The Impact of Brand Personality on Brand-Aroused Feelings

The Impact of Brand Personality on Brand-Aroused Feelings

Dr Karen Miller
School of Management and Marketing, University of Southern Queensland, Toowoomba, QLD,
Email: Miller@usq.edu.au

Brand managers are increasingly using brand personality to differentiate and uniquely position their
brand without really knowing its effects. This study begins to addresses this gap by examining the
effects of brand personality on brand-aroused feelings across the product categories of sport shoes,
mobile phones and surf wear. Using SEM to analyse the data from 324 usable surveys, the findings
indicate that consumers perceive brand personality and brand-aroused feelings as two separate
constructs and that brand personality has a substantial effect on brand-aroused feelings. On the basis
of the findings, this study recommends that brand managers consider positioning their brand as
original, imaginative, considerate and kind if they want to arouse positive brand feelings.

Keywords: brand management; consumer behaviour; positioning strategy; integrated marketing
communications

INTRODUCTION
Over the past decade, there has been increasing interest in brand personality from academics and
practitioners (e.g., Aaker 1997; Caprara, Barbaranelli & Gianluigi 2001; Freling & Forbes 2005a;
Romaniuk 2008; Sweeney & Brandon, 2006), and the intensity of this interest demonstrates the need
to advance its development and investigate its effects. Managers perceive brand personality as a way
to develop a sustainable uniqueness in a marketplace proliferated by brands where it is becoming
increasingly difficult to come up with branded products that are truly different in the eyes of
consumers (Aaker 2003; Sweeney & Brandon 2006). A lack of perceivable differentiation may be
contributing to the reason why an estimated 75% to 85% of new brands will fail (Boyle 2003; Kohli
& Thakor 1997) or why brands are being commoditized and losing money (Barron 2003).
Increasingly, brand personality is used by brand managers as a strategic tool to provide a unique,
consistent, and enduring message to consumers (Matthiesen & Phau 2005) and/or as a positioning tool
to differentiate brands (Sweeney & Brandon 2006; Rekom, Jacobs & Verlegh 2006) , or as a symbolic
device to sustain a competitive advantage (Aaker 1997); however, the consequences of brand
personality haven’t been empirically determined or tested so while much money is going into the
development of brand personality its effects aren’t generally known (Freling & Forbes 2005a).
1 | P a g e

---

Along with brand personality being important to brand managers to endure competition and reduce
competitor erosion, increasingly so is developing an emotional brand (Aggarwal 2004; Algesheimer,
Dholakia and Herrman 2005; Lynch 2004; Thompson, Rindfleisch & Arsel 2006). Called emotional
because consumers may form an intimate bond with a brand, a bond that is passionate, similar to the
emotional bond a consumer experiences with a close circle of friends and/or family (Aggarwal 2004)
and this emotional bond is believed to be based on positive-aroused feelings. Gradually more being
realized, the importance of brand personality and developing positive brand-aroused feelings as
research in these areas enable brand managers to better develop positioning strategies, integrated
marketing communications and build sustainable competitive advantages in the marketplace.

Each in their own right, brand personality and brand-aroused feelings has had limited investigation,
and the effect of one on the other hasn’t been examined. Based on evidence from the psychology
literature (e.g., Meyer & Shack 1989; Yik, Russell & Suzuki 2003; Watson & Clark 1992), it is likely
that brand personality may have an effect on brand-aroused feelings. Brand-aroused feelings refer to a
range of specific conscious affective assessments a consumer experiences when encountering a brand,
such as feelings of happiness or sadness (Frijda 1991; Scherer 1996). An alternate view, the
Interpersonal Circumplex model, incorporates and examines personality and feelings as if they cannot
be separated (Sweeney & Brandon 2006). An issue with this viewpoint is that in the branding context,
consumers may assess a brand as having a personality, but it is unlikely that consumers may assess
whether a brand is happy or sad. Whilst there is general agreement that brands may be perceived to
have personalities (Aaker 1997; Romaniuk 2008; Sweeney & Brandon 2006) and consumers
experience feelings in response to brands (Escalas, Moore & Britton 2004), there is no evidence to
suggest that consumers assess brands to have feelings. Rather, it is likely that brand personality and
brand-aroused feelings are related, yet conceptually different constructs. The purpose of this paper is
to examine the impact of brand personality on brand-aroused feelings, (1) to determine if they are
unique constructs and (2) to determine if brand personality impacts on brand-aroused feelings.

