Académique Documents
Professionnel Documents
Culture Documents
Applied Marketing (Market Research Methods) Topic 12: Multidimensional scaling and conjoint analysis
Dr James Abdey
Overview Multidimensional scaling Formulate the problem Obtain input data Decide on the number of dimensions Conjoint analysis Formulate the problem Construct the stimuli Select a conjoint analysis procedure
Overview
Multidimensional scaling allows the perceptions and preferences of consumers to be clearly represented in a spatial map Conjoint analysis helps to determine the relative importance of attributes that consumers use in choosing products
Obtain input data Decide on the number of dimensions Conjoint analysis Formulate the problem Construct the stimuli Select a conjoint analysis procedure
Multidimensional scaling
Multidimensional scaling (MDS) is a class of procedures for representing perceptions and preferences of respondents spatially by means of a visual display Perceived or psychological relationships among stimuli are represented as geometric relationships among points in a multidimensional space These geometric representations are called spatial maps. The axes of the spatial map denote the dimensions respondents use to form perceptions and preferences for stimuli
Overview Multidimensional scaling Formulate the problem Obtain input data Decide on the number of dimensions Conjoint analysis Formulate the problem Construct the stimuli Select a conjoint analysis procedure
Specify the purpose for which the MDS results would be used Select the brands or other stimuli to be included in the analysis. The number of brands or stimuli selected normally varies between 10 and 20 The choice of the number and specic brands or stimuli to be included should be based on the statement of the market research problem, theory and the judgement of the researcher
Overview Multidimensional scaling Formulate the problem Obtain input data Decide on the number of dimensions Conjoint analysis Formulate the problem Construct the stimuli Select a conjoint analysis procedure
Overview Multidimensional scaling Formulate the problem Obtain input data Decide on the number of dimensions Conjoint analysis Formulate the problem Construct the stimuli Select a conjoint analysis procedure
2 2 . . . 2
3 3 . . . 3
4 4 . . . 4
5 5 . . . 5
6 6 . . . 6
The number of pairs to be evaluated is n(n 1)/2, where n is the number of stimuli
Overview Multidimensional scaling Formulate the problem Obtain input data Decide on the number of dimensions Conjoint analysis Formulate the problem Construct the stimuli Select a conjoint analysis procedure
If attribute ratings are obtained, a similarity measure (such as Euclidean distance) is derived for each pair of brands
Formulate the problem Obtain input data Decide on the number of dimensions Conjoint analysis Formulate the problem Construct the stimuli Select a conjoint analysis procedure
Overview Multidimensional scaling Formulate the problem Obtain input data Decide on the number of dimensions Conjoint analysis Formulate the problem Construct the stimuli Select a conjoint analysis procedure
Preference data
Preference data order the brands or stimuli in terms of respondents preference for some property A common way in which such data are obtained is through preference rankings Alternatively, respondents may be required to make paired comparisons and indicate which brand in a pair they prefer Another method is to obtain preference ratings for the various brands
Overview Multidimensional scaling Formulate the problem Obtain input data Decide on the number of dimensions Conjoint analysis Formulate the problem Construct the stimuli Select a conjoint analysis procedure
Overview Multidimensional scaling Formulate the problem Obtain input data Decide on the number of dimensions Conjoint analysis Formulate the problem Construct the stimuli Select a conjoint analysis procedure
A spatial map
Each point in the spatial map represents a brand, allowing similarities of perceptions or perceptions to be determined
Spatial map of brands
x 1.5
Overview Multidimensional scaling Formulate the problem Obtain input data Decide on the number of dimensions Conjoint analysis Formulate the problem Construct the stimuli
x 1.0 x x
Dimension 2
0.5
0.0
0.5
1.0
2.0
Dimension 1
Overview Multidimensional scaling Formulate the problem Obtain input data Decide on the number of dimensions Conjoint analysis Formulate the problem Construct the stimuli Select a conjoint analysis procedure
