چگونگی ایجاد شخصیت برند قابل تجدید How to create reproducible brand personality scales

The missing reproducibility of brand personality scales For several decades, marketing literature has addressed the assignment of human personality traits to brands. The theory of animism lies behind this approach (Gilmore 1919; Harvey 2005), in which tangible objects are characterized by attributes that are typically associated with humans. The association of human personality traits with brands encourages consumers to characterize brands as their partners or friends (Fournier 1998). Personality scales help marketers position and reposition their brands (Grohmann 2009). Brand personality is a strong predictor of brand equity (Aaker 1997; Grohmann 2009). One facet of brand personality is brand gender (Grohmann 2009). The positive effect of brand gender on brand equity has been demonstrated (Lieven et al. 2014). Thus, brand personality appears to be an ideal instrument to manage brands, to adjust brand personality according to consumers’ perceptions, and to compare a company’s own brands with competitors’ brands. As with any comparison, the measuring tool has to be valid and stable. Otherwise, comparisons are vacuous. Steenkamp and Baumgartner (1998) postulate the generalizability of consumer behavior theories to make them applicable across different countries. This is achieved when the measurements are equivalent and invariant. This holds whether the measurements occur among different brands, industry sectors, consumer groups, or countries and cultures.1 In recent years, scholars made recommendations to increase confidence in scientific findings by testing their reproducibility (Open Science Collaboration 2015). Sev- & Theo Lieven eral attempts have been made to test the reproducibility of the most famous brand personality model (Aaker 1997). The ambiguous results have been widely discussed in other literature. Thus, unlike the concept of brand equity that is shaped rather consistently in literature, with only few equity models not differing substantially from each other (Brady et al. 2008; Keller 1993; Yoo et al. 2000), it seems that a valid and reproducible brand personality model is unavailable. Moreover, the personality models themselves and their conceptualization have been challenged as being unable to measure brand personality (Azoulay and Kapferer 2003). Freling and Forbes (2005), for instance, argued that the definition of brand personality as ‘‘…the set of human characteristics associated to a brand’’ (Aaker 1997, p. 347) was vague and indistinguishable from brand image or brand identity. To narrow the purport of ‘‘brand personality,’’ Azoulay and Kapferer (2003) defined the term more precisely as the ‘‘…set of human personality traits that are both applicable to and relevant for brands’’ (p. 151). Even more precisely, Caprara et al. (2001) presented the most stringent constraint, claiming that ‘‘…personality descriptors [should] load on the same factor when used to describe human personality and brand personalities’’ (p. 381).