Expected price and user image for branded and co-branded sports apparel.
Wu, D. Gloria ; Chalip, Laurence
Expected Price and User Image for Branded and Co-Branded Sports
Apparel
The economic impact of sport is due largely to the ways that sport
ripples through manufacturing and consumer spending (Meek, 1997; Milano
& Chelladurai, 2011). The economic ripples of sport become manifest,
in part, through the effect of sport-related brands, such as sportswear
brands--for example, Nike and Adidas. Sportswear brands have long been
associated with athletes and teams, which typically display the brands
on their apparel. Given the cultural salience and popular sport
associations thereby imparted, sportswear companies began to widen their
net during the 1990s by creating products aimed at an audience that
wanted to project a sporty image, even if not an athlete. Given their
growing economic clout, sport-related brands became attractive as
co-branding partners for non-sport brands.
Co-branding is a brand alliance strategy in which two or more
brands are simultaneously presented to consumers in order to strengthen
their individual brand images (Geylani, Inman, & Ter Hofstede,
2008). This strategy has been recommended as a tool for differentiation
that leverages brands through the transfer of positive associations
among partners in the brand alliance, including brand equity, image, and
awareness (McCarthy & Norris, 1999; Simonin & Ruth, 1998;
Washburn, Till, & Priluck, 2000). Presumably, this should enhance
consumers' perceptions of the value of a product (Simonin &
Ruth, 1995), which would therefore render a price premium (Persson,
2010). Although the logic for a pricing effect is clearly articulated,
no study to date has tested whether the predicted effect occurs.
The significance is elevated by the ways that sport associations
are increasingly used in co-branding contexts to render the predicted
effect. For example, sportswear brands (e.g., Nike, Adidas) are
increasingly being used by fashion designers to enhance the equity in
their brands (Thomas, 2010), while sportswear brands are seeking the
same from their association with high-fashion brands (Mitchell, 2005).
The case of co-branding between sportswear brands and fashion designer
brands is particularly interesting not merely due to the importation of
sport-related associations into a fashion context, and vice versa, but
also because these two different kinds of brands have traditionally had
different or even contradictory brand images, and have even targeted
different market segments (cf. Holmes & Tierney, 2002; Lau &
Axelrod, 2009).
It is possible, however, that the value of co-branding comes not
merely from particular associations that are imported from a brand
partner, but also from the symbolic impact engendered by co-branding,
particularly when two very different kinds of brands are partnered.
Consumers do not buy products just for their practical utility, but also
for the symbolic meaning that products convey (Elliot, 1997; Levy,
1959), which has particular relevance to what consumers are saying about
themselves by the choices they make. It has been argued that individuals
may have a preference for particular brands over others because they
perceive themselves to be similar to others who use brands they favor,
and/or because they seek to identify themselves with others who are
thought to use particular brands (see Sirgy, 1982 for a review, and
Sirgy, 1986 for a comprehensive description of the underlying theory).
Thus, the judgments that consumers make about what brands and, by
extension, brand alliances say about users are important. For example,
if sportswear brands and fashion brands convey different images of
users, then the combination of the two should complicate or elaborate
the image in ways that are not yet well-understood. Further, the image
of users that are projected as a function of brand should affect
perceptions of value, and consequently of price, because consumers are
paying for a particular image of themselves and what they want to
project to the world, as well as the practical utility of the purchase
(Blackston, 2000; Evans, 1968). This is especially, albeit not uniquely,
true for clothing (Braun & Wicklund, 1989).
To date, co-branding studies have focused primarily on
consumers' attitudes towards the brand alliance per se, rather than
how consumers evaluate price and user image with reference to co-branded
products, their single-branded alternatives, or unbranded alternatives.
The purpose of this study was to examine consumers' expectations
regarding price and user image when a sport-related (i.e., sportswear)
brand is paired with a fashion designer brand, and to determine whether
the effect on expected price is, at least in part, a function of the
effect on consumers' projections regarding user image.
Literature Review
Brand User Image
Over forty years of research in marketing (e.g., Elliott, 1994;
Elliott & Wattanasuwan, 1998; Hirschman, 1981; Keller, 1998; Levy,
1959) has shown that the perceptions and associations consumers have
about brands go beyond their functional attributes and benefits, and
include non-functional, symbolic qualities. Researchers have argued that
individuals often use symbolic brand meanings for personal-expression
and social communication (Elliott, 1997; McCracken, 1988, 1993).
This argument is rooted in two complementary streams of work,
symbolic self-completion theory and symbolic interaction. Symbolic
self-completion theory (Wicklund & Gollwizter, 1982) argues that
many behaviors, such as purchase choice, are intended to substantiate
the consumer's definition of themselves. Accordingly, individuals
are particularly likely to value symbols, such as possessions, that
reinforce identities about which they may feel insecure but to which
they are committed. Rather than merely value these possessions,
individuals also like other people to recognize their symbolic worth.
Symbolic interaction theories (Leigh & Gabel, 1992; Blumer,
1969; Kinch, 1967; Mead, 1934; Solomon, 1983) also suggest symbols play
a pivotal role in human behavior. Symbolic interaction contends that
humans value meanings and devise ways to associate meanings to their
actions (Kaiser, Nagasawa, & Hutton, 1991). A symbol can be a
stimulus with a learned meaning and value, and the person's
response to the stimulus is in terms of this meaning (Rose, 1962). There
are three fundamental postulates in the symbolic interaction theory:
"1. A consumer's self-concept is based on perceptions of the
responses of others. 2. A consumer's self-concept functions to
direct behavior. 3. A consumer's perception of the responses of
others to some degree reflects those responses." (Solomon, 1983, p.
320).
Brands can be used by consumers as resources for the symbolic
construction of the self (Elliott & Wattanasuwan, 1998). Claims that
brands could affect user image are grounded in self-congruity theory.
According to self-congruity theory (Sirgy, 1982, 1986), consumers
evaluate and compare their self-concept with the image of a stereotypic
and generalized brand user, which is referred as "user-image,"
and prefer to buy products that match their self-perceptions. By
choosing brands with particular image associations (e.g., elegant,
sporty), consumers seek to project an impression to others of the type
of person they are (or that they want to be seen as)--a process that
also enhances their own self-image and psychological well-being (Sirgy,
2012). Within a product category, the value that a particular brand has
for the consumer depends in no small way on what it is thought to
communicate about the user (Aquirre-Rodriguez, Bosnjak, & Sirgy,
2012; Parker, 2009).
Little is known about the relationship between price and user
image, although it has been shown that consumers associate higher status
with higher prices (Leigh & Gabel, 1992; Wright, Claiborne, &
Sirgy, 1992). This finding has been extended into studies of so-called
luxury brands, which are normally high priced. Research demonstrates
that the positive image associated with luxury brand purchases plays a
key role in their purchase (Liu, Li, Mizerski, & Soh, 2012; Nueno
& Quelch, 1998). Thus, higher priced goods are normally expected to
convey a positive image of the user, but it is not clear how specific
user characteristics projected by a brand might affect consumers'
expectations about price. The fact of a relationship seems clear,
although its specific mechanisms are not yet well understood. Since
sportswear brands and designer brands convey different user images
(Azevedo & Farhangmehr, 2005). Therefore, it is expected that
consumers will describe wearers of each type of brand, as well as the
combination of such brands, differently. That should affect the value
and, consequently, the price that they expect to pay, as noted above.
For the purpose of this study, it is necessary to find a scale that
categorizes and explains the important dimensions represented by
adjectives that people use to describe a product user's personality
characteristics.