2 | P a g e

---

To examine the impact of brand personality on brand-aroused feelings this paper will first address the
constructs of brand personality and brand-aroused feelings, followed by the research design, which
covers the method, measures and sample design before presenting the results. Following the results
will be the discussion, conclusion and implications section, which places the findings in the context of
the theory and discusses the implications of the findings for both theory and practice.

CONCEPTUAL DEVELOPMENT
Brand Personality
Brand personality is a construct that has had considerable attention in the branding literature since
Aaker’s seminal paper on brand personality (1997), which claimed that brands may be perceived as
having human personalities (e.g., Chang & Chieng 2006; Freling & Forbes 2005b; Rekom et al.
2006). This notion of brands having a personality is based on symbolism and animism theory (Aaker
1997; Fournier 1998; Freling & Forbes 2005b), which suggests that like people, brands may be
perceived as having consistent behaviour over time and from this behaviour a personality may be
imbued (Aaker 1997; Fournier 1998; Freling and Forbes 2005a; Wee 2004). Brand personality is in
the eye of the beholder, and consumers may perceive a brand as a “living entity” with a personality;
and this may be based on how a brand is perceived to treat its customers, deliver its marketing
communications, package its products, and is committed to being socially responsible (Aaker 1997;
Batra, Lehmann & Singh 1993; Venable, Rose, Bush & Gilbert 2005). Brand personality may be
inferred from the behaviour of the CEO, employees or customers of a brand, and advertising and other
marketplace behaviour (Aaker 1997; Batra & Homer 2004; Biel, 1993; Diamantopoloulos, Smith &
Grime 2005; Plummer 1984).

Up till now most of the literature has concentrated on exploring the construct of brand personality
(e.g. Aaker 1997; Batra et al. 1993; Caprara et al. 2001; Sweeney & Brandon 2006; Romaniuk 2008),
or associating brand personality with the symbolic benefits of a brand (e.g. Aaker 1997; Chang &
Chieng 2006; Freling & Forbes 2005b; Plummer 1984; Rekom et al. 2006) or with brand social
responsibility (Venable et al, 2005) or determining the difference between fit and personality
3 | P a g e

---

(Helgeson & Supphellen 2004; Lau & Phau 2007). Other studies have investigated the antecedents of
brand personality (Diamantopoloulos et al. 2005; Johar, Sengupta, & Aaker 2005) or the impact that
advertising plays on the development of the symbolic aspects of the brand, including brand
personality (Rajagopal 2006) or the impact of facial image and cosmetic usage on brand personality in
the cosmetic market (Gutherie 2008). So far the only study that investigates, in part, the emotional
impacts of some brand personality traits is Matzler, Bidmon and Graber-Krauter (2006), which found
that the brand personality traits of extraversion and openness have an effect on happiness and pleasure
that consumers feel toward the brand. This study suggested that future research should concentrate on
investigating the impact of all five broad brand personality traits (i.e. openness, extraversion,
conscientiousness, agreeableness and neuroticism/unstableness). As mentioned previously feelings
aroused by a brand may be positive or negative so it would be of value to investigate the impact that
brand personality has on both positive and negative feelings.

