Essay: Limited and stable of generic terms

Introduction

Do brands have a personality ? Can they be described in terms of a set of traits or in terms of a limited and stable set of generic terms such as extraversion as it is done for human beings? This might be argued as advertisement tends to associate personality dimensions to brands (Plummer, 1984) and that marketing managers try to create an image and associate a personality to their brands through an array of strategies such as positioning, branding and advertising. However, in spite of the ancientness of the study of human personality, there has been surprisingly no piece of research exploring the specific components of brand personality before the famous article by Aaker (1997).

Human personality scales have been developed through factor analysis of items describing the self and others, these items having been selected from English language dictionaries (Goldberg, 1990). This lexical approach has led to a number of scales developed to capture dominants personality traits of individuals, such as the Mini-Markers (Saucier, 1994).

As far as brands are concerned, the research conducted by Aaker is a transposition to brands of approaches normally used to measure human personality. Contrasting this approach, we believe that the direct application of a human personality scale to brands can be of interest. What we propose here is to apply the Saucier Mini-Marker scale directly to brands and to study whether this scale is indeed applicable for brand personality measurements. Using a scale developed in English, we also have to question whether this instrument is stable across languages since our study is conducted in French, both in France and Canada (Quebec).

We divide this article into three parts. First, the literature on human personality and its transposition to brands are synthesized. The second part describes data collection procedures and methodology. Finally, we discuss results and both theoretical and managerial implications in the last part of the article.

Conceptual framework

Human personality traits

The origin of personality traits research is ancient since it can be traced back to Théophraste (4th century B.C.). However, in spite of the long research tradition, no unique and universally accepted definition of human personality traits prevails. Generally, they are defined as “tendencies to show consistent patterns of thoughts, feelings and actions” (Costa and McCrae, 1998). They are understood as being psychological cues that determine human action and experiences. Following the work of Allport (1937), Cattell (1950) and Eysenck (1960) considered as the founders of the dominant approach, a number of psychologists believe that the best representation of personality trait structure is given by the Five-Factor model, factors generally named the “Big Five” (John, 1990, see Block, 1995, for a critical vision).

Practically, these factors have been identified through two main approaches: the lexical approach and the hierarchical approach.

The lexical approach (Goldberg, 1990) is based on the hypothesis that all important traits must have been encoded in natural language due to the centrality of these personality traits. A factor analysis of words used to describe personality traits must enable to uncover the structure of personality and to identify fundamental dimensions of personality. However, the lexical approach is difficult to implement because of the vast number of words to be analyzed. Therefore, Saucier (1994) sought to work with a reduced list of adjectives and proposed a list of 40 statistically robust items named “Mini-Markers” allowing to measure the Big Five dimensions of human personality.

In the hierarchical approach, each factor summarizes a great number of characteristics. The five factors are to be found at the highest and most abstract level of the hierarchy. They constitute the structure of human personality and are defined by six conditional and contextual traits named facets. These facets include a great number of distinct and more precise characteristics describing personality. Costa and McCrae (1992) have proposed the Neo-PI-R (Neo Personality Inventory Revised) for measurement purposes.

In spite of the impressive number of studies conducted in psychology aimed at conceptualizing and measuring the structure of human personality, no parallel research has been conducted in the field of consumer behavior before the contribution of Aaker.

Brand Personality

Consumer behavior research has explored the way in which brand personality allows consumers to express their self-concept (Kleine et al., 1995). Practitioners view brand personality as a means of differentiation within a product category and as an important factor affecting preference (Biel, 1993) as well as a common denominator allowing to market a brand across different cultures (Plummer, 1984). The brand personality concept has also been severely criticized. From a conceptual standpoint, what brand personality really is remains ambiguous. How should it be defined? How and when is it different than brand image or than the image of the brand buyer? From a methodological standpoint, how should brand personality be best measured? Should traditional qualitative studies be employed or should adjective lists of human personality inventories be used? Finally, from a managerial standpoint, what are the consequences of the acquisition a personality for a brand and how may marketing activities create or alter it?

Aaker (1997) proposed a theoretical model of the brand personality concept through the determination of the number and the nature of its dimensions. This model is based on a hierarchical approach similar to that of McCrae and Costa (1997). Acknowledging the lack of consensus concerning the definition of the construct and its underlying dimensions, Aaker defines brand personality as “the set of human characteristics associated with a brand”. She identifies 42 traits and five brand personality factors: sincerity, excitement, competence, sophistication and ruggedness. This scale has been shown to be quite stable across different cultural contexts (Ferrandi, et al. 2000; Aaker et al., 2001).

Two other methods have been proposed to measure brand personality. Caprara et al. (1997) used a lexical approach to identify main attributes or markers of brand personality. Allen and Olson (1995) used a narrative approach to understand antecedents and consequences of brand personality.

Although the concepts of human and brand personalities might be similar, both constructs are different in their antecedents and in the roles they play. Human personality traits are created and communicated to others via attitudes, behaviors or physical characteristics (Park, 1986). They are thus inferred directly by others. By contrast, perception of brand personality traits is inferred through direct or indirect contact that consumers have with brands. Brand are inanimate objects which are associated with personality traits through marketing communications. Managers rely on the image of the typical brand user or the set of human characteristics that consumers associate with the typical user, on endorsement by celebrities, on product attributes, symbols, or any means of personification to develop the associations of brand personality (Batra et al., 1993). Contrarily to product attributes which are mainly functional, brand personality tends to have a symbolic function and one of self-expression (Keller, 1993).