Brand User Image Measurement
If the image that is projected by using a brand is important, then
it is necessary to identify appropriate measures of that image. It has
proven difficult to untangle the theoretical issues from measurement
concerns. Aguirre-Rodriguez, Bosnjak, and Sirgy (2012) argued that
self-congruity evaluations during product choice can be either holistic
or piecemeal. In other words, consumers can make global evaluations of
the self-congruity associated with a product, or they can take a
piecemeal approach wherein they identify particular traits they
associate with product alternatives and make a choice based on the
different traits. Although they found that the self-congruity effect is
generally stronger and more complete when consumers use holistic (rather
than piecemeal) evaluations for a product class, they also found that
brand self-congruity effects are stronger when processing is piecemeal
(rather than holistic). In other words, when evaluating a brand,
evaluation with reference to particular traits is stronger than holistic
evaluation. If that is the case, then a piecemeal approach to evaluation
may better capture the detailed trait-by-trait comparisons used when
evaluating brands (or co-brands).
It has been argued elsewhere that there might be a transfer from
brand personality to consumer personality such that by building a
particular brand personality, a brand could contribute to building a
particular, perhaps ideal, self-image for consumers (Fennis, Pruyn,
& Maasland, 2005), particularly in the case of clothing (Fennis
& Pruyn, 2007). If this is true, then it might be appropriate to use
a brand personality scale to measure user image. However, there are two
reasons that might not be the best way to proceed. First, it has been
argued that the current measures of brand personality do not, in fact,
measure brand personality, but instead aggregate multiple dimensions of
brand identity (Azoulay & Kapferer, 2003; Caprara, Barbaranelli,
& Guido, 2001). Second, although brand personality and user image
are expected to be associated through consumers' search for
congruity, they are conceptually distinct components of an overall
process by which consumers evaluate products and brands (cf. Aaker,
1997; Sirgy, 1983).
An alternative approach derives from the extensive work on semantic
differentials deriving from Osgood, Suci, and Tannenbaum (1957), which
consistently finds that three dimensions--which Osgood et al. call
"evaluation, activity, and potency"--usefully and adequately
capture judgments about people and things. Indeed, Mehrabian (1980) has
argued that these three dimensions, which he calls "pleasure,
arousal, and dominance" can (and should) be generally applied to
measure people's judgments about the world. The basis for that
argument begins with Snider and Osgood's (1969) contention that
bipolar adjectives provide an ideal means to capture and describe
people's emotions and attitudes. Conceptual support comes from
semiotic theory which holds that humans employ opposites in the
cognitive identification of concepts; we tend to understand what
something is by first knowing what it is not (Chandler, 2002; Valentine
& Evans, 1993).
A substantial body of work recommends the three dimensional model
derived from semantic differentials. The model has been found to be
highly portable, and to identify and describe the feelings people share
about objects (Brewer, 2004; Westerink, Krans, & Ouwerkerk, 2011).
The three-dimensional structure holds up with a wide variety of
subjects, concepts, and scales (Heise, 1970), across cultures (Osgood,
May, & Miron, 1975), and across applications. The measure has been
shown to describe gendered traits (Kroska, 2002; Langford &
MacKinnon, 2000), leadership styles (Schneider & Schroder, 2012),
and the underlying meaning of occupational prestige scores (MacKinnon
& Langford, 1994). The three dimensions have been shown to be
particularly relevant to assessments of personality, as the evaluative
dimension has been identified with status (Kemper & Collins, 1990),
while the activity dimension has been associated with emotional energy
(Collins, 1990) as well as self-identified expressiveness (Heise, 1989),
and the potency dimension has been related with power (Kemper &
Collins, 1990). The three dimensions have been extensively adopted by
psychologists, consumer researchers, and other social scientists because
they have demonstrated excellent reliability, validity, and parsimony
(Abbott, Shackleton, & Holland, 2008; Heise, 1970).
In this study, we employ these same three dimensions. In previous
work, authors have varied somewhat in the labels that they apply to
each. We have endeavored to retain labels that are comparable to those
used in most previous work, and that seem to best capture what the
measures describe here. So we have called the dimensions
"goodness," "engagement," and "dominance."
Brands and Expected Price
Since brands convey meaning to consumers, they also convey a sense
of value. Forsythe (1991) showed that brand names affect shoppers'
expectations regarding quality and price. Brand affects market share and
consumer choice (Agarwal & Rao, 1996; Ailawadi, Lehmann, &
Neslin, 2003), while price has been shown to be the most useful
indicator of brand equity and associated profitability (Aaker, 1996;
Blackston, 1995). Consequently, differences in the expected price of
differently branded products are useful for determining the effect that
particular brands have on consumer evaluations of a product.
Since co-branded products carry more than a single brand, the
effect on expected price should be elevated, at least if the brand
alliance enhances consumer perceptions of the product. Schema theory
provides the bases for expecting an increase in price when a product is
co-branded, as well as the added complexity of user image, as described
at the beginning of this literature review. Schema theory explains how
people make meaning out of information. A schema can be identified as a
generic or abstract knowledge structure in an individual's memory
that is used to guide encoding, organization, and retrieval of
information (Stein & Trabasso, 1982). It consists of category
attributes of stimulus domains as well as their links, prototypic
exemplars, and affective tags assessing one's attitude toward
members of the category (Robertson & Kassarijian, 1991; Goodstein,
1993). The process of integrating new information into an existing
schema allows for held beliefs to be modified or reinforced.
Accordingly, in the case of co-branding, each brand logo plays the role
of a product cue that activates self-schemata. When consumers receive
information about two different brands, the two different brand images
each affect consumers' perceptions of the user image. Studies of
co-branding do suggest that consumers' attitudes or perceptions
regarding brand-user-image can be changed by the information they
incorporate about the two brands (Ahn, Kim, & Forney, 2010).
This, in turn, should affect expected price. Some evidence comes
from research demonstrating that co-branding can alter consumers'
expected utility for products such that consumers commonly overestimate
the price of a co-branded product (Chang, 2008). On the other hand, if
there is asymmetry in what two brands contribute to a co-branded
alliance or if the two brands match poorly, there can be a suppressive
effect, which might then have a negative effect on expected price
(Venkatesh & Mahajan, 1997, 2009). Thus, while a brand alliance
should affect consumers' expectations regarding price, the
direction of effect could vary as a function of which brands are allied.
Gender Differences
Men and women have been considered as two different but important
market segments, especially in the case of apparel. Gender differences
have been reported in a wide range of studies in advertising and
marketing. Research shows that male and female customers perceive brands
differently (Elliott, 1994; Kamineni, 2005), and women were more likely
to attribute positive characteristics to owners of fashion products
(Mayer & Belk, 1985). Furthermore, females are generally more
innovative in terms of fashion and shopping (Kim & Kim, 2004;
McCracken & Roth, 1989; Stith & Goldsmith, 2006).
These differences extend to perceptions of user characteristics and
prices. Culturally prescribed expectations regarding appropriate
gender-based differences in user characteristics affect product and
brand preferences (Jubas, 2011). Women are also more likely than men to
view high prices as socially acceptable (Maxwell, 1999).
Taken together, these findings demonstrate that male and female
consumers typically evaluate products and brands differently. These
differences extend to projected user characteristics, and to price. For
those reasons, it may be necessary to evaluate effects of brands and
co-branding separately for males and females.
Research Questions
Although there are grounds in preceding studies for expecting that
there might be an effect of combining a sportswear brand with a fashion
brand, no previous work has examined whether there is an effect.
Consequently, nine questions derive from the preceding review:
1. Is there a general branding effect on perceptions of user image
and expected price for a piece of clothing? In other words, is a brand
of any kind (whether single or co-brand) better than no brand?
