WINE TASTER ON 03/09/98 WITH  6 JUDGES AND  4 WINES BASED ON RANKS      
                        Copyright (c) 1994-97 Richard E. Quandt                 
                                                                                
 
                                     FLIGHT 1:
                              Number of Judges =  6
                              Number of Wines  =  4
 
  Identification of the Wine:         The judges's overall ranking:
 
  Wine A is Gaja Barbaresco 1985            ........  2nd place
  Wine B is Hermitage La Chappelle 1985     ........  4th place
  Wine C is Stag's Leap Cab 1985            ........  1st place
  Wine D is Gruaud Larose 1985              ........  3rd place
 
                        The Judges's Rankings
 
  Judge       Wine ->   A   B   C   D                                           
                                                                                
  John L.               1.  4.  2.  3.                                          
  Leslie V.             1.  4.  3.  2.                                          
  Ed B.                 4.  3.  1.  2.                                          
  Orley A.              3.  1.  2.  4.                                          
  Frank V.              1.  4.  2.  3.                                          
  Richard Q.            3.  4.  1.  2.                                          
 
                        Table of Votes Against                                  
                                                                                
              Wine ->   A   B   C   D                                           
 
     Group Ranking ->   2   4   1   3
     Votes Against ->  13  20  11  16
 
     ( 6 is the best possible,  24 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 = 0.2556
 
 
 The probability that random chance could be responsible for this correlation
 is rather large, 0.2035. 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.
 
       Correlation Between the Ranks of
 Each Person With the Average Ranking of Others
 
  Name of Person      Correlation                                               
                                                                                
  John L.               0.8000                                                  
  Frank V.              0.8000                                                  
  Richard Q.            0.6325                                                  
  Leslie V.             0.4000                                                  
  Orley A.             -0.2000                                                  
  Ed B.                -0.2000                                                  
 
 
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 C is Stag's Leap Cab 1985   
  2.   ........  2nd place    Wine A is Gaja Barbaresco 1985   
  3.   ........  3rd place    Wine D is Gruaud Larose 1985     
---------------------------------------------------
  4.   ........  4th place    Wine B is Hermitage La Chap'85   
 
 
 Comments:
 
Comments by Orley Ashenfelter--Tasting held December 1997:                      
The idea of this tasting was interesting:  to pick outstanding wines from      
four regions of the world from what is considered a relatively mature vintage.  
As was evident from the tasting, this really provides a test of what the        
taster likes, rather than any measure of quality.  The wines were served        
from magnums, with the exception of the 1985 Chateau Gruaud Larose, where       
bottles were provided.  It was possible to tell the wines apart, despite all being
made in a very similar style, although the Hermitage revealed the character of  
somewhat unripe syrah that the wines from this area develop in unripe, but      
concentrated years.  This wine was, in the end, a love it or hate it deal.  As  
the results show, I loved it--but most didn't!  All of these were fine wines,   
but most could not help wonder why the Gaja wine sells for so much more than    
the others without any enormous difference in apparent quality.  All the wines  
were drinking nicely, but all could easily be kept in the cellar for another    
5 to to 10 years.  As the results indicate, there were, in fact, no significant 
differences among the wines from a statistical point of view among the top 3    
wines--a view that was shared by most of those present at the tasting.
Return to previous page