Brand personality and human personality share indeed similarities: both are durable and might, at least under given conditions, help explain and predict the actions of individuals belonging to the target (see Fournier, 1998 for brand personality and Park, 1986 for human personality).

Research Design

Aaker’s methodology is a very nice adaptation to brands of what has been developed in the area of human personality measurement. However, as Aaker’s brand personality scale contains dimensions that are not shared by human personality scales, we have opted for the direct transposition to brands of a personality scale developed for human beings, that of Saucier (1994). For this first test, we have used a convenience sample and have studied a limited set of international brands well-known to the respondents on the French and Canadian markets.

Methodology

We hereafter present the nature of the sample and the procedures used to validate the structure of the brand personality scale.

Sample

The study was carried out in France (Nice) and Canada (Quebec). It is based on a convenience sample of 200 undergraduates students in business administration within each country, 55% of whom were female. All the students were 19-22 years old and had to evaluate the personality of 5 brands (Benetton, Levis, Adidas, Nescafé and Bmw) on the scale of Saucier. This scale is particularly interesting, first because it’s structure corresponds to the dominant approach and second, because of it’s parsimony (40 items, each evaluated on a 9 point Likert scale).

Validation of the Personality Scale Structure

In order to test the transposition of personality scale developed by Saucier (1994) to brands in a cross-cultural context, we have followed the recommendations of Churchill (1979) and the usual practices in cross-cultural research. The methodology followed is threefold.

In the first step the 40 original items were translated and back-translated by an interpreting company and done independently by two translators. Individuals differences were solved jointly by both translators. In the second stage we tested the scale structure. Based on the results obtained after performing principal component factor analysis with promax rotation, an iterative procedure allowed to purify the measurements through successive elimination of items ill-represented in the factors. Lastly, validity of scale structure was tested by means of trait validity performed through a confirmatory factor analysis. Trait validity was tested through convergent validity, discriminant validity and a reliability index (see Ferrandi et al; 2000, for a more detailed description of the procedure).

Results

The exploratory principal component analysis on the Saucier scale did not allow to recover the initial structure of the American scale since the number of factors with eigenvalues greater than 1 was lower than the personality traits identified by Saucier (4). Having constrained the structures to 5 factors, 25 items were eliminated due to low communalities. Final scale is thus composed of 15 items. This scale explains 63.66% of total variance. Secondly, we confirmed by means of confirmatory factor analysis the structure obtained at the previous step. Due to the non-normality of our data, systematic bootstrap procedures have been undertaken. Lastly, we tested the invariance structure of the scale between the two countries (Steenkamp and Baumgartner, 1998). We found that a constrained solution (equality of loadings) did provide the best fit to the data.

Table 1 shows the factorial structure of each 15 item scales as well as indicators of convergent validity and reliability. All indicator levels are good and allow to conclude that trait validity of scales is satisfactory.

TABLE 1

Brand Personality Scale Structure

Multi-group structural equation approach

Since we performed a multi-group analysis, we systematically tested for the invariance of the structural parameters between the two groups. The best results were obtained for the solution where the correlation coefficients were left free within each sub-group. The table 2 presents the “bootstrapped estimates” for each group.

TABLE 2

Correlation Coefficients between the brand personality factors

Results indicate that correlation coefficients between pairs of dimensions exhibit the same patterns for the two samples, French coefficients being slightly higher. On the other hand, relationships between neuroticism and agreeableness and between neuroticism and conscientiousness are not significant. In the French sample, the same goes for the relationship between introversion and agreeableness. Finally, relationships between openness and neuroticism and between conscientiousness and introversion have different signs. These differences need further testing.

Conclusion

The main objectives of this research was to apply the scale developed by Saucier (1994) to brand personality measurement and to evaluate and validate this transposed scale in a cross-cultural context (France and Quebec). Results obtained after purification and simplification of the scale are quite encouraging. The human personality scale has been shown to be applicable to brand personality measurement. The French and Canadian purified scales are composed of 15 items that load on 5 independent dimensions and we believe that the dimensions furnish a coherent and plausible representation of personality traits for the brand personality concept. The results are certainly exploratory, given some methodological limitations linked primarily to data collection and to statistical analysis.

As far as data collection is concerned, we have considered a limited number of brands which have been evaluated by a convenience sample. On another hand, the statistical analysis used was limited to principal component factor analysis and confirmatory factor analysis which limits validity of results. Finally, in spite of encouraging results, it could be useful to develop a scale designed for the French language through a lexical approach as done in Italy by Caprara et al. (1997). We would then obtain a bi-polar personality scale suited for the French context.

Beyond these methodological limitations there are some conceptual ones which certainly open room for further research. One direction is to better understand the antecedents of brand personality. How can the latter be managed, modified or acted upon by managers, beyond obvious communication policies? How does brand personality evolve over time? Does the brand personality concept apply equally at the brand level and at the branded product level?