Overview Multidimensional scaling Formulate the problem Obtain input data Decide on the number of dimensions Conjoint analysis Formulate the problem Construct the stimuli Select a conjoint analysis procedure
Overview
In the pairwise approach, also called two-factor evaluations, the respondents evaluate two attributes at a time until all the possible pairs of attributes have been evaluated In the full-prole approach, also called multiple-factor evaluations, full or complete proles of brands are constructed for all the attributes. Typically, each prole is described on a separate index card
Multidimensional scaling Formulate the problem Obtain input data Decide on the number of dimensions Conjoint analysis Formulate the problem Construct the stimuli Select a conjoint analysis procedure
Assume three mobile phone attributes of interest: battery life, screen size and weight Assume each attribute has three different levels, coded 1, 2 and 3:
Battery life: Long, medium, short Screen size: Large, medium, small Weight: Heavy, medium, light
Overview Multidimensional scaling Formulate the problem Obtain input data Decide on the number of dimensions Conjoint analysis Formulate the problem Construct the stimuli Select a conjoint analysis procedure
Battery life 1 1 1 2 2 2 3 3 3
Preference rating 9 7 5 6 5 6 5 7 6
Decide on the number of dimensions Conjoint analysis Formulate the problem Construct the stimuli Select a conjoint analysis procedure
U (X ) =
i =1 j =1
ij xij
Decide on the number of dimensions Conjoint analysis Formulate the problem Construct the stimuli
where
U (X ) = overall utility of an alternative ij = the part-worth contribution or utility associated with the j -th level (j = 1, 2, . . . , ki ) of the i -th attribute (i = 1, 2, . . . , m) xij = 1 if the j -th level of the i -th attribute is present, 0 otherwise ki = number of levels of attribute i m = number of attributes
Overview Multidimensional scaling Formulate the problem Obtain input data Decide on the number of dimensions Conjoint analysis
The attributes importance is normalised to ascertain its importance relative to other attributes, Wi : Wi = Ii
m i = 1 Ii m
Formulate the problem Construct the stimuli Select a conjoint analysis procedure
such that
i =1
Wi = 1
The simplest estimation procedure, and one which is gaining in popularity, is dummy variable regression. If an attribute has ki levels, it is coded in terms of ki 1 dummy variables
Overview
U = b0 + b1 X1 + b2 X2 + b3 X3 + b4 X4 + b5 X5 + b6 X6 where
X1 , X2 = dummy variables representing battery life X3 , X4 = dummy variables representing screen size X5 , X6 = dummy variables representing weight
Multidimensional scaling Formulate the problem Obtain input data Decide on the number of dimensions Conjoint analysis Formulate the problem Construct the stimuli Select a conjoint analysis procedure
For battery life the attribute levels were coded as follows: X1 X2 Level 1 1 0 Level 2 0 1 Level 3 0 0
Weight X5 X6 1 0 0 1 0 0 0 1 0 0 1 0 0 0 1 0 0 1
Obtain input data Decide on the number of dimensions Conjoint analysis Formulate the problem Construct the stimuli Select a conjoint analysis procedure
Overview Multidimensional scaling Formulate the problem Obtain input data Decide on the number of dimensions Conjoint analysis Formulate the problem Construct the stimuli Select a conjoint analysis procedure
Given the dummy variable coding, in which level 3 is the base level, the coefcients may be related to the part-worths: 11 13 = b1 12 13 = b2
Overview Multidimensional scaling Formulate the problem Obtain input data Decide on the number of dimensions Conjoint analysis Formulate the problem Construct the stimuli Select a conjoint analysis procedure
Overview Multidimensional scaling Formulate the problem Obtain input data Decide on the number of dimensions Conjoint analysis Formulate the problem Construct the stimuli Select a conjoint analysis procedure
Overview Multidimensional scaling Formulate the problem Obtain input data Decide on the number of dimensions Conjoint analysis Formulate the problem Construct the stimuli Select a conjoint analysis procedure
Overview Multidimensional scaling Formulate the problem Obtain input data Decide on the number of dimensions Conjoint analysis Formulate the problem Construct the stimuli Select a conjoint analysis procedure