2. Is there a difference between sportswear brands and fashion
brands on perceptions of user image and expected price for a piece of
clothing?
3. Is there a general co-branding effect on perceptions of user
image and expected price for a piece of clothing?
4. Does adding a fashion brand to a sportswear brand help the
sportswear brand with reference to user image and expected price for a
piece of clothing? In other words, is the fashion brand better off when
a sportswear brand is added?
5. Does adding a sportswear brand to a fashion brand help the
fashion brand with reference to user image and expected price for a
piece of clothing? In other words, is the sportswear brand better off
when a fashion brand is added?
6. Is the co-branding effect the same for different fashion
designer brands with reference to user image and expected price for a
piece of clothing?
7. Is the co-branding effect the same for different sportswear
brands with reference to user image and expected price for a piece of
clothing?
8. How well does consumer evaluation of user image for a piece of
clothing predict expected price?
9. Do branding and co-branding effects vary as a function of
gender?
The last question can be examined via MANOVA. Each of the first
seven questions can be answered within the context of a single analysis
(following MANOVA) using planned orthogonal comparisons (one for each
question). Question 8 requires a regression analysis.
Method
Pretest: Choice of Sportswear Brands and Fashion Designer Brands.
Two initial studies of sportswear brands and fashion designer
brands were conducted in order to select the brands to be used in this
study. We sought two of each type of brand in order to determine if the
effects we found were brand-specific or were comparable in each brand
category (Research Question 6 and 7). To better understand the brands
that are both familiar and liked, we sought to pinpoint the level to
which fashion designer and sportswear brand names were familiar and
liked.
Fashion designers were examined in the first calibration study.
Four designers were selected from the list of "Top 10 American
Fashion Designers" (Vercillo, 2009). The four designers--Marc
Jacobs, Ralph Lauren, Tommy Hilfiger, and Oscar de la Renta--had no
affiliation with any sportswear brand, and all four were male and
American, which assured that subsequent findings would not be biased by
previous alliances with a sportswear brand, or differences in designer
gender or nationality. Further, four fake designer names were created
and added into the list to test the overall reliability of
participants' answers.
Forty students at a large American university rated how familiar
they were with the designers and how likeable they thought the designers
were using a 7-point Likert-type scale. A random order of brands was
used on each questionnaire to eliminate order effects. One fake brand
was reported to have positive familiarity by two students. Their
questionnaires were removed from further analysis. The remaining 38
participants' mean ratings were then averaged. The results showed
that Tommy Hilfiger and Ralph Lauren scored highest and near the top of
the scale in categories of familiarity and likability. They were
selected for the final study.
The second study followed similar procedures to determine which
sportswear brands would be used in the final study. Initially, eight
sportswear brands, Lotto, Nike, Starter, Adidas, Puma, Champion, Umbro,
and Harnes, were selected from a list of local store brands. Lists in
which the names were randomized were given to a different set of 40
students at the same university as the previous calibration study.
Moreover, the same scales of familiarity and likability from the first
study were used in this study. Nike and Adidas were determined to be the
most familiar and best liked, so those two were chosen for the final
study.
Participants
College students were deemed the best participants for the main
study, because of their higher acuity of consumer involvement
(O'Cass, 2000), and because they are a key target demographic for
co-branded apparel (Cheng, 2010; Thomas, 2010). Participants were 275
students (110 men and 165 women) ranging in age from 18-42 years
(M=22.65, SD=3.98). Participants were recruited from six undergraduate
liberal arts classes at a large public university in the American
southwest. No participant was in more than one of these classes. Most
(50.5%) were Caucasian; the second highest ethnicity reported (23.6%)
was Asian; the remainder (25.9%) reported an array of other ethnicities.
To avoid prior knowledge about branding, co-branding, or pricing,
students who had obtained coursework in advertising, economics, or
marketing were eliminated, and did not participate. None of the
potential participants was majoring in fashion design, which would also
have been grounds for elimination.
Design
A 3 (fashion designer brand) X 3 (sportswear brand) experimental
design was employed. The three levels of fashion designer brand were:
absence of fashion designer brand logo, presence of fashion designer
logo A (Ralph Lauren), and presence of fashion designer logo B (Tommy
Hilfiger). The three levels of sportswear brand were: absence of
sportswear brand logo, presence of sportswear brand logo A (Nike), and
presence of sportswear brand logo B (Adidas). Thus, there were nine
conditions in this experiment. Each participant was randomly assigned to
one of the nine experimental conditions with the requirement that an
equivalent number of women and men were in each condition.
Guided by Swartz (1983), polo shirts were chosen as an appropriate
stimulus because polo shirts display different brands, and those brands
make a difference in consumer perceptions. Further, polo shirts are
widely worn by both men and women, and are widely available in both
sportswear and fashion designer brands. Following Sproles (1981), nine
black and white drawings were developed to correspond to the nine
experimental conditions. All nine conditions were accompanied by the
description, "All colors and sizes are available." To optimize
logo location, a pretest was conducted with 30 students in an
undergraduate class (who were not included in the main study) in which
logos were located in different places on the shirt. All participants
preferred the unbalanced logo placement (sportswear logo and fashion
designer logo on the left side of the neckline). Therefore, the
unbalanced placement (which is the same as is used in nearly all polo
shirts on the market) was used in this study.
Measures
Two sets of scales needed to be identified or developed in order to
answer the research questions: a scale to measure expected user
characteristics and a scale to measure expected price.
Development of the user image scale. A trait adjective scale to
describe wearers was developed to represent the three dimensions
commonly found in semantic differential work: goodness, engagement, and
dominance. Forty-five differential adjectives were drawn from previous
studies (Hannover & Kuhnen, 2002; Holman, 1980; Peluchette &
Karl, 2007). Following review by a group of 10 students who evaluated
the suitability of each pair as a reasonable descriptor of persons
wearing a polo shirt, 13 adjective pairs were selected to represent the
three target dimensions. This yielded a list of 5 adjective pairs for
the goodness dimension, 3 adjective pairs for the engagement dimension,
and 5 adjective pairs for the dominance dimension, each of which was
rated on a 7-point scale.
The 13 semantic differential items were used in the main study, in
which each respondent was asked to rate a person who might wear the polo
shirt that was pictured. The direction of semantic differentials was
varied to prevent response set.
Confirmatory factor analysis was used to test and refine the scale.
The model with all 13 items yielded marginal fit ([chi square] = 236.27,
df = 62, p<0.001, RMR = 0.11, GFI = 0.88, AGFI = 0.83, CFI = 0.79,
RMSEA = 0.10). Examination of the standardized loadings for each item
onto its latent trait revealed that four items (two for the goodness
dimension and two for the dominance dimension) had loadings below 0.5.
Those items were removed and the model was refit. The resulting model
showed suitable fit ([chi square] = 68.92, df = 24, p<0.001, RMR =
0.07, GFI = 0.95, AGFI = 0.90, CFI = 0.92, RMSEA = 0.08). The average
variance extracted for the three dimensions was substantially higher
than the squared multiple correlations between the dimensions,
indicating that the three dimensions were adequately independent
(0.38<AVE<0.53; .11<[r.sup.2]<.26). The three items
representing each of the three dimensions were averaged to obtain the
score for each dimension that was used in the final study. Scores for
each could therefore range from 1 to 7. Alpha coefficients and
standardized factor loadings (from AMOS) for each subscale are shown in
Table 1.
Price. Using prices for polo shirts in local stores, 12 different
price ranges, from "less than $10" to "$111 or
more," were listed. Each category had a $10 range. Participants
were asked to select the price they expected would be charged for the
polo shirt they were shown. For purposes of data analysis, the midpoint
of the price range selected was entered. Thus, the difference between
groups was the average number of dollars of difference in expected
price. Since no participant picked the highest level price category, it
was possible to estimate a midpoint for all categories chosen.
