0.491 Ed and Grant Not significant
0.479 Bob and Orley Not significant
0.467 Bob and Frank V. Not significant
0.455 John and Dick Not significant
0.394 Grant and Dick Not significant
0.370 Burt and Orley Not significant
0.345 John and Frank L. Not significant
0.309 Frank L. and Burt Not significant
0.273 Bob and Burt Not significant
0.261 Ed and Burt Not significant
0.212 Ed and Orley Not significant
0.188 Frank V. and John Not significant
0.176 Frank L. and Grant Not significant
0.164 John and Grant Not significant
0.164 Frank V. and Frank L. Not significant
0.152 Frank L. and Dick Not significant
0.139 Bob and Grant Not significant
0.127 Frank L. and Ed Not significant
0.103 Burt and Grant Not significant
0.091 Frank L. and Orley Not significant
0.067 John and Burt Not significant
0.067 John and Ed Not significant
0.055 Bob and John Not significant
-0.006 Orley and Grant Not significant
-0.248 Frank V. and Grant Not significant
-0.491 Bob and Frank L. Not significant
COMMENT:
There was no flaw or oxidized wine in the tasting. All wines had good
colors, no off smells, and they had good fruit. Ed says that he is
extremely surprised by the difference the 1961 Paternina Reserva Especial
and the Paternina Grand Reserva of the same year. One question this
raises is whether this is bottle variation or actual quality difference.
Both corks fell apart upon extraction, both were reasonably high shoul-
der. Frank was more surprised that both 1961s showed up so well. Several
people were surprised by the longevity of these wines. John notes that
the third best wine is a 1995, which is unusual because Riojas have to
age more. Ed further notes that these are not fat wines.
Dick further notes that the statistical measures of correlation among the judges
indicate absolutely exceptional agreement among the judges.
Return to previous page