Content uploaded by Gregg Bennett
Author content
All content in this area was uploaded by Gregg Bennett on Sep 19, 2016
Content may be subject to copyright.
734
Journal of Sport Management, 2008, 22, 734-761
© 2008 Human Kinetics, Inc.
Exploring Brand Positioning in a
Sponsorship Context: A Correspondence
Analysis of the Dew Action Sports Tour
Mauricio Ferreira
Texas A&M University
Todd K. Hall
Georgia Southern University
Gregg Bennett
Texas A&M University
In this study, we used correspondence analysis (Greenacre, 1984; Hoffman & Franke,
1986) to examine connections between the title sponsor, brand competitors, and con-
sumer targets exposed to a sponsorship. Demographic characteristics and self-reported
use of 20 soft drink brands were collected from 1,138 attendees of four of the ve
inaugural events of the Dew Action Sports Tour. The analyses consisted of decompos-
ing the cross-tabulated data into latent dimensions and graphically portraying brands
and consumer targets in joint preference maps. Results revealed that consumers dif-
ferentiated the 20 soft drink brands based on two latent dimensions: energy/diet and
convenience. Furthermore, based on proximity of the target market to the title sponsor
in the maps, it appears that Mountain Dew has been relatively effective in positioning
the brand for key target markets in only one of the four cities examined. Theoretical
and managerial implications of the ndings are discussed.
Over the past 20 years, corporate sponsorship has been one of the fastest
growing forms of promotional activity worldwide (IEG, 2004). One of the reasons
for this growth is the opportunity sponsorships provide to cut through the clutter
of traditional media by offering access to well-dened target markets (Gardner &
Shuman, 1987). In fact, corporations seek to associate with properties containing
an attractive consumer base for their products. This product-consumer t is impor-
tant to corporations as it facilitates the activation of the sponsorship, allows the
user imagery of the property to potentially “transfer” to the sponsor products
(Gwinner, 1997), and provides an opportunity to increase product usage among
those consumers who are of interest to sponsors.
Ferreira and Bennett are with the Dept. of Health and Kinesiology, Texas A&M University, College
Station, TX 77845. Hall is with the Dept of Hospitality, Tourism, Family & Consumer Science,
Georgia Southern University, Statesboro, GA 30460.
Brand Positioning in a Sponsorship Context 735
Despite the importance of the t between sponsoring brands and consumers
of the event (Pracejus & Olsen, 2004), very few sponsorship studies have focused
on how brands are positioned in regard to intended audience and their competi-
tors. Consistent with the notion of brand differentiation, it is paramount that brand
managers involved in sponsorships evaluate whether their marketing efforts are
reaching the intended audience.
The purpose of this study, therefore, was to examine the association between
consumer targets and brand consumption for the title sponsor of the Dew Action
Sports Tour. Although this examination can be accomplished in many different
ways, we particularly used correspondence analysis as a positioning analysis tool
to investigate connections between the event sponsor, competitors, and consumers
exposed to the sponsorship. An underlying premise of this inquiry is that sponsor-
ship as a positioning strategy can enhance brand preference among specic, tar-
geted consumer groups. Given that exposure to sponsorship takes place, the degree
to which targeted groups are associated with the consumption of sponsoring
brands indicates how well the sponsorship efforts enhanced brand use among the
targeted group.
This paper makes a specic contribution to the sport management literature
by providing a better understanding of the role sponsorship can play in position-
ing a sponsoring brand to key target markets. This contribution is particularly
accomplished by demonstrating how correspondence analysis can be used and
interpreted to examine the positioning of the Mountain Dew brand through action
sports sponsorships.
Background
Brand Positioning
Brand positioning refers to “a relatively stable set of consumer perceptions (or
meanings) of a brand in relation to competitive alternatives” (Kates & Goh, 2003,
p. 59). A few well-known examples of brand positioning include the common
consumer perceptions of Visa as a widely-accepted credit card, the perceptions of
Apple computers as user-friendly, and the perceptions of the Dallas Cowboys as
America’s team. These are examples of brands that have a distinct position in the
consumers’ minds. Conceptually, brand positioning inuences consumer choice
by “shap[ing] and maintain[ing] a specic brand’s image in the consumer’s mind”
(Green & Muller, 2002, p. 180).
The term positioning was rst introduced by Al Ries and Jack Trout mainly
as a communications tool to breakthrough the clutter of a crowded marketplace
(Ries & Trout, 1986). However, positioning inuences and is inuenced by all
marketing variables, not just communications (Aaker & Shansby, 1982; Bagozzi,
Rosa, Celly, & Coronel, 1998; Kardes, 1998). Although the way brands are pre-
sented to consumers is important, positioning usually follows a set of integrated
strategies to create brand differentiation in the consumers’ mind (Bagozzi et al.,
1998; Kardes, 1998). For example, in an effort to x a new image as a “safety
leader brand without a luxury-car price” into consumers’ minds, Honda made
changes to their marketing actions that were consistent with the intended image.
This entailed not only communicating the intended image, but also providing
safety features in every car and not altering price (Halliday, 2004).
