AN INVESTIGATION INTO THE ASSOCIATION PATTERN TECHNIQUE AS A QUANTITATIVE APPROACH TO MEASURING MEANS-END CHAINS

Means-end chain theory links products to consumers by postulating hier archical relations between attributes of the product, consequences of product use and values of consumers. It has served as an important con ceptual framework for studies in marketing. The authors investigate th e association pattern technique (APT) as a supplement to laddering, th e most popular, qualitative measurement methodology in means-end chain s research. APT is a structured method for measuring means-end chains, suitable for large-scale surveys. It can be used in personal as well as quantitative mail interviews. APT separately measures the attribute -consequence, and the consequence-value links. The independence of att ribute-consequence, and consequence-value links is crucial to the vali dity of APT. Using loglinear models, we investigate this assumption fo r empirical data on four different products. Consistent support for in dependence is found. In addition, we use loglinear models to test the convergent validity of APT and laddering with respect to the content a nd structure of the means-end chains network that they reveal. The res ults show that the content of the APT and laddering networks differs. This result is explained from the different task formats. Most importa ntly, the hypothesis that the structure of APT and laddering networks is the same could not be rejected. (C) 1998 Elsevier Science B.V.