You are on page 1of 17

An Example of Attribute Based MDS Using Discriminant Analysis

Problem : A chocolate company wants to draw a perceptual map using an attribute based procedure, of its consumers perceptions regarding its own brand and two competing brands. Assume that it is Nestle against Cadburys and Amul, for example.

DATA Data was collected from 15 respondents (5 of each brand), on five attributes, namely Price, Quality, Availability, Packaging and Taste. The variables are measured using different scales, but a higher value indicates a favourable rating in each variables measurement.

Input Data (High = Better)


BRAND 1 1 1 1 1 2 2 2 2 2 3 3 3 3 3 PRICE 12 11 10 13 12 10 11 15 13 12 10 12 13 11 11 QUALITY 34 35 36 22 23 14 17 23 14 15 22 24 28 17 18 AVAILABILITY 500 234 250 345 432 234 231 45 35 25 75 80 90 96 59 PACKAGING 5 4 4 5 3 2 3 4 3 2 4 4 5 2 2 TASTE 18 15 14 12 13 15 11 10 12 10 8 7 10 12 6

Means Output by Brand


G u Sais ic ro p t t t s V lidN(lisw e a t is ) U w ig t d n e he Wig t d e he 5 50 0 .0 5 50 0 .0 5 50 0 .0 5 50 0 .0 5 50 0 .0 5 50 0 .0 5 50 0 .0 5 50 0 .0 5 50 0 .0 5 50 0 .0 5 50 0 .0 5 50 0 .0 5 50 0 .0 5 50 0 .0 5 50 0 .0 1 5 1. 0 50 0 1 5 1. 0 50 0 1 5 1. 0 50 0 1 5 1. 0 50 0 1 5 1. 0 50 0 BAD RN 1 P IC R E Q A IT UL Y PCA AKG TSE AT A ABT VL LY P IC R E Q A IT UL Y PCA AKG TSE AT A ABT VL LY P IC R E Q A IT UL Y PCA AKG TSE AT A ABT VL LY P IC R E Q A IT UL Y PCA AKG TSE AT A ABT VL LY Ma en 1. 00 16 0 3. 00 00 0 42 0 . 00 1. 00 44 0 3 22 0 5. 00 1. 00 22 0 1. 00 66 0 28 0 . 00 1. 00 16 0 1 40 0 1. 00 1. 00 14 0 2. 00 18 0 34 0 . 00 86 0 . 00 8. 00 00 0 1. 33 17 3 2. 00 28 0 34 6 . 67 1. 33 15 3 1 20 6 8.67 Sd D v t n t . e iaio 11 0 8 . 41 68 2 2 . 90 . 36 866 23 2 7 . 01 1 47 1 9 1. 64 19 3 4 . 25 37 1 3 . 85 . 36 866 20 3 4 .76 1 84 1 5 0.12 11 0 8 . 41 44 4 4 . 94 13 1 4 . 46 24 8 2 . 03 1. 32 43 5 7 13 7 1 .80 74 5 2 .82 11 5 6 . 24 32 6 8 . 23 1 13 2 6 5.06

Tt l oa

Univariate F tests
Te s ts o f E q u ality o f G r o u p M e an s W ilk s ' La m b d a P R IC E .9 3 6 Q U A LIT Y .4 1 8 PACKAG .7 2 2 TAS TE .4 2 3 A V A LB LT Y .3 1 4 F .4 1 3 8 .3 4 9 2 .3 1 3 8 .1 9 5 1 3 .1 3 1 d f1 2 2 2 2 2 d f2 12 12 12 12 12 S ig . .6 7 1 .0 0 5 .1 4 1 .0 0 6 .0 0 1

Discrim Functions
E ig e n v alu e s C a n o n ic a l Fu n ctio n E ig e n v a lu e % o f V a ria n c eC u m u la tiv e % C o rre la tio n a 1 4 .7 4 9 8 1 .4 8 1 .4 .9 0 9 a 2 1 .0 8 3 1 8 .6 1 0 0 .0 .7 2 1 a . Firs t 2 c a n o n ica l d is c rim in a n t fu n c tio n s w e re u s e d in th e a n a ly s is .

