The Use of Brand Concept Maps and Network Analysis Tools to Examine Brand Associations Networks: An Abstract

Examination of the structure of brand associations is a necessary first step to understand brand image and build a brand with strong equity. Brands create associations in the minds of consumers which constitute brand image which is one of the two components of brand knowledge. According to the customer-based brand equity framework, brand knowledge, which embodies components of brand awareness and brand image, determines consumers’ response to the marketing of the brand. In other words, brand knowledge is the source of brand equity (Keller, 2003, p. 596), and the effectiveness of future branding strategies is influenced by the structure and content of the memory. Therefore, developing insights about the network structure of brand associations stored in the minds of consumers is crucial for brand managers.

The concept mapping technique is an effective way to elicit brand associations networks (John, Loken, Kim, & Monga, 2006). Additionally, the use of network analysis, which offers a myriad of tools that could advance the examination of brand associations, provides a great conceptual toolkit to study brand associations networks (Henderson, Iacobucci, & Calder, 1998, 2002). The use of these elicitation technique (i.e., brand concept mapping) and analysis tools (i.e., network analysis) in combination could provide several useful insights on branding. Therefore, this study proposes a new technique which involves the use of brand concept maps to elicit the network structure of brand associations and the use of network analytic tools to examine those brand associations networks analytically.

In this study, an unstructured concept mapping approach, in which no list of brand associations was provided to the participants, was adapted and the participants were asked to generate a concept map for a focal brand (i.e., Nike). Those individual brand concept maps were aggregated to a brand consensus map using network analysis. A cluster analysis with the Girvan and Newman (2002) algorithm was performed on the aggregated brand consensus map to discover the groupings of brand associations. The clustering results provided high face validity. Additionally, after a median split of the sample, comparisons between brand consensus maps of light versus heavy users of Nike products were made after conducting a quadratic assessment procedure (QAP) test. Light users refer to the participants who owned less than 20 Nike products, while heavy users refer to the respondents who owned 20 or more Nike products. According to the QAP test results, there is a correlation of 0.67 between brand consensus maps of light versus heavy users. This study contributes to the branding literature by advancing our understanding on the examination of brand associations networks systematically.