WINETASTER ON 03/27/00 WITH  8 JUDGES AND  7 WINES BASED ON RANKS,  IDENT=N   
                       Copyright (c) 1995-2000 Richard E. Quandt                  
                                                                                






                                     FLIGHT 1:
                              Number of Judges =  8
                              Number of Wines  =  7


  Identification of the Wine:                   The judges' overall ranking:


  Wine A is Heitz Cellar, Cabernet 96                     ........  7th place
  Wine B is Swanson, Cabernet 97                          ........  6th place
  Wine C is Frogs Leap, Cabernet 97                       tied for  4th place
  Wine D is Gallo Sonoma, Cabernet 96                     tied for  4th place
  Wine E is L Escrime, Cabernet 97                        ........  2nd place
  Wine F is Sterling, Cabernet 97                         ........  1st place
  Wine G is Glen Ellen, Cabernet 97                       ........  3rd place


                        The Judges's Rankings


  Judge       Wine ->   A   B   C   D   E   F   G                               
                                                                                
  Nahomi                7.  3.  6.  5.  1.  4.  2.                              
  Raul                  6.  7.  4.  3.  5.  1.  2.                              
  Lisa                  6.  2.  3.  7.  5.  1.  4.                              
  Gaby                  5.  7.  4.  1.  3.  6.  2.                              
  Stefan                7.  4.  5.  1.  3.  2.  6.                              
  Hans                  7.  5.  2.  4.  1.  3.  6.                              
  Anita                 6.  7.  5.  4.  3.  2.  1.                              
  Karl                  7.  3.  1.  5.  4.  2.  6.                              


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


     Group Ranking ->   7   6   4   4   2   1   3
     Votes Against ->  51  38  30  30  25  21  29


     ( 8 is the best possible,  56 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 = 0.3259




 The probability that random chance could be responsible for this correlation
 is quite small, 0.0158. 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                                             
                                                                                
  Anita                 0.5714                                                  
  Raul                  0.5637                                                  
  Stefan                0.4286                                                  
  Hans                  0.3929                                                  
  Karl                  0.2857                                                  
  Nahomi                0.2342                                                  
  Lisa                  0.1081                                                  
  Gaby                  0.0714                                                  
                                                                                
                                                                                




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 F is Sterling, Cabernet 97             
---------------------------------------------------
  2.   ........  2nd place    Wine E is L Escrime, Cabernet 97            
  3.   ........  3rd place    Wine G is Glen Ellen, Cabernet 97           
  4.   tied for  4th place    Wine D is Gallo Sonoma, Cabernet 96         
  5.   tied for  4th place    Wine C is Frogs Leap, Cabernet 97           
  6.   ........  6th place    Wine B is Swanson, Cabernet 97              
---------------------------------------------------
  7.   ........  7th place    Wine A is Heitz Cellar, Cabernet 96         

   We now test whether the ranksums AS A WHOLE provide a significant ordering.
 The Friedman Chi-square value is 15.6429. The probability that this could
 happen by chance is 0.0158

   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  0.79 for significance at the 0.05   
level and must exceed  0.71 for significance at the 0.1 level                   
                                                                                
                                                                                
                  Nahomi           Raul             Lisa                        
                                                                                
Nahomi            1.000            0.107            0.214                       
Raul              0.107            1.000            0.179                       
Lisa              0.214            0.179            1.000                       
Gaby              0.143            0.393           -0.714                       
Stefan            0.250            0.357            0.036                       
Hans              0.321            0.179            0.214                       
Anita             0.500            0.857            0.071                       
Karl              0.036            0.143            0.714                       
                                                                                
                  Gaby             Stefan           Hans                        
                                                                                
Nahomi            0.143            0.250            0.321                       
Raul              0.393            0.357            0.179                       
Lisa             -0.714            0.036            0.214                       
Gaby              1.000            0.179            0.107                       
Stefan            0.179            1.000            0.571                       
Hans              0.107            0.571            1.000                       
Anita             0.500            0.214            0.214                       
Karl             -0.393            0.393            0.714                       
                                                                                
                  Anita            Karl                                         
                                                                                
Nahomi            0.500            0.036                                        
Raul              0.857            0.143                                        
Lisa              0.071            0.714                                        
Gaby              0.500           -0.393                                        
Stefan            0.214            0.393                                        
Hans              0.214            0.714                                        
Anita             1.000           -0.071                                        
Karl             -0.071            1.000                                        
                                                                                
                     Pairwise correlations in descending order                  
                                                                                
    0.857   Raul             and Anita              Significantly positive      
    0.714   Lisa             and Karl               Significantly positive      
    0.714   Hans             and Karl               Significantly positive      
    0.571   Stefan           and Hans               Not significant             
    0.500   Gaby             and Anita              Not significant             
    0.500   Nahomi           and Anita              Not significant             
    0.393   Raul             and Gaby               Not significant             
    0.393   Stefan           and Karl               Not significant             
    0.357   Raul             and Stefan             Not significant             
    0.321   Nahomi           and Hans               Not significant             
    0.250   Nahomi           and Stefan             Not significant             
    0.214   Stefan           and Anita              Not significant             
    0.214   Lisa             and Hans               Not significant             
    0.214   Nahomi           and Lisa               Not significant             
    0.214   Hans             and Anita              Not significant             
    0.179   Gaby             and Stefan             Not significant             
    0.179   Raul             and Lisa               Not significant             
    0.179   Raul             and Hans               Not significant             
    0.143   Nahomi           and Gaby               Not significant             
    0.143   Raul             and Karl               Not significant             
    0.107   Gaby             and Hans               Not significant             
    0.107   Nahomi           and Raul               Not significant             
    0.071   Lisa             and Anita              Not significant             
    0.036   Lisa             and Stefan             Not significant             
    0.036   Nahomi           and Karl               Not significant             
   -0.071   Anita            and Karl               Not significant             
   -0.393   Gaby             and Karl               Not significant             
   -0.714   Lisa             and Gaby               Significantly negative      










 COMMENT:

  Like the tasting 3 months ago this was a tasting of wines of the lower    
  and medium price range. Like the tasting before the judges were mostly
  inexpierenced at tasting wine.
  The cheapest wines were the Glen Ellen (4$) and 
  Gallo Sonoma (7$). Sterling Vineyards Cabernet was 14$, L Escrime         
  20$, Heitz Cellar 23$, Frogs Leap 27$ and Swanson 31$. Everybody agreed   
  that the poor quality of Heitz was astounding. However, this was almost the only   
  consensus. Except Heitz and Swanson every wine was ranked at least once on the  
  first place. It is remarkable that one of the cheapest wines turned out  
  as clear winner (Sterling, 14 $).                                         
                                                                            
  Accordingly, a simple regression between points (dependent variable) and  
  bottle price (independent varible) indicates almost no relationship:         
               points = 0.4*price + 24.6                                                 
               The R-square is 0.17.

Return to previous page