Procedure
The picture and survey instruments were distributed to participants
in each class. Participants were instructed first to read a cover letter
that explained what would follow along with a consent form. All students
invited to participate consented to do so. The cover letter instructed
participants to examine the picture and to read the words (which
specified that the shirt was available in all colors and sizes)
carefully before turning to the survey. They were also instructed to
refer to the picture whenever they felt that was necessary when
answering the questions. The only time they could not turn back was
during the manipulation check immediately following their viewing of the
picture, at which time they were asked to identify (from a checklist)
which brands they had seen. Eleven respondents failed to identify the
brands correctly, so their data was eliminated from further analysis.
(There were initially 286 participants, but elimination of these 11
brought the total to the 275 reported above.) After the manipulation
check, they responded to the scales for user image and expected price,
followed by items measuring gender, age, major, and ethnicity.
Results
A 3x3x2 MANOVA was used to test the effects of fashion designer
brand, sportswear brand, and gender on the three user image dimensions
and expected price. The three levels of fashion designer brand (Ralph
Lauren, Tommy Hilfiger, none), and the three levels of sportswear brand
(Nike, Adidas, none) and gender (male and female) were the independent
variables. The four dependent variables were expected price and the
three dimensions of the user image scale (Goodness, Engagement,
Dominance). The four dependent variables were sufficiently correlated to
enable confidence in the statistics, but not so highly correlated as to
be redundant (see Table 2.). The three-way interaction was statistically
significant; Wilks' Lambda=0.885; F[16,777]=1.99, p=0.01; partial
eta squared = 0.03. This three-way interaction suggests that other
effects depend on the consumer's gender.
To consider that possibility, which is also Research Question 9, we
examined the gender difference on means and on the relationships between
price and the three user image dimensions. Examination of the univariate
tests showed that the three way interaction was significant for
dominance (F[4,257]=4.78, p=0.001; partial eta squared = 0.07) and
goodness (F[4, 257]=4.53, p=0.001; partial eta squared = 0.07). The
interaction for engagement (F[4,257]=1.39, p=0.23; partial eta squared =
0.02) and price (F[4, 257]=0.78, p=0.54; partial eta squared = 0.01) was
not significant, although there was a significant main effect of gender
on engagement(F[1,257]=3.32, p=0.05; partial eta squared = 0.02). Taken
as a whole, these findings indicate that males and females evaluate the
user image for these shirts differently. It is heuristically useful,
therefore, to treat the two genders as separate populations when
comparing means. Consequently, analyses addressing Research Questions
1-7 are conducted separately for males and females.
Treatment Effects for Males
ANOVAs with planned orthogonal contrasts provided precise tests of
each of the first seven research questions for male participants
(Seltman, 2012, pp. 319-338). The first contrast tested the mean of the
eight conditions with at least one brand against the condition with no
brand. The second tested the mean of the fashion designer brands against
the mean of the sportswear brands. The third tested the mean of the
single branded shirts against the mean of the cobranded shirts. The
fourth tested the mean for shirts with only a sportswear brand against
the mean for all cobranded shirts. The fifth tested the mean for shirts
with only a fashion brand against the mean for all cobranded shirts. The
sixth and seventh tested whether co-branding effects were comparable for
the two fashion designer brands or the two sportswear brands,
respectively. Results are summarized in Table 3. The contrast values
shown in Table 3 are the weighted mean differences for the contrast
required to provide a direct test of each research question. Table 3
also shows Cohen's d, which estimates the size of the difference
between the two means compared in each contrast. The size of effect is
shown in standard deviation units. As a rule of thumb, a d around 0.2 is
considered small, a d around 0.5 is considered medium, and a d of 0.8 or
greater is considered large.
Examination of Table 3 shows that men felt that a brand or co-brand
would, on average, be $24 more expensive than no brand. This was a large
and statistically significant effect. They also felt that a fashion
brand was almost $4 more expensive than a sportswear brand, an amount
that was also large and statistically signifcant. Finally, they felt
that co-branding by adding a fashion brand to a sportswear brand would
increase price by almost $6, an amount that was also large and
statistically significant. However, they did not think that adding a
sportswear brand to a fashion brand would have a significant effect one
way or the other on price. In other words, a fashion brand added
perceived value to the sportswear brand, but a sportswear brand had
neither a positive nor a negative impact when added to the fashion
brand. Other comparisons were also insignificant.
Brands also made a significant difference in the ways that males
perceived wearers of the shirts. Having a brand of any kind rendered a
perception that the wearer was significantly more engaged. If the
product was single branded, a fashion brand rendered a significantly
higher rating for engagement, dominance, and goodness than did a
sportswear brand. These effects were medium to large. Co-branding had no
significant effect on perceptions of goodness or engagement, but it had
a significant negative impact on perception of dominance. In other
words, a single branded shirt rendered a higher perception of dominance
than did a cobranded shirt. This was due primarily to the effect that
co-branding had for the fashion brand, for which perceptions of
dominance were substantially reduced, whereas the effect for the
sportswear brand was insignificant. The same was true in the case of
goodness and engagement. In short, to the degree that goodness,
engagement, and dominance are desirable qualities to represent,
co-branding was detrimental for fashion brands, but neither helpful nor
harmful for sportswear brands on average. However, under the cobranded
condition, Nike was substantially (and significantly) better off than
Adidas in terms of perceptions of wearer dominance and engagement, and
Ralph Lauren was moderately (and significantly) better off than Tommy
Hilfiger in terms of wearer goodness.
Treatment Effects for Females
ANOVAs with planned orthogonal contrasts provided precise tests of
each of the first seven research questions for female participants
(Seltman, 2012, pp. 319338). Contrasts were the same as those for males.
Results are summarized in Table 4. The contrast values shown in Table 4
are the weighted mean differences for the contrast required to answer
each research question directly, and Cohen's d provides an estimate
of effect size in standard deviation units.
Examination of Table 4 shows that as with the men, the presence of
any brand whatsoever (single or co-brand) was expected to render a
substantially and significantly higher price (almost $21 higher) than
when no brand was displayed. Females also felt that a fashion brand was
worth significantly more ($2.55 more) than a sportswear brand. This was
a large effect. They felt that when a fashion brand co-branded with a
sportswear brand, it was worth significantly less ($3 less) than when it
was displayed on its own. This was a moderate effect. They did not feel
that a sportswear brand was helped or hurt significantly by being paired
with a fashion brand.
Branding and co-branding had fewer effects on females'
perceptions regarding the wearer of the shirt than either did for males.
Females felt that a fashion brand demonstrated moderately (but
significantly) higher engagement than did a sportswear brand, and they
felt that allying a sportswear brand with a fashion brand moderately
(but significantly) lowered the fashion designer brand's
demonstration of engagement. They felt that any form of co-branded shirt
demonstrated moderately lower wearer goodness than did a single-branded
alternative. Finally, if there was to be co-branding, they felt that
Nike was moderately (but significantly) better off than was Adidas in
terms of goodness and dominance.
Effects of User Image on Expected Price
Although mean differences between treatment groups differ as a
function of gender, the effect of user image dimensions on price may
not. This is Research Question 8. To test whether the relationship
varies as function of gender, tests for linear restrictions were
conducted (Katos, Lawler, & Seddighi, 2000, pp. 4059). The
relationship between the three user image dimension variables and
expected price does not significantly differ by gender (F[4, 267]=0.726,
p>0.05). Therefore, to answer Research Question 8, the regression
analysis is conducted for males and females jointly.