736 Ferreira, Hall, and Bennett
The crucial element to positioning is differentiation. There are many ways in
which a brand can be perceived to be different, including among others the attri-
butes/benets it provides, its price, whether it is a pioneer brand, when it is con-
sumed, and who consumes the brand (Aaker & Shansby, 1982). One particular
strategy that is relevant to the current study is positioning by the product user
(Aaker & Shansby, 1982). This positioning strategy attempts to associate the
product with users or user groups. Many product categories such as soft drinks,
beers, cosmetics, and personal hygiene products employ such strategy. For exam-
ple, a recent marketing campaign employed by Pepsi highlighted the brand’s Diet
Pepsi line extension in an attempt to appeal to a specic consumer segment over
the age of 35 years old (Lippert, 2003). More recently Miller High Life endeav-
ored to reinforce or reclaim its position with the blue-collar workforce by creating
commercials wherein their own blue-collar workers took Miller High Life product
away from trendy restaurants and night clubs, rather than delivering it (Mullman,
2006). Regardless of the positioning strategy employed, an important objective of
brand positioning efforts is to increase the consumption of a particular product or
brand among a specic segment of consumers.
Brand Positioning and Sponsorship
One avenue for creating product differentiation through positioning strategy is to
employ sponsorship. Sponsorship is “an investment, in cash or in kind, in an activ-
ity, in return for the exploitable commercial potential associated with that activ-
ity” (Meenaghan 1991, p. 36). Marketers have increasingly used sponsorship as a
promotional and positioning tool to reach targeted markets due to its unique abil-
ity to cut through the clutter of traditional media (Cliffe & Motion, 2005; Gardner
& Shuman, 1987; Nicholls, Roslow, & Dublish, 1999). Among the many pro-
posed benets and/or objectives of corporate sponsorship is the ability to create an
emotional attachment with the target audience, which can be difcult to accom-
plish via traditional media. In essence, sponsorship allows the corporation to
approach the target audience through a medium in which they are already inter-
ested. In return, companies who invest in sponsorships may experience high levels
of preference among those who attend a sponsored event.
While sponsorship has many varying objectives including creating aware-
ness, building goodwill, and enhancing corporate image and brand equity (Gard-
ner & Shuman, 1987; Meenaghan, 1991; Pracejus, 2004), it is often used as a
communication tool to differentiate and position a brand apart from its competi-
tors. For instance, in July of 2007, PepsiCo sponsored “Live Earth,” a 24-hr series
of concerts held around the globe to support the ght against global warming
(Hein, 2007a). Not only did the sponsoring of such a cause enable PepsiCo to
reach specic targets, but it also allowed PepsiCo to borrow from the image of
being “green” or an environmentally conscious organization, in an attempt to dif-
ferentiate from their primary competitor, Coca-Cola.
Exploring Event and Sponsor Fit
When deciding on what sponsorships best t the corporation’s goals, rms often
employ many criteria to decide what to sponsor (IEG, 1999). However, brand
Brand Positioning in a Sponsorship Context 737
image and audience compatibility are often required to assess the viability of a
sponsorship engagement (Howard & Crompton, 2004). Keller (1993) dened
brand image as “perceptions about a brand as reected by the brand associations
held in memory” (p. 3). According to Keller (1993), brand images are formed by
three types of brand associations held in memory: attributes, benets, and brand
attitudes. In the case of a soft drink, for example, we might form brand attitudes
toward a brand like Mountain Dew based on its product-related attributes (sugar
content, caffeine, taste, color) or nonproduct related attributes such as price, pack-
aging, as well as the perceptions we may have of those who consume Mountain
Dew (e.g., trendy, cool, hip-hop) and usage occasions (e.g., evenings, at events,
etc.). Brand positioning through sponsorships is intended to function in a way that
the brand image of sponsors is leveraged or facilitated by the event image. The
success of the leverage may depend on the degree of congruence between the
brand and event images (Keller, 1993). Studies of sponsorship, branding, and
endorsements have supported this notion of brand image t, where high levels of
congruence may enhance consumer attitudes toward rms (Keller & Aaker, 1992;
Speed & Thomson, 2000).
In addition to brand and event image t, rms also seek brand and audience
t. That is, rms are often interested in sponsorship opportunities that show a
match between attendees or participants of an event and the consumers of the
brand (Cornwell, Weeks, & Roy, 2005; Howard & Crompton, 2004). Sociodemo-
graphic variables such as age and gender often play an important role in audience
t because of the reliance by both the companies and sport properties on these
variables to target and sell their products (Fennell & Allenby, 2004; Fennell,
Allenby, Yang, & Edwards, 2003). Brand and audience t in a sponsorship rela-
tionship is particularly important because the promotional strategy is aimed at
specic consumer segments and can be a part of a positioning strategy, especially
positioning by the product user, as discussed above. Sponsorships offer compa-
nies the opportunity to build goodwill (Meenaghan, 1991, 2001) among those
who are current (conrmed) or potential brand consumers. Action sports sponsor-
ships, for example, give companies access to teenagers and young adults.
In our view, sponsorship decision making can be complex and often requires
market research information to minimize the risks of making erroneous marketing
decisions. The assessment of the relationship between target markets and brand
use is one legitimate way to evaluate current efforts, especially if target markets
are considered as an important element of sponsorship decision making. There-
fore, a brand positioning analysis can be helpful to inform post-sponsorship evalu-
ative decisions such as renewals, new strategies, and brand portfolio
management.
As companies engage in sponsorships, important questions to ask are those
that can help managers better understand brand consumption and its relationships
with intended target markets and competition. For example, questions such as
whether the intended targeted markets have been persuaded through the sponsor-
ship and how well sponsors are positioned or differentiated relative to competition
can provide valuable information for future decisions regarding sponsorships.
