1. Conjoint Analysis, Related Modeling, and Applications
Authors: John R. Hauser & Vithala R. Rao
Presenter: jellylover

2. 1 2 3 4 5 Contents Introduction Origins challenges Summary03 08 09 18 20
Page
Introduction
Origins
Basic elements
& their evolution
challenges
Summary
& conclusions 1 2 3 4 5

3. Origins: papers
R. Duncan Luce and John W. Tukey, “Simultaneous Conjoint Measurement: A New Type of Fundamental Measurement,” Journal of Mathematical Psychology, February 1964, p.I.
The first marketing-oriented paper on conjoint measurement was by Paul E. Green and Vithala R. Rao, “Conjoint Measurement for quantifying judgmental Data,” Journal of Marketing Research, August 1971, p. 355.
Accordingly, it now seems useful to adopt the name, “conjoint analysis”to cover models and techniques that emphasize the transformation of subjective responses into estimated parameters.
Green, Paul E. and V. Srinivasan (1978), “Conjoint Analysis in Consumer Research: Issues and Outlook,” Journal of Consumer Research, 5, 2, (September), 1 2 3

4. Introduction
Example:
Processes of Conjoint Analysis

5. Introduction 1 Price? Brand name? Only available online?
Consider an example…
Introduction 1
Price?
Brand name?
Only available online?

6. Respondents' evaluation (rank number)
Experimental Design for evaluation of a Xiao Mi
brand name
retail price (yuan)
only available online
Respondents' evaluation (rank number) 1 mi
1699 Y 3 2
Motorola
HTC 4
1999 12 5 11 6 10 7
2499 14 8 15 9 13 N 16 17 18 2

7. 3
We assumed that the overall preference was an additive sum of the “partworths” of the features,
represented each feature by a series of dummy variables,
used monotonic regression to estimate the contribution of each feature to overall preference.
For more information about MONOTONIC REGRESSION, please refer to:
Green, Paul E. and V. Srinivasan (1978), “Conjoint Analysis in Consumer Research: Issues and Outlook,” Journal of Consumer Research, 5, 2, (September),

8. 4 Results of computer analysis of experimental data Utility Utility
Brand name
Retail Price
Utility
Only Available
Online
Retail Price
Brand name
Only Available Online
Utility Ranges

9. For More Information, please refer to the paper blew:
Green, Paul E. and Jerry Wind (1975), “New Way to Measure Consumers’ Judgments,” Harvard Business Review, (July-August),

10. CONTENTS Basic elements & their evolution
Part 1 Decomposing the Product
and Service
Part 2 Representation Stimuli
Part 3 Have Mercy on the
Respondent
Part 4 Formats of Data Collection
Part 5 Estimation Methods
Basic elements
& their evolution

11. Part 1 Decomposing the Product and Service

12. Elements into which the product is decomposed
Consideration 1
Elements into which the product is decomposed
As complete as feasible
Understandable to respondents
Useful to product-development team
As separable as feasible
Selecting Criteria

13. Consideration 2
The function by which the elemental decomposition is mapped onto overall preference
The conjoint function can be represented by an additive (or multiplicative) decomposition
Satisfied
Preferential
independence
The conjoint function is more difficult to estimate and interactions are necessary
N.S.
Preference is defined over risky features

14. Part 2 Representation of Stimuli

15. Representation of Stimuli Evolution
Combinations of physical models, photographs and verbal descriptions (Wind, et al., 1989)
Virtual prototypes to web-based respondents (Dahan and Srinivasan 2000)
Verbal and pictorial descriptions on cards
Videotapes (Vavra, Green, et al., 1999)

16. Part 3 Have Mercy on the Respondents

17. Realize self-explicate tasks Two stages trade off analysis
Hierarchical
integration ……
Realize
1. It is difficult for customers to rank more than a dozen profiles
2. The accuracy degrades as the number of questions increases

18. Part 4 Formats of Data Collection

19. Formats of Data
Collection
Full Profile
Evaluations

20. Formats of Data
Collection
Partial Profile
Evaluations

21. Formats of Data
Collection
Stated
Preference

22. Formats of Data
Collection
Self-explicated
Methods

23. Configurators Example： http://nb.zol.com.cn/ Formats of Data
Collection
Configurators
Example：

24. Part 5 Estimation Methods

25. Regression-based Methods
Estimation
Methods
Regression-based Methods
Simplicity and Availability of software with which to perform estimations
Require more observations**

26. Random-Utility Models
Estimation
Methods
Random-Utility Models
Hierarchical Bayes Estimation
Direct Computation Based on Self-Explicated Importance
Estimation Based on New Optimization Methods

27. Challenges Pragmatic Issues
Analysis of tradeoffs between complexity of analysis, cost and difficulty of data collection, and managerial application.
Analysis methods VS New forms of data
Meta-analyses of the varied applications under a variety of managerial problems
External validity & comparative strengths of the various methods
Methods to handle large numbers of features

28. Challenges Conceptual Issues
Relative merits of the merits of the various data-collection methods
Price
Competitors→ Equilibrium, non-equilibrium models
Diffusion
Scenario problem

29. Methodological Issues
Challenges
Methodological Issues
Anticipate further exploration of hybrid methods that combine data from multiple data sources
Use of information from other respondents to inform each respondent’s estimates
Improved methods for adaptive data collection
Cost of the study VS the value of the study

30. You can find a copy of the papers listed in the slides by sending an Email to lincoln@pku.edu.cn

31. Thanks for your time

32. 更多精美ppt下载请点击：ppt宝藏-www.pptbz.com PPT宝藏致力于优秀的ppt分享