In order to examine the relationship between evaluations of user
image and expected price, expected price was regressed on the three user
image dimensions. The three user image dimensions provided small but
significant prediction of expected price ([R.sup.2] = 0.07, p<0.01).
Examination of Table 5 shows that both engagement and dominance provide
significant prediction of expected price. The higher a
participant's rating of the wearer's engagement or dominance,
the higher they expect the price to be.
Discussion
Findings here are partially but not fully consistent with Blackett
and Boad's (1999) suggestion that co-branded products should
command a premium above the prices enjoyed by equivalent products
bearing only a single brand. Although display of any brand or co brand
rendered a much higher expected price than did the display of no brand,
allying sportswear brands with fashion designer brands was only helpful
for sportswear brands, and only among males. The price males expected to
pay for fashion designer brands was neither helped nor hurt by
co-branding. For females, on the other hand, the expected price for
fashion designer brands was hurt by co-branding with a sportswear brand,
and the sportswear brands' expected price was not significantly
helped by co-branding. The reasons for these effects are unclear, as the
observed asymmetries are not adequately accounted for by current
theories of brand associations or gender differences.
The percentage of variance attributable to gender in the three-way
interaction between brand types and gender was small, and was not
significant in univariate tests on each variable. Although the overall
pattern of results here is consistent with the expectation that males
and females will evaluate clothing (in this case in terms of projected
user characteristics) differently, the effect is subtle. It is not
merely that males and females provide different ratings; the pattern of
rating differences varies by gender, which suggests that they are
processing clothing brands differently. The differences are apparent
when user characteristics are considered, but not in terms of price.
Future work should explore differences in the cognitive processing males
and females apply when evaluating brand cues, particularly with
reference to images and evaluations of a product other than price.
Cognitive consistency theory (Schewe, 1973) predicts that consumers
will seek to maintain consistency and internal harmony among their
attitudes. Accordingly, when evaluating a product with two brands,
consumers should assimilate their attitudes towards the parent brands
such that their attitudes towards the co-brand would be an average of
the parent brand attitudes (Levin, Davis, & Levin, 1996). That seems
to be the case in part. Females reduce the value they associate with a
fashion designer brand, and males increase the value they associate with
a sportswear brand. Thus, for females the lower value of the sportswear
brand does reduce the perceived value of the fashion designer brand when
the two are paired, as cognitive consistency theory predicts. Similarly,
for males, the higher perceived value of the fashion designer brand does
increase the perceived value of the sportswear brand, as the theory
predicts.
However, the effects are asymmetric. The decrease in the value of a
fashion brand that females perceive, and the increase in the value of
the sportswear brand that males perceive are not matched by comparable
effects for the sportswear brand (in the case of women), or the fashion
designer brand (in the case of men). Cognitive consistency theory does
not accommodate asymmetric effects of this kind.
Perhaps the asymmetry can be explained by gender role expectation
regarding apparel preferences and shopping styles (cf. Bae & Miller,
2009; Belk, Mayer, & Driscoll, 1984; Campbell, 1997; Workman &
Studak, 2006), as well as gender differences in information processing
when shopping (Laroche, Saad, Cleveland, & Browne, 2000). Thus,
women would establish the fashion brand as the point of reference
because that is popularly deemed most gender appropriate for them, while
men would take the sports brand as their point of reference because that
is popularly deemed most gender appropriate for them. The effect of a
co-brand would depend on how evaluation of the partner brand would raise
or lower the value of the reference brand, but the same would not be
true for a brand that is not a reference brand. In other words, the
effect of co-branding may depend on which kind of brand is preferred in
the first place, rather than on simple averaging of the value for
partner brands or on aggregation of the two brands' equity. Future
research should test that possibility.
Worth, Smith, and Mackie (1992) suggest that gender self-image
affects evaluation of products described in gender-relevant terms.
Regardless of the traditional image of the product, and regardless of
the gender of the perceiver, people prefer a product that is described
in terms that matched the gender attributes that they perceived as both
characteristic of and important to themselves. This may suggest that
women respond less favorably toward fashion brands co-branding with a
sport brand because they associate sportswear brands with a masculine
user image and fashion brands with a feminine user image. On the other
hand, if this were true, it should also be the case that females would
give more positive ratings to the fashion and sportswear cobranded
product than the sportswear product without a fashion co-brand. That was
not the case here. Although gender schematicity may provide some insight
into the gender differences observed here, it may not fully predict our
findings.
Nevertheless, gender differences are certainly apparent here in the
ways that user image varies as a function of brand and co-brand. When
considered in terms of the number of significant effects, males'
image of users was far more affected than was females' image of
users. It would tempting, therefore, to suggest that males are more
sensitive to what clothing brands symbolize than are women, but that
would be inconsistent with other research which demonstrates the
opposite (e.g., Belk et al., 1984). In general, women demonstrate higher
involvement in clothing purchases than do men (O'Cass, 2000).
Consequently, their attitudes and the evaluative criteria they apply are
more highly crystallized (Belk, 1978). That may be what is demonstrated
by the findings here, as females provided highly focused evaluations of
the wearer, particularly in terms of goodness, whereas males spread
their evaluations in a far less focused manner, as would be expected in
the presence of a less crystallized evaluative framework. Further, women
did not differentiate the shirts in terms of dominance, but men did.
From the standpoint of gender role expectations, dominance is popularly
perceived to be more relevant to men than to women (Rudman & Glick,
2008)--a matter that resonates into judgment and choice when the two
genders evaluate consumption alternatives (Coleman, 2012). It is clear
that males and females interpret brand symbolisms differently,
particularly when evaluating clothing. The fact of such differences
should inform future research into perceptions of brands and brand
alliances. Such work may benefit by treating males and females as
distinct populations.
It has long been understood that price and status are interrelated
(Lichtenstein, Ridgway, & Netemeyer, 1993). It is not clear whether
price signals status, or the status indicators in a garment signal
higher value, which is then reflected in price. The user characteristics
measured here are not direct indicators of status, but they are
reflections of ways that the user would be perceived, which should be
associated with status (Izard, 1959). The small but significant effects
found here for engagement and dominance on price are consistent with
those interrelationships. In other words, the effect of perceived user
characteristics on price may be a consequence of ways that personality
and price are relate to status. That should be tested in future
research. Further, since the effect of personality on status is
independent of the effect that appearance has on status, at least for
males (Anderson, John, Keltner, & Kring, 2001), the effect of
single-branded and cobranded on ratings of appearance might also be
useful to incorporate.
The effects of engagement and dominance on price did not differ for
males and females. That is somewhat surprising given the moderating
effect of gender on mean ratings found here, and other work showing that
males and females evaluate clothing differently (Elliott, 1994;
Kamineni, 2005; Mayer & Belk, 1985). The failure to find a
difference in the test for linear restrictions may be an artifact of the
small effect that perceived user characteristics had here. If user image
dimensions are incorporated in future research, stronger measures of
user image may prove useful. Although the AVE and alphas for our
measures were acceptable, they were somewhat low. Stronger measures may
yield more portent exploration of the causes and effects of user image.
In general, our two sportswear brands and our two fashion designer
brands behaved comparably. Thus, for the most part, our findings are not
brand-dependent. Nevertheless, there was a moderate but significant
difference in the ways that the brand alliance was evaluated when Nike
was paired with a fashion designer brand than when Adidas was paired
with a fashion brand. Further the direction was different for men than
for women. The size of effect in the context of other effects
demonstrated here suggests that this effect is not substantial, and it
was not found in the case of fashion designer brands. Nevertheless, the
finding does demonstrate that brand alliances can impact different
brands differently. The basis for such differences is not well
understood, and warrants further study.