Mountain Dew Sponsorships and Target Markets. The Mountain Dew brand is
currently the title sponsor of the Dew Action Sports Tour (DAST), which began its
738 Ferreira, Hall, and Bennett
inaugural event series in 2005 with ve events in Louisville, Denver, Portland,
San Jose, and Orlando. Each tour stop was sponsored by a major rm attempting
to connect with the youth market, including Panasonic, Toyota, Vans, Right Guard,
and Sony’s PlayStation brand. The event series, which is owned and broadcast by
the National Broadcast Company (NBC), provided action sports athletes with an
annual tour championship much like the PGA Tour and Nextel NASCAR series
(Bennett & Lachowetz, 2004).1 Action sports athletes compete in various sports
(BMX—Dirt, Vert, and Park, Skateboarding—Vert and Park, and Freestyle Moto-
cross) at each event, accumulating points resulting with a champion in each sport
at the end of each season.
Before this event series, there were a handful of events for action sports ath-
letes, the most notable being ESPN’s X Games. The X Games are the marquee
action sports event, having been in existence for over a decade. The X Games have
been a staple for ESPN in regard to broadcasting hours, and they have helped fuel
the growth of action sports participation among the youth segment (Bennett &
Lachowetz, 2004). The Dew Action Sports Tour is NBC’s response to the contin-
ued growth of action sports and market share enjoyed by ESPN.
As the fastest growing soft drink in terms of sales in the 1990s (Gladden &
McDonald, 2006), the Mountain Dew brand has used action sports to position its
core product (as well as the product extensions Diet Mountain Dew and MD Amp)
as a youthful, high energy, high action brand (Holt, 2003). In using action sports
as a tool to accomplish its positioning strategy, Mountain Dew is targeting the
youth market, males ages 10–24 (Wade Martin personal interview as cited by
Gladden & McDonald, 2006), which consists of more than 58 million consumers
(McCarthy, 2001).
As Mountain Dew uses action sports sponsorship as a tool to communicate
with the youth male market, this study assesses the potential associations between
Mountain Dew, consumption, and target markets. In addition, this study also
examines how well Mountain Dew is positioned or differentiated relative to com-
petition. Therefore, the following questions are explored in this investigation:
R1: Are Dew Action Sports Tour attendees overall (and for each city of the
tour) more likely to consume Mountain Dew versus other brands?
R2: Does a relationship between brand consumption and different target mar-
kets exist (overall and for each city of the tour)?
R3: How is Mountain Dew positioned relative to competing brands in key
target markets (males 24 years of age and younger) overall and for each city
of the tour?
Method
Respondents
Respondents (n = 1138) were attendees at four of the ve events of the Dew
Action Sports Tour, as data were not collected at one tour stop (Denver). Data
were collected onsite between noon and 8:00 p.m. over the three days when the
events were held in June, August, September, and October of 2005. Mall intercept
Brand Positioning in a Sponsorship Context 739
methodology was employed by trained collectors at all four locations. In each
city, there were two to four collectors near an information booth located in a cen-
tral, high-trafc area in close proximity to the only entrance to the various events,
and ve to nine collectors within the individual arenas while waiting for an event
to commence. All collectors were instructed to work in pairs to solicit every
respondent (not only those they thought were more likely to ll out a survey) pass-
ing through their strategic points. The survey took less than ten minutes to com-
plete and respondents were offered the incentive of a prize drawing for their par-
ticipation. The prizes varied by city and ranged from a bicycle to a mini
motorcycle.
Table 1 details the demographic prole of the entire sample and displays the
demographic differences between each city. The overall sample consisted of a
predominance of Caucasian (75%) males (68%). Members of Generation Y (ages
12–24) comprised 50% of the sample. To identify demographic differences across
cities, a Chi-square analysis was conducted between each demographic character-
istic and cities. The Chi-square adjusted residuals were then used to determine
values that show signicant relationship beyond what would be expected by
chance alone. As shown in Table 1, the Chi-square values indicate that the cities
differed signicantly on all demographic characteristics. According to the adjusted
residuals, the most remarkable differences include the following: San Jose was
comprised of more Hispanics (37%), more males (77%), more teenagers 17 years
of age and under (46%), and more respondents still in middle/high school (44%)
than expected. Orlando was also more diverse than expected, with Hispanics com-
prising 15% of the sample. Moreover, Orlando had more females (37%) and older
respondents in the 37+ years of age group (36%) than expected. Portland was
comprised of more teenagers under 17 years of age (44%) and more Caucasians
(82%) than expected. Finally, Louisville was comprised of more females (35%),
more individuals in the older 37+ years of age group (30%), more Caucasians
(91%), and more individuals with very high income, $85,000 or higher (18%)
than expected.
Measures
To address the research questions of this study, a questionnaire was constructed
and administered to the sample as described above. The questionnaire contained
items related to involvement with action sports, consumption of soft drinks, and
demographic segmentation, including gender and age.
Target Market Composite Characteristics. In the demographic section of the
questionnaire, respondents were asked to provide their gender and age at the time
of the event. The gender measure was a dichotomous variable, which was chosen
by respondents (1 = female; 0 = male), while the age measure required an open-
ended response. The respondents were then categorized into four groups of rela-
tively1 equal size according to quartiles (bands of 25 percentile frequencies) across
the entire database: ages 17 and under, ages 18–24, ages 25–36, and ages 37 and
older.
