Applied Conjoint Analysis

Size: px

Start display at page:

Download "Applied Conjoint Analysis"

Della Blair
6 years ago
Views:

1 Vithala R. Rao Applied Conjoint Analysis < ) Springer

2 1 Problem Setting Introduction Marketing Decisions and Role of Consumer Choice A Framework for Understanding Consumer Choice Origins of Conjoint Analysis Some Terminology Principal Types of Conjoint Analysis Focus of this Book Industry Uses of Conjoint Analysis An Illustration of Conjoint Method Application of Choice-Based Conjoint Analysis Implementation of a Conjoint Study Another Illustration Features of Conjoint Analysis Taxonomy of Conjoint Methods Overview of Subsequent Chapters 27 Appendix: A Selection of Applications of Conjoint Analysis in Areas Other than Marketing 30 References 34 2 Theory and Design of Conjoint Studies (Ratings Based Methods) Introduction Designing a Conjoint Study Types of Attributes and Partworth Functions Selection of Attributes and Levels Stimulus Set Construction General Considerations Statistical Designs for Generating Full Profiles Full Factorial Designs Fractional Factorial Designs Orthogonal Main Effects Plans 47 ix

3 2.5.6 Incomplete Block Designs Random Sampling Generating "Acceptable" Designs Data Collection Methods Full Profile Approach Trade-off Matrix Method Paired Comparison Methods Self-explication Methods Adaptive Methods Hybrid Methods Stimulus Presentation Reliability and Validity Summary 65 Appendix 1: Illustration of a Ratings-Based Conjoint Questionnaire 68 Appendix 2: Measures of Efficiency of an Experimental Design 73 Appendix 3: Several Orthogonal Plans 74 References 77 3 Analysis and Utilization of Conjoint Data (Ratings Based Methods) Introduction Analysis Models for Ratings Data Notation Additive Utility Model Utility Model with Interactions Coding for Categorical Attributes Model Selection Level of Analysis Individual Level Analysis (Approaches IA and IB) Subgroup Level Analysis (Approaches IIA and IIB) Pooled Analysis for the Sample as a Whole (Approaches IIIA and IIIB) Some Comparisons Methods for Simulation l Illustration Estimating the Hybrid Conjoint Model Notation Models Comparison Individualized Hybrid Conjoint Models Notation Model for Adaptive Conjoint Analysis Polyhedral estimation Ill 3.8 Methods for Ranking and Categorical Response Data Summary : 113

4 xi Appendix 1: Computation of Trade-offs from Utility Functions in Attributes 113 Appendix 2: Specification of Utility Functions 115 Appendix 3: Hierarchical Bayesian Method for Ratings-Based Conjoint Analysis 117 Appendix 4: Linear Programming' Approach to Ranked Response Data 121 Appendix 5: A Method for Analyzing Categorical Response Data: Categorical Conjoint Analysis 122 References * Choice Based Conjoint Studies: Design and Analysis Introduction The Choice Process Choice Experiments for Conjoint Analysis: An Illustration Design of Choice Sets and Data Collection for Choice-Based Conjoint Studies Two Types of Designs Factors to Be Considered in Choice Set Design Examples of Designs Used in Some Past Studies Criteria for Evaluating Designs A Taxonomy of Choice Set Designs Strategies for Designing Choice Sets Methods Based on Linear Models Methods Based on Nonlinear Models for with Assumed Beta Values Bayesian Methods Based on Nonlinear Model for a Prior Distribution for Betas Other Methods Which Method to Use for Developing Designs? Incentive-Aligned Methods Partial Profile Choice Experiments Analysis Methods for Choice-Based Conjoint Data Multinomial Logit Model for Choice-Based Conjoint Data Modeling Utility Interpretation of Coefficients Data Structure Model Fit and Test Some Examples of MNL Analyses Some Alternatives to MNL for Stated Choice Data Bayesian Methods for Choice-Based Conjoint Analysis Which Conjoint Approach (Ratings-Based or Choice-Based)? Software for Design and Analysis Summary 172 Appendix 1: Illustration of Designing Choice Sets 173