With the exception of work on licensed products, sport marketing
research has not yet concerned itself with the ways that brands
associated with sport, but that are not team or league brands, ripple
through the economy. As the example of emerging alliances between
sportswear brands and fashion designer brands demonstrates, sport
associations continue to penetrate the economy beyond sport competitions
and licensed products. The effects of such alliances are neither
straightforward nor well understood. By exploring the ways that
sport-derived associations ramify, we can better understand sport's
impact on markets and consumer behavior.
References
Aaker, D. (1996). Measuring brand equity across products and
markets. California Management Review, 38(3), 102-120.
Aaker, J. L. (1997). Dimensions of brand personality. Journal of
Marketing Research, 24, 347-356.
Abbott, M., Shackleton, J., & Holland, R. (2008). Measuring the
brand category through semantic differentiation. Journal of Product and
Brand Management, 17, 223-234.
Agarwal, M., & Rao, V. (1996). An empirical comparison of
consumer-based measures of brand equity. Marketing Letters, 7, 237-247.
Ahn, S., Kim, K. J., & Forney, J. C. (2010). Fashion
collaboration or collision? Examining the match-up effect in
co-marketing alliances. Journal of Fashion Marketing and Management, 14,
6-20.
Ailawadi, K. L., Lehmann, D. R., & Neslin, S.A. (2003). Revenue
premium as an outcome measure of brand equity. Journal of Marketing,
67(4), 117.
Anderson, C., John, O. P., Keltner, D., & Kring, A. M. (2001).
Who attains social status? Effects of personality and physical
attractiveness in social groups. Journal of Personality and Social
Psychology, 81, 116-132.
Aquirre-Rodriguez, A., Bosnjak, M., & Sirgy, M. J. (2012).
Moderators of the self-congruity effect on consumer decision-making: A
meta-analysis. Journal of Business Research, 65, 1179-1188.
Azevedo, A., & Farhangmehr, M. (2005). Clothing branding
strategies: Influence of brand personality on advertising response.
Journal of Textile and Apparel, Technology and Management, 4(3), 1-13.
Azoulay, A., & Kapferer, J. N. (2003). Do brand personality
scales really measure brand personality? Journal of Brand Management,
11, 143-155.
Bae, S., & Miller, J. (2009). Consumer decision-making styles
for sport apparel: Gender comparisons between college consumers.
International Council for Health, Physical Education, Recreation, Sport,
and Dance, 4(1), 40-45.
Belk, R. (1978). Assessing the effects of visible consumption on
impression formation. Advances in Consumer Research, 5, 39-47.
Belk, R., Mayer, R., & Driscoll, A. (1984). Children's
recognition of consumption symbolism in children's products.
Journal of Consumer Research, 10, 386-397.
Blackett, T., & Boad, B. (1999). Co-branding: The science of
alliance. New York, NY: St. Martin's Press.
Blackston, M. (1995). The qualitative dimension of brand equity.
Journal of Advertising Research, 35(4), RC2-RC7.
Blackston, M. (2000). Observations: Building brand equity by
managing the brand's relationships. Journal of Advertising
Research, 40(6), 101-105.
Blumer, H. (1969). Symbolic interactionism: Perspective and method.
Englewood Cliffs, NJ: Prentice Hall.
Braun, O. L., & Wicklund, R. A. (1989). Psychological
antecedents of conspicuous consumption. Journal of Economic Psychology,
10, 161-187.
Brewer, W. F. (2004). Charles Osgood: The psychology of language.
In L. Hoddeson (Ed.), No boundaries: University of Illinois vignettes
(pp. 210-225). Urbana, IL: University of Illinois Press.
Campbell, C. (1997). Shopping, pleasure and the sex war. In P. Falk
& C. Campbell (Eds.), The shopping experience (pp. 166-176). London,
UK: Sage.
Caprara, G. V., Barbaranelli, C., & Guido, G. (2001). Brand
personality: How to make the metaphor fit? Journal of Economic
Psychology, 22, 377-395.
Chandler, D. (2002). Semiotics: The basics. London, UK: Routledge.
Chang, W. L. (2008). A value-based pricing system for strategic
co-branding goods. Kybernetes, 37, 978-996.
Cheng, A. (2010, May 14). Adidas tones up Reebok for growth. Market
Watch. Retrieved from http://articles.marketwatch.com/2010-0514/
industries/30752868_1_reebok-adidas-group-namesake-brand
Coleman, C. A. (2012). Construction of consumer vulnerability by
gender and ethics of empowerment. In C. C. Otnes & L. T. Zayer
(Eds.), Gender, culture, and consumer behavior (pp. 3-32). New York, NY:
Taylor & Francis.
Collins, R. C. (1990). Stratification, emotional energy, and the
transient emotions. In T.D. Kemper (Ed.), Research agendas in the
sociology of emotions (pp. 27-57). Albany, NY: The State University of
New York Press.
Elliott, R. (1994). Exploring the symbolic meaning of brands.
British Journal of Management, 5, S13-S19.
Elliott, R. (1997). Existential consumption and irrational desire.
European Journal of Marketing, 31, 285-296.
Elliott, R., & Wattanasuwan, K. (1998). Brand as symbolic
resources for the construction of identity. International Journal of
Advertising, 17, 131-144.
Evans, F. B. (1968). Automobiles and self imagery: Comment. Journal
of Business, 41, 484-485.
Fennis, B. M., & Pruyn, A. T. H. (2007). You are what you wear:
The impact of brand personality on consumer impression formation.
Journal of Business Research, 60, 634-639.
Fennis, B. M., Pruyn, A. T. H., & Maasland, M. (2005).
Revisiting the malleable self: Brand effects on consumer
self-perceptions of personality traits. Advances in Consumer Research,
32, 371-377.
Forsythe, S. (1991). Effect of private, designer and national brand
name on shoppers' perception of apparel quality and price. Clothing
and Textiles Research Journal, 9(2), 1-6.
Geylani, T., Inman, J. J., & Ter Hofstede, F. (2008). Image
reinforcement or impairment: The effects of co-branding on attribute
uncertainty. Marketing Science, 27, 730-744.
Goodstein, R. C. (1993). Category-based applications and extensions
in advertising: Motivating more extensive ad processing. Journal of
Consumer Research, 20, 87-99.
Hannover, B., & Kiihnen, U. (2002). "The clothing makes
the self via knowledge activation. Journal of Applied Social Psychology,
32, 25132525.
Heise, D. R. (1970). The semantic differential and attitude
research. In G. F. Summers (Ed.), Attitude measurement (pp. 235-253).
Chicago, IL: Rand McNally.
Heise, D. R. (1989). Effects of emotion displays on social
identification. Social Psychology Quarterly, 52, 10-21.
Hirschman, E. C. (1981). Comprehending symbolic consumption: Three
theoretical issues. In E. C. Hirschman & M. B. Holbrook (Eds.),
Symbolic consumer behavior (pp. 4-6). Ann Arbor, MI: Association for
Consumer Research.
Holman, R. H. (1980). Clothing as a communication: An empirical
investigation. Advances in Consumer Research, 7, 372-377.
Holmes, S., & Tierney, C. (2002, November 4). How Nike got its
name back. Business Week, 129-131.
Izard, C. E. (1959). Personality correlates of sociometric status.
Journal of Applied Psychology, 43(2), 89-93.
Jubas, K. (2011). Shopping for identity: Articulations of gender,
race and class by critical consumers. Social Identities, 17, 319-333.
Kaiser, S. B., Nagasawa, R. H., & Hutton, S. S. (1991).