Because our interest here was mainly to determine the relationship between
brand consumption and target groups, and not the relationship between demo-
graphic variables, a composite variable representing target groups was created
740
Table 1 Demographic Characteristics
Portland San Jose Orlando Louisville Overall
(n=203) (n=217) (n=166) (n=552) (n=1138) Chi-Square df p
Gender
Male 69% 77% 63% 65% 68%
Female 31% 23% 37% 35% 32% 250 3 <.001
Age
<Under 17 yrs old 44% 46% 10% 25% 30%
18-24 16% 18% 30% 19% 20%
25-31 11% 10% 12% 14% 13%
32 + years 29% 26% 48% 42% 37% 1,870 9 <.001
Ethnicity
African American 3% 4% 3% 4% 3%
Asian or Asian American 4% 8% 1% 1% 3%
Hispanic 5% 37% 15% 2% 10%
Native American 2% 5% 1% 1% 2%
White, not Hispanic 82% 44% 77% 91% 75%
Other 5% 3% 3% 1% 2% 5,236 15 <.001
Education
Enrolled in middle/high school 43% 44% 14% 24% 30%
HS grad 16% 19% 19% 20% 19%
Trade/tech diploma 2% 4% 6% 5% 4%
(continued)
741
Table 1 (continued)
Portland San Jose Orlando Louisville Overall
(n=203) (n=217) (n=166) (n=552) (n=1138) Chi-Square df p
Some college 14% 16% 23% 24% 21%
2-year degree 6% 5% 6% 7% 6%
College grad 9% 7% 24% 14% 13%
Graduate degree 8% 3% 5% 5% 5%
Professional degree 2% 1% 3% 2% 2% 1,648 21 <.001
Income
<$15,000 17% 35% 19% 14% 19%
$15,000 - $24,999 16% 9% 8% 13% 12%
$25,000 - $39,999 19% 12% 21% 19% 18%
$40,000 - $59,999 23% 13% 25% 23% 22%
$60,000 - $84,999 13% 13% 17% 14% 14%
$85,000 + 11% 19% 10% 18% 16% 1,061 15 <.001
742 Ferreira, Hall, and Bennett
using the individual variables age and gender. The composite variable was cre-
ated, in which each value of the new variable represented an age by gender com-
bination. Because age and gender have four and two levels respectively, the new
variable had eight values. This new composite variable with eight targets sufces
for separate analysis within each city. To allow a comparison of these eight target
groups across the four targeted cities, a composite variable with 32 groups (eight
target groups x four cities) were also created. These composite variables not only
allow the focus of the analysis to be on the relationship between brands and tar-
gets (as opposed to individual demographic characteristics), but also it allows a
simple correspondence analysis to be used.
Self-Reported Brand Use. Survey respondents were asked to identify which, if
any soft drink brands, from a list of 20, they had consumed in the previous two-
week period. The soft drinks listed represent a variety of popular cola, noncola,
diet, sport, and energy brands.
Using Correspondence Analysis to Explore Audience
and Sponsor Fit
Chi-square and correspondence analyses were conducted to examine the relation-
ship between brand consumption and target groups who participated in the survey.
Correspondence analysis (Greenacre, 1984; Hoffman & Franke, 1986) is an
exploratory data reduction method designed to portray cross-tabulated categorical
data in a joint space map. The relationship between categorical variables is uncov-
ered by factoring and decomposing categorical data into latent dimensions, simi-
lar to principal components analyses. The results of the analysis are graphically
portrayed in a low-dimensional map where similar objects (described by rows and
columns as points) are plotted close together and different objects (rows and col-
umns) are plotted relatively far apart. This graphical portrayal of both rows and
columns in the same map is particularly useful in revealing the nature of relation-
ship (or correspondence) between the objects (Bendixen, 1996). For tables of
three or more variables, either multiple variables can be combined into a single
composite variable (the approach used in this paper) or a generalized multivariate
extension of correspondence analysis, Multiple Correspondence Analysis (MCA),
can be employed (Kaciak & Louviere, 1990).
Correspondence analysis is often applied as a compositional or attribute-
based approach, where respondents are asked to make direct associations between
products and attributes using questions of “pick-any” type (Arimond & Elfessi,
2001; Hair, Anderson, Tatham, & Black, 1998). Many marketing and tourism-
related studies have particularly employed correspondence analysis to understand
the relationship between products, brands, or destinations, and their correspond-
ing features or attributes (Arimond & Elfessi, 2001; Calantone, Benedetto, Hakam,
& Bojanic, 1989; Hoffman & Franke, 1986; Kaciak & Louviere, 1990). However,
correspondence analysis can also be used as a decompositional or attribute-free
approach when product attributes are inferred from an analysis of cases-by-brands
matrices of nonnegative data such as brand usage data (Elrod, 1991; Elrod &
Winer, 1992). Decompositional approaches consist of measuring perceptions
indirectly in such a way that they are inferred from the choices, brand uses, prefer-
Brand Positioning in a Sponsorship Context 743
ence rankings, or similarity judgments individuals make regarding brands. The
most common decompositional approaches are associated with multidimensional
scaling (MDS) techniques, which are often applied to judgment data (similarity or
preferences) at the individual or group level (Hair et al., 1998). Martin (1994), for
example, used similarity comparisons between sports to create a perceptual map
of the consumer’s sport schema using MDS techniques. In addition, Milne,
McDonald, Sutton, and Kashyap (1996) examined participation patterns using
transformed frequency data as input for a MDS analysis. Correspondence analysis
was particularly used in this paper as a decompositional approach by using self-
reported brand usage data as the sampling unit for contingency tables between
brands and targets.
