Skip to main content

2000 | Buch | 2. Auflage

Market Segmentation

Conceptual and Methodological Foundations

verfasst von: Michel Wedel, Wagner A. Kamakura

Verlag: Springer US

Buchreihe : International Series in Quantitative Marketing

insite
SUCHEN

Über dieses Buch

Modern marketing techniques in industrialized countries cannot be implemented without segmentation of the potential market. Goods are no longer produced and sold without a significant consideration of customer needs combined with a recognition that these needs are heterogeneous. Since first emerging in the late 1950s, the concept of segmentation has been one of the most researched topics in the marketing literature. Segmentation has become a central topic to both the theory and practice of marketing, particularly in the recent development of finite mixture models to better identify market segments.
This second edition of Market Segmentation updates and extends the integrated examination of segmentation theory and methodology begun in the first edition. A chapter on mixture model analysis of paired comparison data has been added, together with a new chapter on the pros and cons of the mixture model. The book starts with a framework for considering the various bases and methods available for conducting segmentation studies. The second section contains a more detailed discussion of the methodology for market segmentation, from traditional clustering algorithms to more recent developments in finite mixtures and latent class models. Three types of finite mixture models are discussed in this second section: simple mixtures, mixtures of regressions and mixtures of unfolding models. The third main section is devoted to special topics in market segmentation such as joint segmentation, segmentation using tailored interviewing and segmentation with structural equation models. The fourth part covers four major approaches to applied market segmentation: geo-demographic, lifestyle, response-based, and conjoint analysis. The final concluding section discusses directions for further research.

Inhaltsverzeichnis

Frontmatter

Introduction

Frontmatter
Chapter 1. The Historical Development of the Market Segmentation Concept
Abstract
After briefly introducing the concept and history of market segmentation, we review the criteria for effective segmentation and introduce the topics to be discussed in this book.
Michel Wedel, Wagner A. Kamakura
Chapter 2. Segmentation Bases
Abstract
This chapter classifies segmentation bases into four categories. A literature review of bases is provided and the available bases are evaluated according to the six criteria for effective segmentation described in chapter 1. The discussion provides an introduction to parts 3 and 4 of the book, where some special topics and new application areas are examined.
Michel Wedel, Wagner A. Kamakura
Chapter 3. Segmentation Methods
Abstract
This chapter classifies current segmentation methods and techniques into four categories. It provides an overview and evaluation of the methods in each of the four categories. Also, it provides an introduction to part 2 of this book, which describes cluster analysis, mixture, mixture regression and mixture scaling methods, currently the most powerful approaches to market segmentation.
Michel Wedel, Wagner A. Kamakura
Chapter 4. Tools for Market Segmentation
Abstract
In this chapter we reiterate the main conclusions drawn from the two preceding chapters about segmentation bases and methods, and introduce further developments, to be presented in the subsequent chapters.
Michel Wedel, Wagner A. Kamakura

Segmentation Methodology

Frontmatter
Chapter 5. Clustering Methods
Abstract
In this chapter, we describe the hierarchical and nonhierarchical descriptive clustering methods traditionally used for market segmentation. We also discuss fuzzy clustering and clusterwise regression methods. The latter two sections also provide an introduction to chapters 6 and 7.
Michel Wedel, Wagner A. Kamakura
Chapter 6. Mixture Models
Abstract
We now discuss the main statistical approach to clustering and segmentation: mixture models. We examine a general form of such mixtures and describe the EM estimation algorithm. We consider problems of the model and its estimation that are related to identification, local optima, standard errors and the number of segments. Those issues provide the basic foundation for the chapters that follow.
Michel Wedel, Wagner A. Kamakura
Chapter 7. Mixture Regression Models
Abstract
We review mixture models that relate a dependent variable to a set of exogenous or explanatory variables. Also, we describe a generalized linear regression mixture model that encompasses previously developed models as special cases. The model allows for a probabilistic classification of observations into segments and simultaneous estimation of a generalized linear regression model within each segment. Previous applications of the approach to market segmentation are extensively reviewed.
Michel Wedel, Wagner A. Kamakura
Chapter 8. Mixture Unfolding Models
Abstract
We describe a general mixture unfolding approach that allows simultaneously for a probabilistic classification of observations into segments (similar to the GLIMMIX models described in the preceding chapter) and the estimation of an internal unfolding model within each segment. This multidimensional scaling (MDS) based methodology is formulated in the framework of the exponential family of distributions, whereby a wide range of data types can be analyzed. Possible re-parameterizations of stimulus coordinates by stimulus characteristics, as well as of probabilities of segment membership by subject background variables, are permitted. We also review previous applications of the approach to market segmentation problems.
Michel Wedel, Wagner A. Kamakura
Chapter 9. Profiling Segments
Abstract
In this chapter we further expand on the mixture models provided in the preceding three chapters. We discuss procedures that simultaneously profile segments. Those procedures, called concomitant variable mixture approaches, apply to any of the MIX, GLIMMIX and STUNMIX model frameworks described in the preceding chapters. They allow for a simultaneous profiling of the segments derived with external “concomitant variables” and alleviate some disadvantages of previously used two-step procedures.
Michel Wedel, Wagner A. Kamakura
Chapter 10. Dynamic Segmentation
Abstract
In the segmentation methods discussed so far, we make the implicit assumption that the segments are stationary in structure and characteristics. Segment change is an important but currently under-researched area. In this chapter we describe approaches to track segment changes over time. Three types of models are identified and discussed. Two of the dynamic segmentation approaches build directly on the concomitant variable methods described in the preceding chapter.
Michel Wedel, Wagner A. Kamakura