Fashion, post-modernity and personal appearance: A symbolic
interactionist formulation. Symbolic Interaction, 14, 165-185.
Kamineni, R. (2005). Influence of materialism, gender and
nationality on consumer brand perceptions. Journal of Targeting,
Measurement and Analysis for Marketing, 14, 25-32.
Katos, A., Lawler, K. A., & Seddighi, H. (2000). Econometrics:
A practical approach. London, UK: Routledge.
Keller, K. L. (1998). Strategic brand management: Building,
measuring, and managing brand equity. Upper Saddle River, NJ: Prentice
Hall.
Kemper, T. D., & Collins, R. (1990). Dimensions of
microinteraction. American Journal of Sociology, 96, 32-68.
Kim, E. Y., & Kim, Y. K. (2004). Predicting online purchase
intentions for clothing products. European Journal of Marketing, 38,
883-897.
Kinch, J. W. (1967). A formalized theory of the self-concept. In J.
G. Manis & B. N. Meltzer (Eds.), Symbolic interaction: A reader in
social psychology (pp. 232-240). Boston, MA: Allyn & Bacon.
Kroska, A. (2002). Does gender ideology matter? Examining the
relationship between gender ideology and self- and partner-meanings.
Social Psychology Quarterly, 65, 248-265.
Langford, T., & MacKinnon, N. (2000). The affective bases for
the gendering of traits: Comparing the United States and Canada. Social
Psychology Quarterly, 63, 34-48.
Laroche, M., Saad, G., Cleveland, M., & Browne, E. (2000).
Gender differences in information search strategies for a Christmas
gift. Journal of Consumer Marketing, 17, 500-522.
Lau, V., & Axelrod, N. (2009, November 24). Who wears the
clothes? Balancing brand image and customer reality. Women's Wear
Daily, p. 1.
Leigh, J. H., & Gabel, T. G. (1992). Symbolic interactionism:
Its effects on consumer behavior and implications for marketing
strategy. Journal of Consumer Marketing, 9, 27-38.
Levin, A. M., Davis, J. C., & Levin, I. P. (1996). Theoretical
and empirical linkages between consumers' responses to different
branding strategies. Advances in Consumer Research, 23, 296-300.
Levy, S. J. (1959). Symbols for Sale. Harvard Business Review,
37(4), 117-124.
Lichtenstein, D. R., Ridgway, N. M., & Netemeyer, R. G. (1993).
Price perceptions and consumer shopping behavior: a field study. Journal
of Marketing Research, 30, 232-245
Liu, F., Li, J., Mizerski, D., & Soh, H. (2012).
Self-congruity, brand attitude, and brand loyalty: A study on luxury
brands. European Journal of Marketing, 46, 922-937.
MacKinnon, N. J., & Langford, T. (1994). The meaning of
occupational prestige scores: A social psychological analysis and
interpretation. The Sociological Quarterly, 35, 215-245.
Maxwell, S. (1999). Biased attributions of a price increase:
Effects of culture and gender. Journal of Consumer Marketing, 16, 9-23.
Mayer, R., & Belk, R. (1985). Fashion and impression formation
among children. In M. Solomon (Ed.), The psychology of fashion (pp.
293-307). Lexington: MA, Lexington Books.
McCarthy, M., & Norris, D. (1999). Improving competitive
position using branded ingredients. Journal of Product and Brand
Management, 8, 267-283.
McCracken, G. (1988). Culture and consumption: New approaches to
the symbolic character of consumer goods and activities. Bloomington,
IN: Indiana University Press.
McCracken, G. (1993). The value of the brand: An anthropological
perspective. In D. Aaker & A. Biel (Eds.), Brand Equity and
Advertising (pp. 125-142). Hillsdale, NJ: Lawrence Erlbaum Associates.
McCracken, G., & Roth, V. (1989). Does clothing have a code?
Empirical findings and theoretical implications in the study of clothing
as a means of communication. International Journal of Research in
Marketing, 6, 13-33.
Mead, G. H. (1934). Mind, self, and society. Chicago, IL:
University of Chicago Press.
Meek, A. (1997). An estimate of the size and supported economic
activity of the sports industry in the United States. Sport Marketing
Quarterly, 6, 15-21.
Mehrabian, A. (1980). Basic dimensions for a general psychological
theory: Implications for personality, social, environmental and
developmental studies. Cambridge, MA: Oelgeschlager, Gunn & Hain.
Milano, M., & Chelladurai, P. (2011). Gross domestic sport
product: The size of the sport industry in the United States. Journal of
Sport Management, 25, 24-35.
Mitchell, R. (2005, May 9). Is fashion design a team sport? Brand
Channel. Retrieved from
http://www.brandchannel.com/features_effect.asp?pf_id=262
Nueno, J. L., & Quelch, J. A. (1998). The mass marketing of
luxury. Business Horizons, 41(6), 61-68.
O'Cass, A. (2000). An assessment of consumers product,
purchase decision, advertising and consumption involvement in fashion
clothing. Journal of Economic Psychology, 21,545-576.
Osgood, C. E., May, W. H., & Miron, M. S. (1975).
Cross-cultural universals of affective meaning. Urbana, IL: University
of Illinois Press.
Osgood, C. E., Suci, G. J., & Tannenbaum, P. H. (1957). The
measurement of meaning. Urbana, IL: University of Illinois Press.
Parker, B. (2009). A comparison of brand personality and brand
user-imagery congruence. Journal of Consumer Marketing, 26, 175-184.
Peluchette, J., & Karl, K. (2007). The impact of workplace
attire on employee self perceptions. Human Resource Development
Quarterly, 18, 354360.
Persson, N. (2010). An exploratory investigation of the elements of
B2B brand image and its relationship to price premium. Industrial
Marketing Management, 39, 1269-1277.
Rose, A. M. (Ed.). (1962). Human behavior and social processes: An
interactionist approach. Boston, MA: Houghton-Mifflin.
Robertson, T., & Kassarjian, H. H. (1991). Handbook of consumer
behavior. Englewood Cliffs, NJ: Prentice-Hall.
Rudman, L. A., & Glick, P. (2008). The social psychology of
gender: How power and intimacy shape gender relations. New York, NY:
Guilford Press.
Schewe, C. D. (1973). Selected social psychological models for
analyzing buyers. Journal of Marketing, 37, 31-39.
Schneider. A., & Schroder, T. (2012). Ideal types of leadership
as patterns of affective meaning: A cross-cultural and over-time
perspective. Social Psychology Quarterly, 75, 268-287.
Seltman, H. J. (2012). Experimental design and analysis.
Pittsburgh, PA: Carnegie-Mellon University.
Sirgy, M. J. (1982). Self-concept in consumer behavior: A critical
review. Journal of Consumer Research, 9, 287-300.
Sirgy, M. J. (1983). Social cognition and consumer behavior. New
York, NY: Praeger.
Sirgy, M. J. (1986). Self-congruity: Toward a new theory of
personality and cybernetics. New York, NY: Praeger.
Sirgy, M. J. (2012). The psychology of quality of lifehedonic
well-being, life satisfaction, and eudaimonia. Dordrecht, NL: Springer.
Simonin, B. L., & Ruth, J. A. (1995). Bundling as a strategy
for new product introduction: Effects on consumers' reservation
prices for the bundle, the new product, and its tie-in. Journal of
Business Research, 33, 219230.
Simonin, B. L., & Ruth, J. A. (1998). Is a company known by the
company it keeps? Assessing the spillover effects of brand alliances on
consumer brand attitudes. Journal of Marketing Research, 35, 30-42.
Snider, J. G., & Osgood, C. E. (Eds.). (1969). Semantic
differential technique: A sourcebook. Chicago, IL: Aldine.