As a data reduction technique, correspondence analysis offers at least three
advantages over other methods. First, unlike any other data reduction method,
correspondence analysis is particularly useful in identifying how categorical vari-
ables are related by yielding a simultaneous graphical portrayal of two-way or
multiway contingency tables. Second, the method is very exible regarding data
requirements, where input data can be simply dichotomous (e.g., yes/no) data,
which are easier to obtain and useful when many variables (e.g., brands) are used
in a study. Finally, when used as a decompositional approach to examining self-
reported or actual brand usage data, the results derived from correspondence anal-
ysis reveal a map that best reects consumer preference. However, one disadvan-
tage of this approach is that the interpretation of the dimensions may not be simple,
relying on researchers’ previous knowledge and judgment. Further technical dis-
cussion regarding correspondence analysis can be found in articles by Carroll,
Green, and Schaffer (1987), Greenacre (1984, 1989), and Hoffman and Franke
(1986).
In this study, simple correspondence analysis was rst performed on contin-
gency tables, which were formed by tabulating the frequency of self-reported
brand use of twenty soft drink brands by eight consumer targets for each city sepa-
rately (a total of four maps). Each cell of the contingency table contained the
number of afrmative responses to the brand consumption question regarding 20
brands (e.g., dichotomous variables that indicate whether individuals consumed a
brand in the last two weeks). To depict an overall relationship between brands and
targets while accounting for variation across cities, a fth preference map was
developed by including a composite variable that incorporates location, yielding
32 levels (gender × age × cities). All correspondence analyses were performed
using PROC CORRESP in SAS 9.1. For the sake of saving space, we have only
reported the map of Louisville in detail and briey discuss the overall map2, which
contains the data from all four cities. They will be discussed in the next section.
Results
Chi-square Analysis—Relationship between Brand
and Consumption
Tables 2, 3, and 4 contain the frequency of individual responses to the brand con-
sumption question for each brand and for each of the four host cities. Chi-square
744
Table 2 Frequency of Soft Drink Consumption
Portland Consumption San Jose Consumption
(n=203) (n=217)
No Yes No Yes
Soft Drinks Count % Count % Count % Count %
Mountain Dew 76 37% 127 63% 100 46% 117 54%
Diet Mountain Dew 190 94% 13 6% a 195 90% 22 10% b
Coke 100 49% 103 51% 118 54% 99 46%
Diet Coke 177 87% 26 13% a 183 84% 34 16% b
Pepsi 73 36% 130 64% 83 38% 134 62%
Diet Pepsi 171 84% 32 16% a 185 85% 32 15% b
Dr. Pepper 101 50% 102 50% 109 50% 108 50%
Diet Dr. Pepper 179 88% 24 12% a 184 85% 33 15% b
Gatorade 80 39% 123 61% 85 39% 132 61%
Powerade 150 74% 53 26% a 155 71% 62 29% b
Sprite 114 56% 89 44% a 121 56% 96 44%
7-Up 143 70% 60 30% a 166 76% 51 24% b
Sierra Mist 122 60% 81 40% a 128 59% 89 41%
Starbucks Double Shot 156 77% 47 23% a 155 71% 62 29% b
Sobe 113 56% 90 44% 132 61% 85 39%
Adrenaline Rush 166 82% 37 18% a 176 81% 41 19% b
Arizona RX 176 87% 27 13% a 171 79% 46 21% b
Snapple 130 64% 73 36% a 153 71% 64 29% b
Amp (Mountain Dew) 145 71% 58 29% a 167 77% 50 23% b
Red Bull 141 69% 62 31% a 143 66% 74 34% b
Column Totals 2703 67% 1357 33% a 2909 67% 1431 33% b
Brand Positioning in a Sponsorship Context 745
analyses indicate a signicant relationship between brands and consumption cat-
egories (yes and no) in Portland, 2 (19, N = 203) = 561.820, p < .001, in San Jose,
2 (19, N = 217) = 467.468, p < .001, in Orlando, 2 (19, N = 116) = 492.573, p <
.001, in Louisville, 2 (19, N = 552) = 1399.234, p < .001, and across all cities, 2
(19, N = 1138) = 2717.76, p < .001. Adjusted Chi-square residual values for each
city and across all cities revealed a higher than expected self-reported consump-
tion for Mountain Dew (residual values ranging from 6.733 to 19.353, all p <
.001), Coke (residual values ranging from 4.067 to 9.148, all p < .001), Pepsi
(residual values ranging from 9.063 to 14.105, all p < .001), Dr. Pepper (residual
values ranging from 2.706 to 9.731, all p < .001), Gatorade (residual values rang-
ing from 8.419 to 13.230, all p < .001), Sprite (residual values ranging from 3.229
to 4.704, all p < .001), and Sierra Mist (residual values ranging from 1.072 to
3.735, all p < .05, except in Orlando). Nevertheless, Chi-square residuals indi-
cated a signicant lower than expected consumption of all diet soft drinks (resid-
ual values ranging from –8.373 to –4.168, all p < .001) and certain energy drinks,
Table 3 Frequency of Soft Drink Consumption
Orlando Consumption
(n=166)
No Yes
Soft Drinks Count % Count %
Mountain Dew 75 45% 91 55%
Diet Mountain Dew 155 93% 11 7% c
Coke 84 51% 82 49%
Diet Coke 141 85% 25 15% c
Pepsi 73 44% 93 56%
Diet Pepsi 147 89% 19 11% c
Dr. Pepper 108 65% 58 35%
Diet Dr. Pepper 146 88% 20 12% c
Gatorade 67 40% 99 60%
Powerade 125 75% 41 25% c
