Assessing Mutual Association between Alternative Market Segmentation Bases

Using a strategy-aligned fuzzy competitive analysis approach for market segment evaluation and selection

“…Even general studies of market segmentation have paid little or no attention to the evaluation and selection stages (Beane & Ennis, 1987;Weinstein, 1987;Wind, 1978). Authors generally limit themselves to analyzing how to evaluate segment stability (Bettman, 1971;Calentone & Sawyer, 1978;Lehmann, Moore, & Elrod, 1982;MacLachlan & Johansson, 1981), congruence (Green, 1977), internal homogeneity and profitability (Eckrich, 1984;Van Auken & Lonial, 1984;Beik & Buzby, 1973), to mention only the most relevant.…”

Segment Congruence Analysis: An Information Theoretic Approach

“…Furthermore, in discussing the criteria for evaluating alternative bases for segmentation, they treat variables and bases interchangeably. Both Green and Carmone (1977) and Van Auken and Lonial (1984) treat specific variables as bases, in segment congruence analysis. of the set of variables (e.g., deal prone and brand loyal, deal prone and brand disloyal, etc.)…”

“…Both Green and Carmone (1977) and Van Auken and Lonial (1984) used the likelihood-ratio chi-square in this way.…”

“…In the applications used by Green and Carmone (1977) and Van Auken and Lonial (1984), all of the variables were dichotomized. In such case, only one coefficient is generated for each two dimensional contingency property provides a powerful structural modeling tool for researchers concerned with the analysis of nominally scaled data, in all fields (Broekstra, 1978, 1981, 1982, Krippendorff, 1979, 1981, Klir, 1976, 1986.…”

“…Van Auken and Lonial (1984) Using the same four hypotheses as Green and Carmone's, they concluded that the four segmentation bases attitudes, benefits, problems, and usage, are mutually associated (In reality, in spite of their claim that all four variables were mutually associated, the hypothesis of mutual independence is rejected even if only two of the variables are significantly associated.) Then using the one-variable independence log-linear models, they concluded that the problems base was independent of the other three, but attitudes, benefits, and usage were not independent of the others.…”