Solomon, M. R. (1983). The role of products as social stimuli: A
symbolic interactionism perspective. Journal of Consumer Research, 10,
319-329.
Sproles, G. B. (1981). The role of aesthetics in fashion-oriented
consumer behavior. In G.B. Sproles (Ed.), Perspectives of fashion (pp.
120-127). Minneapolis, MN: Burgess.
Stein, N. L., & Trabasso, T. (1982). What's in a story? An
approach to comprehension. In R. Glaser (Ed.), Advances in the
psychology of instruction (Vol. 2, pp. 213-268). Hillsdale, NJ: Erlbaum.
Stith, M. T., & Goldsmith, R. E. (2006). Race, sex, and fashion
innovativeness: A replication. Psychology & Marketing, 6, 249-62.
Swartz, T.A. (1983). Brand symbols and message differentiation.
Journal of Advertising Research, 23(5), 59-64.
Thomas, J. (2010, January 20). Armani and Reebok to roll out
co-branded clothing. Marketing Magazine. Retrieved from
http://www.marketingmagazine.co.uk/news/978635/Armani-Reebok-roll-co-branded-clothing/
Valentine, V., & Evans, M. (1993). The dark side of the onion:
Rethinking "rational" and "emotional" responses.
Journal of the Market Research Society, 35, 125-143.
Venkatesh, R., & Mahajan, V. (1997). Products with branded
components: An approach for premium pricing and partner selection.
Marketing Science, 16, 146-165.
Venkatesh, R., & Mahajan, V. (2009). Design and pricing of
product bundles: A review of normative guidelines and practical
approaches. In V.
R. Rao (Ed.), Handbook of pricing research in marketing (pp.
232-257). Northampton, MA: Edward Elgar.
Vercillo, K. (2009, December). Top ten American fashion designers.
Hub Pages. Retrieved from
http://kathrynvercillo.hubpages.com/hub/Top10-American-Fashion-Designers
Washburn, J., Till, B., & Priluck, R. (2000). Co-branding:
Brand equity and trial effects. Journal of Consumer Marketing, 17,
591-604.
Westerink, J., Krans, M., & Ouwerkerk, M. (Eds.). (2011).
Sensing emotions in context: The impact of context on behavioral and
physiological experience measurements (Vol. 12). Berlin, DE: Springer.
Wicklund, R. A., & Gollwizter, P. M. (1982). Symbolic
self-completion. Hillsdale, NJ: Erlbaum.
Workman, J. E., & Studak, C. M. (2006). Fashion consumers and
fashion problem recognition style. International Journal of Consumer
Studies, 30, 75-84.
Worth, L. T., Smith, J., & Mackie, D. M. (1992). Gender
schematicity and preference for gender-typed products. Psychology &
Marketing, 9, 17-30.
Wright, N. D., Claiborne, C. B., & Sirgy, M. J. (1992). The
effects of product symbolism on consumer self-concept. Advances in
Consumer Research, 19, 311-318.
D. Gloria Wu, PhD candidate, is a teaching assistant in sport
management in the Department of Kinesiology and Health Education at the
University of Texas at Austin. Her research interests include sport
branding and sport event design.
Laurence Chalip, PhD, is a professor and chair of the Department of
Recreation, Sport and Tourism at the University of Illinois at
Urbana-Champaign. His research focuses on sport policy.
Table 1
Measurement Model for User Image Scale
Standardized
Factor Loadings
Goodness Alpha = .74
Deceitful--Trustworthy 0.54
Unreliable--Dependable 0.73
Disreputable--Respectable 0.74
Engagement Alpha = .70
Laid back--Assertive 0.58
Playful--Serious 0.57
Easy-going--Competent 0.67
Dominance Alpha = .71
Weak--Powerful 0.79
Submissive--Influential 0.82
Apathetic--Ambitious 0.51
N = 275.
Table 2
Correlations Among Dependent Variables
Goodness Engagement Dominance Price
Goodness 1
Engagement 0.282 ** 1
Dominance 0.412 ** 0.227 ** 1
Price 0.052 0.186 ** 0.212 ** 1
** p < .01
Table 3
Planned Contrasts for Male Participants on User Image and
Expected Price
Research Value of SE 101) Cohen's t
Question Contrast
Goodness 1 1.69 2.05 0.68 0.26
2 1.73 0.54 10.34 ** 1.06
3 -1.41 0.73 3.73 0.43
4 0.32 0.85 0.14 0.10
5 -3.14 0.96 10.75 *** 0.96
6 -0.98 0.50 3.92 * 0.60
7 -0.68 0.50 1.90 0.42
Engagement 1 5.75 2.29 6.32 * 0.74
2 2.15 0.60 12.89 *** 1.11
3 -1.31 0.82 2.59 0.34
4 0.84 0.95 0.78 0.22
5 -3.46 1.07 10.52 ** 0.89
6 -0.10 0.55 0.03 0.05
7 -1.24 0.55 5.04 * 0.64
Dominance 1 3.10 2.02 2.36 0.45
2 1.37 0.53 6.71 * 0.79
3 -1.70 0.72 5.60 * 0.49
4 -0.33 0.84 0.16 0.10
5 -3.07 0.94 10.63 * 0.89
6 0.04 0.49 0.01 0.02
7 -1.13 0.49 5.41 * 0.65
Expected Price 1 24.34 4.13 34.68 *** 1.78
2 3.87 1.08 12.74 *** 1.13
3 1.87 1.48 1.61 0.27
4 5.74 1.72 11.11 *** 0.84
5 -2.00 1.93 1.07 0.29
6 0.52 1.00 0.27 0.15
7 -0.75 1.00 0.57 0.22
* p < .05; ** p < .01; *** p < .001
Table 4
Planned Contrasts for Female Participants on User Image and
Expected Price
Research Value of SE 156) Cohen's t
Question Contrast
Goodness 1 -2.76 1.43 3.74 0.48
2 0.62 0.34 3.20 0.43
3 -1.78 0.48 13.99 *** 0.62
4 -1.17 0.59 3.98 * 0.35
5 -2.40 0.59 16.48 *** 0.41
6 0.23 0.33 0.48 0.16
7 0.95 0.33 8.28 ** 0.66
Engagement 1 2.69 1.86 2.10 0.37
2 1.18 0.45 6.97 ** 0.66
3 -1.00 0.62 2.61 0.28
4 0.18 0.76 0.06 0.05
5 -2.19 0.77 8.08 ** 0.61
6 -0.48 0.43 1.27 0.27
7 -0.01 0.43 < .001 0.01
Dominance 1 1.71 1.70 1.00 0.26
2 -0.22 0.41 0.28 0.12
3 -0.40 0.57 0.49 0.13
4 -0.62 0.70 0.78 0.19
5 -0.18 0.71 0.06 0.05
6 -0.52 0.39 1.74 0.31
7 0.98 0.39 6.25 * 0.59
Expected Price 1 20.86 3.12 44.61 *** 1.68
2 2.55 0.76 11.37 *** 0.82
3 -0.52 1.05 0.25 0.08
4 2.02 1.28 2.49 0.33
5 -3.07 1.30 5.61 * 0.49
6 -0.61 0.72 0.70 0.20
7 0.38 0.72 0.28 0.13
* p < .05; ** p< .01; *** p < .001
Table 5
Regression Results
Dimension [R.sup.2] F(3, 271) B
Expected .07 6.821 **
Price
Goodness -.182
Engagement .314
Dominance .446
SE Beta t(271)
Expected
Price
.150 -.080 -1.212
.120 .161 2.623 **
.139 .209 3.218 ***
** p < .01; *** p <.001