Sprite 97 58% 69 42%
7-Up 143 86% 23 14% c
Sierra Mist 117 70% 49 30% c
Starbucks Double Shot 135 81% 31 19% c
Sobe 134 81% 32 19% c
Adrenaline Rush 152 92% 14 8% c
Arizona RX 147 89% 19 11% c
Snapple 144 87% 22 13% c
Amp (Mountain Dew) 141 85% 25 15% c
Red Bull 127 77% 39 23% c
Column Totals 2458 74% 862 26% c
746
Table 4 Frequency of Soft Drink Consumption
Louisville Consumption Aggregate Consumption
(n=552) (n=1138)
No Yes No Yes
Soft Drinks Count % Count % Count % Count %
Mountain Dew 198 36% 354 64% 449 39% 689 61%
Diet Mountain Dew 457 83% 95 17% d 997 88% 141 12% e
Coke 303 55% 249 45% d 605 53% 533 47% e
Diet Coke 440 80% 112 20% d 941 83% 197 17% e
Pepsi 252 46% 300 54% 481 42% 657 58%
Diet Pepsi 442 80% 110 20% d 945 83% 193 17% e
Dr. Pepper 297 54% 255 46% d 615 54% 523 46% e
Diet Dr. Pepper 469 85% 83 15% d 978 86% 160 14% e
Gatorade 261 47% 291 53% 493 43% 645 57%
Powerade 398 72% 153 28% d 828 73% 309 27% e
Sprite 352 64% 200 36% d 684 60% 454 40% e
7-Up 459 83% 92 17% d 911 80% 226 20% e
Sierra Mist 358 65% 193 35% d 725 64% 412 36% e
Starbucks Double Shot 482 87% 70 13% d 928 82% 210 18% e
Sobe 433 79% 118 21% d 812 71% 325 29% e
Adrenaline Rush 480 87% 71 13% d 974 86% 163 14% e
Arizona RX 496 90% 56 10% d 990 87% 148 13% e
Snapple 485 88% 67 12% d 912 80% 226 20% e
Amp (Mountain Dew) 463 84% 88 16% d 916 81% 221 19% e
Red Bull 413 75% 139 25% d 824 72% 314 28% e
Column Totals 7938 72% 3096 28% d 16008 70% 6746 30% e
Note. Column subscripts within each sample indicate a signicant difference in proportion of yes responses between the brand and Mountain Dew at p <.05 (Bonferroni
adjusted).
Brand Positioning in a Sponsorship Context 747
including Amp, Arizona Rx, Starbucks Double-Shot, and Adrenaline Rush (resid-
ual values ranging from –8.133 to –1.504, all p < .001, except for Amp in
Portland).
As displayed in Tables 2, 3, and 4, Mountain Dew had the highest proportion
of self-reported consumption (“yes” responses) in Louisville (61%), the second
highest in Portland (63%), and the third highest in both San Jose (54%) and
Orlando (55%). Across all cities, Mountain Dew (61%) had the highest proportion
of self-reported consumption, followed by Pepsi (58%), and Gatorade (57%). To
determine whether there were signicant differences between the proportions of
self-reported consumption for Mountain Dew and the proportions for other brands,
two-tailed Z-tests for dependent proportions (Wild & Seber, 1993) were per-
formed overall and for each of the four cities. Across all cities, Bonferroni adjusted
results indicated a signicantly higher proportion of self-reported consumption of
Mountain Dew than any other brand at alpha .05 (Z tests ranging from 4.44 and
14.35, all p < .05), except when compared with the proportions of Pepsi (Z = 0.87,
p = .38) and Gatorade (Z = 1.20, p = .23). Similar results were found in the Lou-
isville data. Although Mountain Dew was the second or third most consumed
brand in the other cities, its proportions were not statistically different than other
leading brands such as Coke, Pepsi, and Gatorade. Therefore, in regard to our rst
question, Mountain Dew was most likely to be consumed than other leading
brands, with an exception for Pepsi and Gatorade, across the entire sample and for
the Louisville sample in particular.
Results of Correspondence Analyses by City
The cross-tabulation matrix used in the analysis, which relates the brands (rows)
to target group categories (columns) for the Louisville sample, is displayed in
Tables 5 and 6. The frequencies reported are the afrmative responses of the
respective soft drink brand consumption. Table 7 shows the results of the decom-
position of the matrix into dimensions that underlie the relationship between
brands and target groups. There were a total of seven dimensions that recovered
perfectly the data matrix. The eigenvalues for the rst two dimensions accounted
for cumulative proportions of inertia equal to 75.55%. Because the third dimen-
sion only added to an additional 9.92% of the variance coupled with the desire to
keep a low-dimensionality solution to facilitate interpretation, two dimensions
were deemed appropriate for the data. As recommended by Bendixen (1996), a
signicant relationship between brands and targets was established by examining
the Chi-square statistic and an approximation of a correlation coefcient, which is
given by the square root of the trace. As suggested by Bendixen (1996), correla-
tion coefcients higher than .20 were considered signicant. The overall Chi-
square showed that the relationship between brands and target groups was signi-
cant (2 = 213.824, df = 133, p < .001), with a correlation coefcient of .26. This
result provides a partial answer to our second question indicating that a relation-
ship between brand consumption and target markets exist at least for the Louis-
ville sample.
Figure 1 shows the joint display of brands and target groups in a map for
Louisville. Focusing on the brands rst, the gure shows four clustering of brands.
Diet brands including, Diet Coke, Diet Pepsi, Diet Mountain Dew, and Diet Dr.