Brand Aroused Feelings
Along with brand personality, brand aroused feelings have been receiving growing attention in the
domain of marketing (Batra & Homer 2004; Escalas et al. 2004; Gobe 2001; Pham, Cohen, Pracejus,
Hughes, Mick & Baumgartner 2001; O’Cass & Frost 2002). Crucially, to highlight the importance of
brand-aroused feelings, Travis (2000) argues that products are different from brands, because, unlike
products, a brand arouses feelings. Travis (2000) argues that how a brand makes a consumer feel is a
crucial question, one that brand managers need to understand. Reviewing the literature on brand-
aroused feelings much of the research has come from the work of Escalas et al. (2004) and O’Cass
and colleagues (e.g. O’Cass & Lim 2001; O’Cass & Frost 2002; O’Cass & Grace 2003), and there
seems to be some agreement that the feelings aroused may be characterized as positive, warm or
negative (Escalas et al. 2004; O’Cass & Lim 2001). Findings indicate that in response to a brand
encounter, a consumer may experience specifically aroused feelings such as delight, happiness, or
inspiration, which are generally associated with positive aroused feelings. Alternatively, in response
to a brand, warm feelings may be experienced, such as feeling moved, sentimental or warm-hearted.
In other cases, a consumer may have negative feelings aroused, such as being offended, disgusted or
4 | P a g e

---

sad. For example, in response to an encounter with a brand (e.g. Coca-Cola), the mention of a brand
name (e.g. Coca-Cola) or seeing its brand identifiers, such as its symbol, or listening to its jingle,
Batra and Homer (2004) found that a consumer might experience positive affective assessments such
as happiness and/or pleasure.

Like personality, the notion of aroused feelings has its origins in the psychology literature and a
number of academics, for instance, Allik and Realo (1997), Eysenck (1967), Meyer and Shack (1989),
Yik et al. (2003) and Watson and Clark (1992) have clearly delineated personality and feelings as
distinct constructs, finding that one of the consequences of personality is its effect on aroused
feelings. Overall, this body of work indicates that around 40% (or more) of the variance in feelings is
accounted for by an individual’s personality. In fact, Meyer and Shack (1989) and Watson and Clark,
(1992) argue that the construct of personality should encompass all of the five factors (openness,
contentiousness, extraversion, agreeableness and neuroticism), which contribute to personality, to
ensure that the effect of personality on feelings is accurately accounted for – as encompassing some of
the five factors is likely to attenuate the results and impact on error.

The idea of personality having an effect on feelings has been carried over into the marketing
literature, where researchers such as Mooradian (1996) have investigated the effects of an individual’s
personality on the feelings aroused by advertisements. Mooradian (1996) found that personality
explained 37% of the variance in positive feelings, 36% of the variance in warm feelings and 44% of
the variance in uneasy/negative feelings. Interestingly, Mooradian’s (1996) results are similar to those
reported in the psychology literature on the relationship between personality and feelings.

In the context of brands, Smothers (1993), Biel (1993), Aaker (1997) and Wee (2004) suggest that
brand personality should arouse feelings. However, there has been no known empirical support for
this relationship. The only evidence about the effect of brand personality on brand-aroused feelings
has come from marketplace observations. For instance, Smothers (1993) found from his observations
that the personality of the brand, Nike, seemed to evoke strong positive and negative feelings. He
5 | P a g e

---

argued Nike’s brand personality aroused intense positive feelings of love and devotion; and there
were feelings of pride when consumers were able to save their money to buy a pair of Nike Air
Jordans. He suggested that the personality trait of achievement was associated with Nike through the
connection with Michael Jordan. Smothers (1993) argued that the personality of the brand, Nike,
aroused strong negative feelings when consumers were not able to afford Nike products. Consumers
are likely to find some brand personalities more appealing than others; if consumers find the
personality of a brand appealing, then it is likely that they will experience positively aroused feelings
as a response. Alternatively, if they find a brand personality unappealing, they may experience
negatively aroused feelings in response. As brand-aroused feelings will vary in intensity, as well as in
direction, it is also possible that in response to a brand’s personality a consumer will experience warm
aroused brand feelings. Given the findings in the psychology and marketing literature, and
suppositions in the branding literature, the argument proposed here is that, in response to a brand’s
personality, a consumer will experience brand-aroused feelings. Therefore it is hypothesized that:

H1:
that brand personality and brand-aroused feelings are unique constructs
H2:
brand personality will have a significant effect on brand-aroused feelings
H2a:
positive brand personality traits such as openness, conscientiousness, extraversion
and agreeableness will have a significant positive effect on positive brand aroused
feelings
H2b:
negative brand personality trait of unstableness will have a negative effect on
negative brand aroused feelings
H3:
the impact of brand personality on brand aroused feelings will be similar across
various product categories

