conjoint analysis
DESCRIPTION
conjoint AnalysisTRANSCRIPT
Conjoint Analysis
Formulate the ProblemSneaker Attribute and LevelsAttribute Level No DescriptionSole 1 Rubber
2 Polyurethane3 Plastic
Upper 1 Leather2 Canvas3 Nylon
Price 1 30 USD2 60 USD3 90 USD
CONSTRUCT THE STIMULI- PAIRWISE APPROACH
UPPER
SOLE
RUBBER POLYURETHANE
PLASTIC
LEATHER
CANVAS
NYLON
PRICE
SOLE
RUBBER POLYURETHANE
PLASTIC
30 USD
60 USD
90 USD
UPPER
PRICE
30 USD 60 USD 90 USD
LEATHER
CANVAS
NYLON
Full-Profile Approach to Collecting Conjoint Analysis
EXAMPLE OF A SNEAKER PRODUCT PROFILE
SOLE MADE OF RUBBERUPPER MADE OF NYLONPRICE 30 USD
CodingX one X two
Level 1 1 0Level 2 0 1Level 3 0 0
Sneaker Profiles and their Ratings
ATTRIBUTE LEVELSPROFILE NO.
SOLE UPPER PRICE PREFERENCE RATINGS
1 1 1 1 92 1 2 2 73 1 3 3 54 2 1 2 65 2 2 3 56 2 3 1 67 3 1 3 58 3 2 1 79 3 3 2 6
Conjoint Analysis Model
Where
For us
Sneaker data coded for Dummy Variable Regression
Preference ATTRIBUTES Ratings SOLE UPPER PRICE Y 9 1 0 1 0 1 0 7 1 0 0 1 0 1 5 1 0 0 0 0 0 6 0 1 1 0 0 1 5 0 1 0 1 0 0 6 0 1 0 0 1 0 5 0 0 1 0 0 0 7 0 0 0 1 1 0 6 0 0 0 0 0 1
The model estimated may be presented as:
where
Estimation of the parameters
1
0.667 2.333
Context of dummy variables•Given the dummy variable coding in which
level 3 is the base level, the coefficients may be reduced to part worths
•Each dummy variable coefficient represents the difference in the part worth for that level minus the part worth for the base level. For sole we have the following:
•To solve for the part-worths, an additional constraint is necessary. The part-worths are estimated on an interval scale, so the origin is arbitrary. Therefore the additional constraint that is imposed is of the form:
α11 + α12 + α13 = 0
These equations for the first attribute, Sole, are:
= -0.333
Soα11 = 0.778
α12 = −0.556
α13 = −0.222
Similarly we can solve for other attributes
RESULTSLEVEL
ATTRIBUTE No. Description Utility Importance
Sole 3 Rubber 0.7782 Polyurethane -0.5561 Plastic -0.222 0.286
Upper 3 Leather 0.4452 Canvas 0.1111 Nylon -0.556 0.214
Price 3 30 USD 1.1112 60 USD 0.1111 90 USD -1.222 0.500
Part-worthsDifference between
the highest and lowest Highest 0.778 1.334Lowest -0.556
-0.222Highest 0.445
0.111lowest -0.556 1.001Highest 1.111
0.111Lowest -1.222 2.333
sum of part-worths 4.668
Relative Importance of attributesSOLE 1.334/4.668=0.286
UPPER 1.001/4.668=0.214
PRICE 2.333/4.668=0.5
Preference for Attribute Levels
Rubber Polyurethane Plastic
-0.8-0.6-0.4-0.2
00.20.40.60.8
1
Sole
Sole
Preference for Attribute Levels
Leather Canvas Nylon
-0.8
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
Upper
Upper
Preference for Attribute Levels
30 USD 60 USD 90 USD
-1.5
-1
-0.5
0
0.5
1
1.5
Price
Price
Assessing Reliability and Validity•Look at the R square•Test-retest reliability•If an aggregate analysis has been
conducted, the estimation sample can be split and conjoint analysis conducted on each sub-sample. The results can be compared across sub-samples to assess the stability of conjoint analysis solution.
Sneaker data coded for Dummy Variable Regression
Preference ATTRIBUTES Ratings SOLE UPPER PRICE Y R Pl 𝑋3 𝑋4 𝑋5 𝑋6 9 1 0 1 0 1 0 7 1 0 0 1 0 1 5 1 0 0 0 0 0 6 0 0 1 0 0 1 5 0 0 0 1 0 0 6 0 0 0 0 1 0 5 0 1 1 0 0 0 7 0 1 0 1 1 0 6 0 1 0 0 0 1
Estimation of the parameters
𝑏0 = 3.889 𝑟= 1.333 𝑝𝑙 = 0.333 𝑏3 = 1 𝑏4 = 0.667 𝑏5 =2.333 𝑏6 = 1.333
These equations for the first attribute, Sole, are:
α11 − α12 = 𝑏1 = 1 α13 − α12 = 𝑏2= 0.333
Sneaker data coded for Dummy Variable Regression
Preference ATTRIBUTES Ratings SOLE UPPER PRICE Y P Pl 𝑋3 𝑋4 𝑋5 𝑋6 9 0 0 1 0 1 0 7 0 0 0 1 0 1 5 0 0 0 0 0 0 6 1 0 1 0 0 1 5 1 0 0 1 0 0 6 1 0 0 0 1 0 5 0 1 1 0 0 0 7 0 1 0 1 1 0 6 0 1 0 0 0 1
Estimation of the parameters
𝑏0 = 3.889 𝑃= −1.333 𝑝𝑙 = −1 𝑏3 = 1 𝑏4 = 0.667 𝑏5 =2.333 𝑏6 = 1.333
These equations for the first attribute, Sole, are:
α12 − α11 = 𝑏1 = −1.333 α13 − α11 = 𝑏2= -1.0