748
Table 5 Cross-Tabulation Matrix for Louisville Data
Male 17 Years & Younger Male 18-24 Male 25-36 Male 37+
Brand Count Row % Count Row % Count Row % Count Row %
Mountain Dew 89 25% 44 12% 65 18% 59 17%
Diet Mountain Dew 19 20% 4 4% 15 16% 20 21%
Coke 66 27% 33 13% 37 15% 37 15%
Diet Coke 13 12% 3 3% 17 15% 25 22%
Pepsi 70 23% 37 12% 39 13% 47 16%
Diet Pepsi 15 14% 6 5% 14 13% 22 20%
Dr. Pepper 72 28% 30 12% 36 14% 31 12%
Diet Dr. Pepper 18 22% 4 5% 9 11% 14 17%
Gatorade 72 25% 35 12% 48 17% 49 17%
Powerade 46 30% 15 10% 17 11% 24 16%
Sprite 57 29% 23 12% 28 14% 17 9%
7-Up 33 36% 7 8% 13 14% 10 11%
Sierra Mist 54 28% 19 10% 26 14% 26 14%
Starbucks Double Shot 16 23% 8 11% 9 13% 9 13%
Sobe 46 39% 15 13% 7 6% 11 9%
AdrenalineRush 26 37% 8 11% 7 10% 5 7%
Arizona RX 16 29% 4 7% 3 5% 9 16%
Snapple 13 19% 5 7% 10 15% 10 15%
Amp (MD) 39 44% 18 20% 6 7% 4 5%
Red Bull 42 30% 21 15% 17 12% 13 9%
Column Totals 822 27% 339 11% 423 14% 442 14%
749
Table 6 Cross-Tabulation Matrix for Louisville Data
Female 17 Years and
Younger Female 18-24 Female 25-36 Female 37+
Row
Totals
Brand Count Row % Count Row % Count Row % Count Row %
Mountain Dew 19 5% 22 6% 23 7% 32 9% 353
Diet Mountain Dew 4 4% 10 11% 8 8% 15 16% 95
Coke 16 6% 17 7% 23 9% 20 8% 249
Diet Coke 8 7% 13 12% 13 12% 20 18% 112
Pepsi 16 5% 26 9% 29 10% 35 12% 299
Diet Pepsi 4 4% 11 10% 12 11% 26 24% 110
Dr. Pepper 16 6% 23 9% 22 9% 23 9% 253
Diet Dr. Pepper 7 9% 9 11% 8 10% 13 16% 82
Gatorade 14 5% 21 7% 20 7% 31 11% 290
Powerade 8 5% 11 7% 13 9% 18 12% 152
Sprite 10 5% 21 11% 20 10% 22 11% 198
7-Up 6 7% 10 11% 4 4% 8 9% 91
Sierra Mist 11 6% 18 9% 20 10% 18 9% 192
Starbucks Double Shot 6 9% 9 13% 9 13% 4 6% 70
Sobe 9 8% 10 8% 8 7% 12 10% 118
AdrenalineRush 3 4% 5 7% 8 11% 9 13% 71
Arizona RX 2 4% 8 14% 4 7% 10 18% 56
Snapple 7 10% 9 13% 7 10% 6 9% 67
Amp (MD) 2 2% 10 11% 4 5% 5 6% 88
Red Bull 13 9% 12 9% 8 6% 13 9% 139
Column Totals 181 6% 275 9% 263 9% 340 11% 3085
750
Table 7 Results of Correspondence Analysis – Louisville
Dimension I (Energy/Diet) Dimension II (Convenience)
Coordinate
Squared
Correlation Contribution Coordinate
Squared
Correlation Contribution
Mountain Dew 0.002 0.000 0.000 -0.188 0.898 0.306
Diet Mountain Dew 0.322 0.853 0.082 0.029 0.007 0.002
Coke -0.057 0.160 0.007 -0.109 0.601 0.073
Diet Coke 0.499 0.964 0.231 0.057 0.012 0.009
Pepsi 0.048 0.237 0.006 -0.014 0.021 0.001
Diet Pepsi 0.469 0.798 0.200 0.160 0.093 0.069
Dr. Pepper -0.079 0.705 0.013 -0.016 0.027 0.002
Diet Dr. Pepper 0.232 0.578 0.037 0.152 0.249 0.046
Gatorade 0.032 0.048 0.002 -0.119 0.672 0.100
PowerAde -0.011 0.008 0.000 0.049 0.144 0.009
Sprite -0.087 0.233 0.012 0.073 0.165 0.026
7-Up -0.154 0.287 0.018 0.067 0.055 0.010
Sierra Mist -0.022 0.051 0.001 0.018 0.037 0.002
Double Shot -0.019 0.004 0.000 0.001 0.000 0.000
Sobe -0.279 0.594 0.076 0.186 0.263 0.100
Adrenaline Rush -0.188 0.337 0.021 0.179 0.305 0.056
Arizona RX 0.109 0.078 0.006 0.331 0.713 0.150
Snapple 0.136 0.185 0.010 0.006 0.000 0.000
Amp -0.563 0.848 0.231 0.138 0.051 0.041
Red Bull -0.202 0.570 0.047 0.000 0.000 0.000
Male 17 years and younger -0.225 0.875 0.344 0.056 0.055 0.064
(continued)
751
Table 7 (continued)
Dimension I (Energy/Diet) Dimension II (Convenience)
Coordinate
Squared
Correlation Contribution Coordinate
Squared
Correlation Contribution
Male 18-24 -0.261 0.717 0.192 -0.100 0.106 0.083
Male 25-36 0.081 0.127 0.023 -0.194 0.731 0.388
Male 37+ 0.243 0.778 0.217 -0.086 0.098 0.081
Female 17 years and
younger
-0.001 0.000 0.000 -0.014 0.002 0.001
Female 18-24 0.044 0.038 0.004 0.153 0.464 0.157
Female 25-36 0.120 0.263 0.031 0.056 0.057 0.020
Female 37+ 0.259 0.617 0.189 0.158 0.229 0.207
Total
Principal Inertia
(Eigenvalues)
0.039 0.013 0.052
Percentage(Trace) 56.41% 19.14% 75.55%
752
Figure 1 — Louisville Preference Map
Brand Positioning in a Sponsorship Context 753
Pepper, formed one group in the upper right quadrant of the map. On the left
quadrants of the map, energy brands such as AMP and Red Bull formed another
group. Mountain Dew, Coke, and Gatorade formed a third group in the lower
quadrants. Finally, Arizona Rx, Sobe, and Adrenaline Rush formed another group
in the upper quadrants of the map. The contributions to the inertia and correlations
between brands and dimensions (cosine square) shown in Table 7 were used to
interpret the map. The main contributions and highest correlations highlighted in
Table 7 conrm that the diet brands and Dew Amp and Red Bull were the main
contributors to dimension I, and that Mountain Dew, Coke, Gatorade, Arizona Rx,
and Adrenaline Rush were the main contributors to dimension II. Based on these
results, we tentatively interpret the rst dimension as “diet/energy” and the second
as “convenience.” The labeling of the second dimension is derived from the fact
that on one side of the dimension, there is a cluster of highly available and con-
sumed brands, Mountain Dew, Coke, and Gatorade, and the other with niche
brands such as Arizona Rx and Adrenaline Rush that are not as readily available
in vending machines in schools or stores.