RESEARCH DESIGN
To investigate the impact of brand personality on brand-aroused feelings, the method chosen was a
survey because the data can be collected quickly, inexpensively, and there is a reasonable degree of
sample control (Maholtra & Peterson 2005). To systematically develop the survey measures the
Netemeyer, Bearden and Sharma (2003) four-step scaling procedures process was used. The first step
6 | P a g e

---

is construct definition and brand personality was defined as a consumer’s unique characterization of a
brand in terms of the extent to which he or she assesses that it displays inner-psychological human
traits, this definition along with the definitions of each of the five factors that form brand personality
are displayed in Table 1.

Table 1: Five Broad Factors that Form Brand Personality
Definition
Brand a consumer’s unique characterization of a brand in terms of the extent to
personality which he or she assesses that it displays inner-psychological human traits
Openness The degree to which a brand is assessed as displaying novel solutions,
expressing original ideas, and using imagination in the marketplace.
Conscientiousness The degree to which a brand is assessed as displaying organization,
orderliness, and efficiency in carrying out its business in the marketplace.
Extraversion The degree to which a brand is assessed as displaying outgoingness,
gregariousness and sociableness in the marketplace.
Agreeableness The degree to which a brand is assessed as displaying consideration,
kindness and empathy in the marketplace.
Unstableness The degree to which a brand is assessed as displaying inconsistency,
unreliability or undependability in the marketplace.

The second step searches the literature for appropriate scales and according to Netemeyer et al. (2003)
it is important that the items reflect/form the construct definition. While there has been some use of
the scale developed by Aaker (1997) these measures were discounted because they measure of a
broader array of human characteristics associated with a brand, and are not solely a measure of brand
personality. Saucier’s (1994) scale from the psychology literature was used as it measures personality
and was a trimmed version of Goldberg’s (1993) scale considered to be one of the most robust
measures of personality (given the consistency of findings across a broad array of contexts; Hofstee,
Kiers, De Raad, Goldberg & Ostendorf 1997; Hofstee, de Raad et al 1992; Johnson & Ostendorf,
1993; Saucier 1992). Given the independence of these five factors in forming personality (see Barrick,
Parks & Mount 2005; Digman 1990; Goldberg 2006; Saucier 2002), when assessing brand
personality, the weighted score from one of the five factors should not affect the weighted score of
another factor. This means that a low weighted contribution from openness does not necessarily
indicate a highly weighted contribution from conscientiousness and vice-versa. All of this implies that
a brand’s personality will have varying degrees of contribution from each of the five factors. The
7 | P a g e

---

uniqueness of these weighted contributions is what forms a consumer’s unique characterization of a
brand in terms of the degree to which it is seen as displaying inner psychological human traits.

Regarding brand-aroused feelings, they were defined as a conscious affective assessment a consumer
experiences when encountering a brand. After an extensive search and analysis of the feeling scales in
the literature, the scale that appeared to be the most suitable was that developed by Edell and Burke
(1987), which includes the factors of positive aroused feelings, negative aroused feelings and warm
aroused feelings. The measurable items (i.e. adjectives such as happy, pleasure, warm-hearted,
disgusted or sad) reflect the first-order factors (positive, warm and negative). To operationalise the
construct of brand-aroused feelings according to Escalas et al. (2004), the adjectives should be scaled
to reflect the varying degrees of intensity experienced, from weakly aroused to strongly aroused
feelings. Thus, a high score on positively aroused feelings would indicate that it was strongly felt,
compared with a low score, which would indicate the positive feeling was felt by the consumer to a
lesser degree or not at all.

The third step in measurement and scale development is refining and trimming the measures, this was
done via a panel of expert judges, focus groups and a pilot test of the survey retaining a total of 40
items. For brand personality 27 items were retained and included five measures for conscientiousness
and extraversion, example items include orderly and outgoing, four items were retained for
unstableness and an example item would be erratic. For openness six items were retained and an
example item is innovative and for agreeableness seven items were retained and an example item is
considerate. For brand-aroused feelings 13 items were retained, which included five measures for
positive aroused feelings an example item is happy, four items were retained for negative and warm
feelings and example items include: sad, and warm-hearted.