5 Appendix 2:> Design Plans for Pre-specified Holistic Alternatives Using Fractional Factorial Method 174 Appendix 3: Illustration of Design Efficiency in Choice-Based Conjoint Designs 174 Appendix 4: Illustration of Managerial Efficiency 176 Appendix 5: Empirical Illustration of Availability Designs 178 Appendix 6: Weighted Least Squares Method 180 References Methods for a Large Number of Attributes Introduction Alternative Methods for Massive Number of Attributes Details of the Methods and Applications Methods of Category A: Profile Methods Methods of Category B: Attribute Simplification Methods Methods of Category C: Self Explicated Methods Category D: Methods Combining Several Approaches Category E: Upgrading Methods Category F: SVM Methods A Comparison of Methods Summary 221 References Applications for Product and Service Design and Product Line Decisions Introduction General Problem of Product and Product Portfolio Design An Unified Framework for Product Design Role of Choice Simulators Applications for New Product Design Application 1: Design of a Truck Application 2: Design of a SLR Camera Application 3: Design of a Course at a University Application 4: Design of Microfinance Products Application 4: Design of Dental Benefit Plans Application 5: Design of a Hotel Application 6: Design of Electronic Toll Collection System Optimal Design of a Pharmaceutical Product Applications for Product Line Decisions Application 1: Redesigning Product Lines at the Sunbeam Appliance Company Application 2: Redesign of Product Lines for an Herbicide Company Conclusions 255

6 xiii Appendix 1: A Mathematical Formulation of the Product Design and Positioning Problem 256 Appendix 2: Details on the SIMOPT Model 259 Appendix 3: Description of Algorithms for Product Line Design 263 References Applications for Product Positioning and Market Segmentation Introduction Methods of Forming Segments with Conjoint Results Applications for Product Positioning Application: Positioning of an Antidepressant Drug Market Segmentation Applications Application 1: Segments of Camera Buyers Application 2: Segments of Food Processor Buyers Application 3: Segments of Buyers of an Antifungal Medication Application 4: Segments from a Choice-Based Conjoint Study Comparison of Different Conjoint Segmentation Approaches Conclusion 290 References Applications for Pricing Decisions Introduction Conjoint Method for Determining Price Elasticities (Brand/Price Trade-off) Conjoint Method for Competitor Reaction Elasticities Method Based on Reservation Prices Measurement of Price Effects Using Ratings-Based Approach Using Choice-Based Approach More Applications Application 1: Bidding for a Contract Application 2: Pricing Digital Content Product Lines Application 3: Multipart Pricing Summary 313 Appendix 1: Technical Details for Estimating Self- and Cross-Price/Demand Relationships 313 Appendix 2: Estimation of Mean and Variance from Truncated Normal Distribution 315 References Applications to a Miscellany of Marketing Problems Introduction Competitive Strategy Decisions Distribution and Personal Selling Decisions 321

7 9.3.1 Store Location Decisions Setting Sales Quotas Choice of a Distribution Channel Web Page Design Legal Decisions Measuring Damage Due to Patent Infringement An Application to a Class Action Suit Resource Allocation Decisions Allocation of Push Marketing Mix Budget for a Brand Market Value of an Attribute Improvement (MVAI) Measurements for Marketing Strategies Measuring Brand Equity Customer Satisfaction Summary References Recent Developments and Future Outlook Introduction Experimental Designs for Mixture and Mixture-Amount Conjoint Approaches Barter Conjoint Method Conjoint Poker Best-Worst Scaling Peer Influence Measurement Incorporating Non-compensatory Choice Processes Combining Preference and Choice Data Applications Self-designed Products Bundle Choice Models Future Outlook Summary 360 References Beyond Conjoint Analysis: Advances in Preference Measurement Introduction: Beyond Conjoint Analysis Problem Helping Companies Helping Consumers Helping Policy Makers and Health Care Professionals Helping Academic Researchers Design and Data Collection Optimal Experimental Design: Beyond A-Efficiency and D-Efficiency New Forms of Interactions 368

8 xv Dealing with a Large Number of Attributes and Products Combining Multiple Sources of Data Model Specification, Estimation, and Action Taking Social Interactions into Account Meta-attributes More Flexible Utility Functions Incorporating Behavioral Effects Modeling Learning, Dynamics and Preference Formation ' Recent Tools for Estimation From Model to Action In Conclusion..."Every Generation Needs a New Revolution" References 377 Index 383

What Is Conjoint Analysis? DSC 410/510 Multivariate Statistical Methods. How Is Conjoint Analysis Done? Empirical Example

What Is Conjoint Analysis? DSC 410/510 Multivariate Statistical Methods Conjoint Analysis 1 A technique for understanding how respondents develop preferences for products or services Also known as trade-off