Special Topics in Market Segmentation

Frontmatter
Chapter 11. Joint Segmentation
Abstract
The mixture models discussed so far are applicable for segmentation along a single dimension or basis, or at most, for segmentation along one criterion while using another dimension for segment-membership predictions (e.g., concomitant-variable mixture models). In this chapter we describe an extension of the latent-class model for joint segmentation, which define segments along multiple segmentation bases.
Michel Wedel, Wagner A. Kamakura
Chapter 12. Market Segmentation with Tailored Interviewing
Abstract
In this chapter we introduce the concept of tailored interviewing as a method to reduce respondent burden and interviewing cost. We then present a tailored adaptive interviewing procedure designed for market segmentation that is based on a mixture model approach. An application to life-style segmentation is provided, in which we demonstrate that the respondent burden is reduced by almost 80%.
Michel Wedel, Wagner A. Kamakura
Chapter 13. Model-Based Segmentation Using Structural Equation Models
Abstract
We describe a very general approach to response-based segmentation. Market segments are formed in the context of model structures in which multiple dependent variables are involved, and both dependent and independent variables are potentially measured with error: structural equation models. The method subsumes several specialized models, such as mixtures of simultaneous equations and confirmatory factor analysis. We start with a brief introduction to structural equation models.
Michel Wedel, Wagner A. Kamakura
Chapter 14. Segmentation Based on Product Dissimilarity Judgements
Abstract
Dissimilarity judgements have been represented in the marketing and psychometric literatures as either spaces or trees. The dominant tree representations have been the ultrametric and the additive trees. We describe an extension of those approaches to accommodate differential points of view of consumer segments, by allowing for different segments that have different perceptions of the stimuli, represented as mixtures of spaces, or mixtures of trees, or mixtures of both.
Michel Wedel, Wagner A. Kamakura

Applied Market Segmentation

Frontmatter
Chapter 15. General Observable Bases: Geo-Demographics
Abstract
In this chapter, we describe an increasingly popular form of market segmentation that combines multiple segmentation bases: demographics, geography, lifestyle and consumption behavior. Another distinct characteristic of this type of market segmentation is that its focus is on neighborhoods rather than individual consumers.
Michel Wedel, Wagner A. Kamakura
Chapter 16. General Unobservable Bases: Values and Lifestyles
Abstract
Lifestyle segmentation has been one of the most popular forms of market segmentation in the literature. We devote this chapter to a review of psychographics as it is applied to values and lifestyle segmentation.
Michel Wedel, Wagner A. Kamakura
Chapter 17. Product-Specific Observable Bases: Response-Based Segmentation
Abstract
The applications of market segmentation discussed in the preceding chapters are based on the assumption that values, lifestyles and geo-demographic profiles are associated with consumer behavior. That assumption is crucial if the segments identified on the basis of those variables are to differ in their response to the marketing mix, thus enabling managers to target specific marketing strategies to chosen segments. In this chapter we review applications in which segments are formed directly on the basis of consumer preferences and their response to price and sales promotions.
Michel Wedel, Wagner A. Kamakura
Chapter 18. Product-Specific Unobservable Bases: Conjoint Analysis
Abstract
We start by providing a brief introduction to conjoint analysis. Conjoint segmentation is the most common approach in the class of unobservable productspecific bases. We describe its use for market segmentation, and review the procedures that have been proposed for metric conjoint segmentation. Then we report the results of a recently published Monté Carlo study that demonstrates the relative strengths of a variety of proposed segmentation methods for metric conjoint analysis. We conclude the chapter by discussing segmentation for conjoint choice and rank-order data.
Michel Wedel, Wagner A. Kamakura

Conclusions and Directions for Future Research

Frontmatter
Chapter 19. Conclusions: Representations of Heterogeneity
Abstract
In this chapter we discuss an issue that has recently received much interest in marketing: is the discrete distribution of heterogeneity, imposed by the assumption of the existence of market segments, adequate?
Michel Wedel, Wagner A. Kamakura
Chapter 20. Directions for Future Research
Abstract
In this final chapter we briefly review the past and dwell upon segmentation strategy. We provide an agenda for future research.
Michel Wedel, Wagner A. Kamakura
Backmatter
Metadaten
Titel
Market Segmentation
verfasst von
Michel Wedel
Wagner A. Kamakura
Copyright-Jahr
2000
Verlag
Springer US
Electronic ISBN
978-1-4615-4651-1
Print ISBN
978-1-4613-7104-5
DOI
https://doi.org/10.1007/978-1-4615-4651-1