The final step is finalizing the measures and determining the scale. In this study a seven-point lexical
scale (Oliver, Angleitner & Ostendorf, 1988) was selected because the reported reliabilities of the
existing scales were good; and according to Viswanathan et al. (2004) seven-point scales sufficiently
8 | P a g e

---

capture similarities and differences that are meaningful to participants. Lexical scales use
alphabetically arranged adjectives and are advantageous, because the degrees to which the scale items
load on their factors have been substantiated in prior research, even though in the survey they are
placed in alphabetical order (Oliver et al. 1988; Saucier & Goldberg 1996).

Generation Y (18-25 year old) consumers were selected as the sample to survey because there is
limited understanding regarding this cohort (Vahie & Paswan 2006), they are different to other
generations because they have been exposed to brands from birth onwards as they were born between
1980 and 1994 (Martin & Turley 2004), they are the second largest cohort and account for 28% of the
population (McCrindle 2007) and have significant economic and social influence in the marketplace
(Wolburg and Pokrwwcznski 2001). Focus groups conducted with Generation Y consumers found
that mobile phones (Nokia), sport shoes (Nike) and surf wear (Billabong) were important products
and brands and that everyone had some degree of experience with them. Thus, for the purposes of
generalisability (Raykov & Marcoulides 2006; Volckner & Sattler 2007) these three product
categories were chosen to examine the impact of brand personality on brand-aroused feelings.
Following the sampling procedures of Wolburg and Pokrywczynsk (2001), which studied generation
Y university students, one class from five Australian universities across three states (NSW, QLD and
ACT) were asked to participate. The size of the sample was dependent on the number of students
present on the day, and on a pre-arranged day and time the surveys were administered voluntarily to
self-identified generation Y consumers in the first ten minutes of class, consumers were given one of
three surveys, and the surveys were identical except for the name of the brand.

RESULTS
Returned were 324 usable surveys, 107 for Nokia, 106 for Nike and 111 for Billabong. As the survey
instrument used lexical scales (placing the brand personality and brand aroused feeling items in
alphabetical order), to assess if brand personality and brand-aroused feelings were separate constructs
a Harman’s one factor test was used. Often associated with common method variance, because a
single source of data collection (ie a survey) may contaminate the effects of the exogenous construct
9 | P a g e

---

on the endogenous, the Harman’s one factor test assesses whether one or more factors are produced
(Podsakoff & Organ, 1986). The results of the Harman’s one factor test showed no evidence of
common method variance, and provided evidence of eight factors, five for brand personality and three
for brand-aroused feelings with the initial items loading on their respective factors and together
explaining 65% of the variance.

To examine the impact of brand personality on brand-aroused feelings the structural equation
modeling technique of PLS (Partial Least Squares) was used as it considers all the path coefficients
simultaneously, and enables the researcher to avoid biased and inconsistent parameter estimates for
equations (Chin 1998; Fornell & Bookstein 1982; White, Varadarajan & Dacin 2003). To assess the
measurement model, according to Johnson, Herrmann and Huber (2006), the factor loadings should
exceed .707, the construct reliability should exceed .70 and the average variance extracted should
exceed .50. The measurement results were found to surpass the minimum requirement as Table 2
shows that the factor loadings ranged between .72 and .87, the composite reliabilities (CR) ranged
between .83 and .93, and the average variance extracted (AVE), which measures the shared variance
captured by the construct (Dillon & Goldstein, 1984) ranged between .61 and .74 indicating that the
variance captured was reliably measuring the construct purported. Assessing convergent and
discriminant validity, Fornell and Larker (1981) argue that the AVE is a useful estimate in assessing
convergent validity, as the AVE was found to be above .50, convergent validity was supported. Also
supported was discriminant validity because the squared multiple correlation between the two
constructs was less than either of their individual AVE; indicating that the constructs have more
internal (extracted) variance than variance shared between constructs (Fornell & Larker, 1981), a
result that further supports H1, the independence of brand personality and brand-aroused feeling
constructs.




10 | P a g e

---