Significance Test

W ilk s ' L am b d a W ilk s ' T e s t o f F u n c tio n (La m b d a C h i-s q u a r e s) 1 th r o u g h 2 .0 8 4 2 4 .8 2 7 2 .4 8 0 7 .3 3 6 df 10 4 S ig . .0 0 6 .1 1 9

Standardised Coeffs.
Standardized Canonical Dis criminant Function Coefficients Functio n PRICE QUALITY PACKAG TASTE AVALBLTY 1 .207 .988 -.398 -.136 .999 2 .7 01 -.454 -.293 .986 -.122

Var. Loadings on Functions


Structure Matrix Function AVALBLTY QUALITY PACKAG TASTE PRICE 1 .664 * .517 * .268 * .431 -.044 2 .294 -.336 -.203 .668 * .235 *

Pooled within-groups correlations between discriminating variables and standardized canonical discriminant functions Variables ordered by absolute size of correlation within function. *. Largest absolute correlation between each variable and any discriminant function

Centroids of Brands on Functions


Functions at Group Centroids Function BRAND 1 2 3 1 2.745 -1.596 -1.149 2 .123 1.073 -1.196

Unstandardized canonical discriminant functions evaluated at group means

Plot of Brands on 2 Dimensions


C n ica D a on l iscrim a t F n n in n u ctio s
2 2 1 1 0

-1

B AD RN
G u C n id ro p e tro s

Function 2

-2

3 2

-3 -4 -2 0 2 4 6

F unction 1

Putting Variables/Attribute Vectors on the Above Map Vectors which represent the original attributes can be located on the above map. If there are more than 3 brands, we may get more than 2 dimensions, and may have to draw more than one plot of the above type. To plot the attributes on the map above, we can use the standardized coefficients of the original variables in the discriminant function. For example, for Taste, the standardized coefficients are -.136 and .986 on Dimensions 1 and 2 respectively. So we can locate this point (-.136, . 986) on the map, and draw an arrow from the origin to that point.

This will be labeled the Taste vector, and similarly, all other vectors can be located, one for each of the five attributes - Price, Quality, Availability, Packaging and Taste. The length of the arrow represents its effect in discriminating on each dimension. Longer arrows pointing more closely towards a given group centroid represent variables most strongly associated with the group (or Brand, in this case). Vectors pointing in the opposite direction from a given group centroid represent lower association with a group.

Variables with longer vectors in a given dimension, and those closest to a given axis (dimension represented by the discriminant function) are contributing more to the interpretation of that dimension. Looking at all variables that contribute to a given axis (dimension), we can label the dimension as a combination of those variables. In this case, the interpretation in terms of the variables and their correlation to dimensions 1 and 2 can be found from the graph which follows (on next page).

Plot of Brands and Attribute Vectors


1.5 Cadbury Taste 1 Price 0.5 Dimension 2 Nestle 0 -0.5 Quality -1 Amul -1.5 Dimension 1 1 Availability 2 3

0 -2 -1 Packaging

As seen from the graph, Nestle, Cadbury and Amul, the three brands have their unique positions on the map. In addition, on the same map, we now have plotted values of the attributes on the same 2 dimensions (each discriminant function represents a dimension). As we can see, Dimension 1 seems to be a combination of Availability (closest to the x-axis) and Quality. This is also evident from the standardized discriminant coefficients for Availability (.999) and Quality (.988) on Dimension 1, from the earlier output table.

Dimension 2 seems to comprise of Taste and Price, the two vectors (arrows) that are closest to the vertical axis. This is also evident from the standardized coefficients, of .986 and .701 respectively, for Taste and Price on Dimension 2, from the earlier output table. Packaging is not useful in defining any of the two dimensions, as its arrow is not close to either of the two dimensions. Brands and their Association with Attributes/Dimensions Nestle seems to be stronger on Dimension 1 (Availability and Quality), and Cadbury on Dimension 2 (Taste and to a lesser extent, Price). Amul scores low on both dimensions compared to its competitors.

You might also like