WINETASTER ON 2/10/97 WITH 7 JUDGES AND 6 WINES BASED ON RANKS, IDENT=N Copyright (c) 1995-2011 Richard E. Quandt, V. 1.65

FLIGHT 1: Number of Judges = 7 Number of Wines = 6

Identification of the Wine: The judges' overall ranking:

Wine A is Baron Philippe 1983 ........ 3rd place Wine B is Clos Rene 1982 tied for 1st place Wine C is Marbuzet 1982 ........ 6th place Wine D is Marbuzet 1983 tied for 1st place Wine E is Baron Philippe 1982 ........ 4th place Wine F is Clos Rene 1983 ........ 5th place

The Judges's Rankings

Judge Wine -> A B C D E F Burt 3. 6. 1. 5. 4. 2. David 3. 2. 6. 5. 1. 4. John 2. 3. 6. 1. 5. 4. Orley 5. 1. 6. 2. 3. 4. Ed 4. 1. 5. 2. 3. 6. Bob 1. 5. 4. 2. 6. 3. Dick 3. 1. 6. 2. 4. 5.

Table of Votes Against Wine -> A B C D E F

Group Ranking -> 3 1 6 1 4 5 Votes Against -> 21 19 34 19 26 28

( 7 is the best possible, 42 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 = 0.2070

The probability that random chance could be responsible for this correlation is rather large, 0.2031. 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 0.8986 John 0.7714 Ed 0.4348 Orley 0.3714 David 0.0857 Bob -0.0580 Burt -0.9429

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. tied for 1st place Wine D is Marbuzet 1983 2. tied for 1st place Wine B is Clos Rene 1982 3. ........ 3rd place Wine A is Baron Philippe 1983 4. ........ 4th place Wine E is Baron Philippe 1982 5. ........ 5th place Wine F is Clos Rene 1983 --------------------------------------------------- 6. ........ 6th place Wine C is Marbuzet 1982 We now test whether the ranksums AS A WHOLE provide a significant ordering. The Friedman Chi-square value is 7.2449. The probability that this could happen by chance is 0.2031 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 Burt David John Burt 1.000 -0.543 -0.600 David -0.543 1.000 0.029 John -0.600 0.029 1.000 Orley -0.943 0.486 0.486 Ed -0.943 0.429 0.486 Bob 0.200 -0.486 0.657 Dick -0.943 0.429 0.771 Orley Ed Bob Burt -0.943 -0.943 0.200 David 0.486 0.429 -0.486 John 0.486 0.486 0.657 Orley 1.000 0.829 -0.314 Ed 0.829 1.000 -0.257 Bob -0.314 -0.257 1.000 Dick 0.829 0.886 0.086 Dick Burt -0.943 David 0.429 John 0.771 Orley 0.829 Ed 0.886 Bob 0.086 Dick 1.000 Pairwise correlations in descending order 0.886 Ed and Dick Significantly positive 0.829 Orley and Ed Not significant 0.829 Orley and Dick Not significant 0.771 John and Dick Not significant 0.657 John and Bob Not significant 0.486 John and Orley Not significant 0.486 David and Orley Not significant 0.486 John and Ed Not significant 0.429 David and Ed Not significant 0.429 David and Dick Not significant 0.200 Burt and Bob Not significant 0.086 Bob and Dick Not significant 0.029 David and John Not significant -0.257 Ed and Bob Not significant -0.314 Orley and Bob Not significant -0.486 David and Bob Not significant -0.543 Burt and David Not significant -0.600 Burt and John Not significant -0.943 Burt and Ed Significantly negative -0.943 Burt and Orley Significantly negative -0.943 Burt and Dick Significantly negative

COMMENT: Fact 1: tHE 83 Marbuzet and the 82 Clos Rene tied for first place. Fact 2: In the three possible com parisons, 83 beat 82; 2:1. B ut adding in the last flight, the Gloria, it comes out 2:2. The comparison of 82 and 83 in Ch. Gloria mirrors for Ed the comparisons of most major chateaux in the two vintages. The 82s are darker in color, slightly fuller in body, more sensuous and less astringent than the high quality 83s. Bob says that it is very close between 82 and 83. In the Gloria there is no dispute that the 82 is superior. Bob is shocked at the rating of the 82 Marbuzet to the extent of wondering whether it might have been a bad bottle.

Return to previous page

Report

WINETASTER ON 12/19/11 WITH 7 JUDGES AND 2 WINES BASED ON RANKS, IDENT=N Copyright (c) 1995-2011 Richard E. Quandt, V. 1.65

FLIGHT 2: Number of Judges = 7 Number of Wines = 2

Identification of the Wine: The judges' overall ranking:

Wine A is Ch.Gloria 1983 ........ 2nd place Wine B is Ch. Gloria 1982 ........ 1st place

The Judges's Rankings

Judge Wine -> A B Burt 2. 1. David 1. 2. John 2. 1. Orley 2. 1. Ed 2. 1. Bob 2. 1. Dick 2. 1.

Table of Votes Against Wine -> A B

Group Ranking -> 2 1 Votes Against -> 13 8

( 7 is the best possible, 14 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 = 0.5102

The probability that random chance could be responsible for this correlation is quite small, 0.0588. 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 Burt 1.0000 Ed 1.0000 John 1.0000 Orley 1.0000 Bob 1.0000 Dick 1.0000 David -1.0000

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 B is Ch. Gloria 1982 --------------------------------------------------- 2. ........ 2nd place Wine A is Ch.Gloria 1983 We now test whether the ranksums AS A WHOLE provide a significant ordering. The Friedman Chi-square value is 3.5714. The probability that this could happen by chance is 0.0588 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 1.00 for significance at the 0.05 level and must exceed 1.00 for significance at the 0.1 level Burt David John Burt 1.000 -1.000 1.000 David -1.000 1.000 -1.000 John 1.000 -1.000 1.000 Orley 1.000 -1.000 1.000 Ed 1.000 -1.000 1.000 Bob 1.000 -1.000 1.000 Dick 1.000 -1.000 1.000 Orley Ed Bob Burt 1.000 1.000 1.000 David -1.000 -1.000 -1.000 John 1.000 1.000 1.000 Orley 1.000 1.000 1.000 Ed 1.000 1.000 1.000 Bob 1.000 1.000 1.000 Dick 1.000 1.000 1.000 Dick Burt 1.000 David -1.000 John 1.000 Orley 1.000 Ed 1.000 Bob 1.000 Dick 1.000 Pairwise correlations in descending order 1.000 Bob and Dick Significantly positive 1.000 Burt and John Significantly positive 1.000 Burt and Orley Significantly positive 1.000 Burt and Ed Significantly positive 1.000 Burt and Bob Significantly positive 1.000 Burt and Dick Significantly positive 1.000 Orley and Bob Significantly positive 1.000 Orley and Dick Significantly positive 1.000 Ed and Bob Significantly positive 1.000 Ed and Dick Significantly positive 1.000 Orley and Ed Significantly positive 1.000 John and Orley Significantly positive 1.000 John and Ed Significantly positive 1.000 John and Bob Significantly positive 1.000 John and Dick Significantly positive -1.000 Burt and David Significantly negative -1.000 David and John Significantly negative -1.000 David and Orley Significantly negative -1.000 David and Ed Significantly negative -1.000 David and Bob Significantly negative -1.000 David and Dick Significantly negative

Return to previous page