Question: (50 Points)

ReMarQ Inc. recently conducted a study of the ground caffeinated coffee market because several of its clients operate in this market. The study was undertaken with several objectives in mind, including the identification of benefits that consumers seek and the comparison of consumer beliefs regarding benefits of several brands offered in the market.

A questionnaire was designed to accomplish these objectives. This questionnaire was mailed to 400 individuals previously identified as consumers of ground caffeinated coffee (personally drinking at least one cup per day). Of those mailed out, 328 were returned; 300 of these were judged to be usable responses. Please see the data file called Coffee.sav. Var1 to Var19 refer to consumer ratings for various attributes of coffee by a sample of 300 consumers. Specific question asked: On a scale of 1 to 10 where 1 refers to NOT IMPORTANT and 10 refers to EXTREMELY IMPORTANT, please indicate how important it is to you that a ground coffee have each of the following characteristics:

Var1	Rich Taste
Var2	Always Fresh
Var3	Gets Day Off to Good Start
Var4	Full-Bodied Taste
Var5	Rich Aroma in the Cup
Var6	Good Value for the Money
Var7	Best Coffee in the Morning
Var8	Rich Aroma in the Can/Bag
Var9	Smooth Taste
Var10	Highest Quality Coffee
Var11	Premium Brand
Var12	Not Bitter
Var13	Coffee That Brightens Day Most
Var14	Has No Aftertaste
Var15	Economy Brand
Var16	Rich Aroma While Brewing
Var17	Best Ground Coffee Available
Var18	Enjoy Drinking with Meal
Var19	Costs Less Than Other Brands

1. Factor analyze the 19 attribute preference data. Save the factor scores for each consumer. Use six factors. Copy and paste the Rotated Component Matrix here (10 points).

2. Highlight the highest factor loadings of each variable on the factors and name the factors that you generated (15 points).

3. Now conduct a cluster analysis to segment the population based on the saved factor scores in (a). For this just use a K-mean clustering with K=3 (3 segments). Save the cluster membership for each household. How many individuals in each cluster? To what extent can you interpret the clusters? Specifically, can you name the clusters based on their average factor scores? Copy and paste your SPSS results here (25 points).