WINETASTER ON 11/05/19 WITH 8 JUDGES AND 6 WINES BASED ON RANKS, IDENT=N Copyright (c) 1995-2019 Richard E. Quandt, V. 1.65
A Vertical Tastingo of Ch. Musar
FLIGHT 1: Number of Judges = 8 Number of Wines = 6
Identification of the Wine: The judges' overall ranking:
Wine A is 2001 ........ 6th place Wine B is 1995 ........ 4th place Wine C is 1997 ........ 5th place Wine D is 1994 ........ 1st place Wine E is 2002 tied for 2nd place Wine F is 1998 tied for 2nd place
The Judges's Rankings
Judge Wine -> A B C D E F Mike 3. 6. 2. 1. 5. 4. Bob 6. 2. 4. 1. 3. 5. Stephen 4. 3. 5. 1. 6. 2. Burt 6. 2. 5. 3. 1. 4. Zaki 3. 2. 5. 1. 4. 6. Orley 5. 6. 4. 3. 2. 1. Ed 5. 3. 4. 6. 1. 2. Dick 5. 3. 6. 1. 4. 2.
Table of Votes Against Wine -> A B C D E F
Group Ranking -> 6 4 5 1 2 2 Votes Against -> 37 27 35 17 26 26
( 8 is the best possible, 48 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 = 0.2321
The probability that random chance could be responsible for this correlation is quite small, 0.0982. 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 0.7537 Bob 0.5508 Burt 0.5218 Stephen 0.3479 Zaki 0.1429 Orley 0.0857 Mike -0.0580 Ed -0.0580
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 D is 1994 --------------------------------------------------- 2. tied for 2nd place Wine E is 2002 3. tied for 2nd place Wine F is 1998 4. ........ 4th place Wine B is 1995 5. ........ 5th place Wine C is 1997 --------------------------------------------------- 6. ........ 6th place Wine A is 2001 We now test whether the ranksums AS A WHOLE provide a significant ordering. The Friedman Chi-square value is 9.2857. The probability that this could happen by chance is 0.0982 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 0.89 for significance at the 0.05 level and must exceed 0.83 for significance at the 0.1 level Mike Bob Stephen Mike 1.000 0.029 0.314 Bob 0.029 1.000 0.314 Stephen 0.314 0.314 1.000 Burt -0.543 0.714 -0.086 Zaki 0.143 0.657 0.371 Orley 0.143 -0.086 0.086 Ed -0.771 -0.143 -0.486 Dick 0.029 0.543 0.829 Burt Zaki Orley Mike -0.543 0.143 0.143 Bob 0.714 0.657 -0.086 Stephen -0.086 0.371 0.086 Burt 1.000 0.257 0.200 Zaki 0.257 1.000 -0.543 Orley 0.200 -0.543 1.000 Ed 0.543 -0.600 0.429 Dick 0.429 0.371 0.371 Ed Dick Mike -0.771 0.029 Bob -0.143 0.543 Stephen -0.486 0.829 Burt 0.543 0.429 Zaki -0.600 0.371 Orley 0.429 0.371 Ed 1.000 -0.086 Dick -0.086 1.000 Pairwise correlations in descending order 0.829 Stephen and Dick Not significant 0.714 Bob and Burt Not significant 0.657 Bob and Zaki Not significant 0.543 Burt and Ed Not significant 0.543 Bob and Dick Not significant 0.429 Orley and Ed Not significant 0.429 Burt and Dick Not significant 0.371 Stephen and Zaki Not significant 0.371 Orley and Dick Not significant 0.371 Zaki and Dick Not significant 0.314 Mike and Stephen Not significant 0.314 Bob and Stephen Not significant 0.257 Burt and Zaki Not significant 0.200 Burt and Orley Not significant 0.143 Mike and Orley Not significant 0.143 Mike and Zaki Not significant 0.086 Stephen and Orley Not significant 0.029 Mike and Dick Not significant 0.029 Mike and Bob Not significant -0.086 Stephen and Burt Not significant -0.086 Bob and Orley Not significant -0.086 Ed and Dick Not significant -0.143 Bob and Ed Not significant -0.486 Stephen and Ed Not significant -0.543 Mike and Burt Not significant -0.543 Zaki and Orley Not significant -0.600 Zaki and Ed Not significant -0.771 Mike and Ed Not significant
COMMENT: The present tasting is very similar to the one held in January 2019: the only difference is that our host could not duplicate the 2007 Musar. It was replace by the 1998 Musar; hence five of the six wines were idsntical in the two tastings. The basic purpose of the experiment was to assess the intertemporal consistency of the choices made by the group. It should also be noted that one person present in January was replaced by a different taster; thus seven of the eight tasters were identical. The Spearman rho rank correlation between the two sets of results (truncated to five) turns out to be -0.32; the question of statistical significance does not even arise since the two sets of rankings are negative correlated, which is certainly a disappointing result. This is driven predominantly by tghe ranks assigned to the 2001 Musar which was highly rated in January but dead last in November. Within the November tasting the results were much more consistent: five of the tasters ranked the 1994 first, (although one taster ranked it last). The wines were all pleasurable to drink, even the ones not ranked highly.
Return to previous page