Also shown in Table 7, the main target groups that contributed to dimension
I were male teenagers (12–17 years old) and young males (18–24 years old)
stretched along the left most side of dimension I (along with the energy drinks),
and older 37+ years of age males and females on the right most side of dimension
I (along with diet drinks). The most important contributors to the second dimen-
sion were females across different age groups positioned on the top of the map
and males between 25 and 36 years of age on the bottom of the map (along with
Mountain Dew, Coke, and Gatorade). Males between 18 and 24 years of age and
37 years of age and older also contributed to dimension II, but the contributions
were relatively much smaller compared with males between 25 and 36 years of
age. When interpreting the map displayed in Figure 1, it is important to consider
the joint contributions that each individual variable makes to explain the map. For
example, females 37 years of age and older contribute positively to both dimen-
sions, therefore this target’s position is located at the right (along diet drinks) and
upper quadrant of the map (along with niche drinks).
The title sponsor, Mountain Dew, contributes primarily to dimension II and is
shown to be competing in the same preference space as Coke and Gatorade. Its
consumption (along with the consumption of Coke and Gatorade) was reported to
be more strongly associated with males between 25 and 36 years of age who
attended the Louisville event. Therefore, as a response to our third question, these
results indicate that Mountain Dew was “better” positioned toward the adult male
market (25–36 years old), along with Coke and Gatorade, than other brands for
the Louisville sample. On the other hand, energy drinks seem to be better posi-
tioned toward the teenager market (17 years of age & younger) and young adults
(18–24 years old), while the diet drinks tend to be better positioned to females and
older adults. Given this scenario, it is not surprising to see Mountain Dew brand
extensions, AMP and Diet Mountain Dew, lling the gaps to reach the teenager,
younger adults, and female target markets in Louisville.
The analysis of the data gathered in the other three cities of Portland, Orlando,
and San Jose showed mixed results regarding the relationship between brands and
target groups. While the correlation coefcients showing dependency between
brands and targets were above .28 for each city, the Chi-square statistics were not
754 Ferreira, Hall, and Bennett
signicant (2 values ranging from 110.72 to 116.88, df ranging from 112 to 133,
p > .83). For comparative purposes, Pearson correlations between the four city
maps for dimensions I and II derived from a correspondence analysis were com-
puted. Results showed consistency across maps in how brands were positioned
along dimension I: energy on one extreme and diet drinks on the other extreme
(Mountain Dew generally centrally located). Pearson correlation coefcients were
higher than .58 for brand locations on dimension I between maps. However,
results were inconsistent in regard to dimension II as evidenced by low correla-
tions between the maps.
Results of Correspondence Analysis Across Cities
To derive an overall picture of the relationships between brands and targets
accounting for the variation across cities, a preference map was developed using
a composite variable that represented any gender, age, and city combination (32
levels). This map required a new level of analysis beyond the one conducted for
each city separately. The most striking benet of this overall map is the identica-
tion of the locations in which the brand-target relationships are stronger. Because
the map is much more complex than the ones within each city, it is up to the ana-
lyst to decide whether the contrasts by location overcome the added complexity
and difculty of interpretation.
The overall map is displayed in Figure 2. To avoid clutter, only the targets
with most contributions to the map were plotted along with the brands. The Chi-
square showed that the relationship between brands and target groups was signi-
cant (2 = 738.195, df = 589, p < .001). The eigenvalues for the rst two dimen-
sions accounted for cumulative proportions of inertia equal to 53.94%.
With a close resemblance to the map for Louisville, the overall map shows
four clusters of brands. The rst cluster of diet brands included, Diet Coke, Diet
Pepsi, and Diet Dr. Pepper, in the upper right quadrant of the map (although not in
the upper quadrant, Diet Mountain Dew is also on the far right side of the map).
On the left side along dimension I, energy brands such as AMP, Sobe, and to a
lesser extent Adrenaline Rush and Red Bull formed another group. Mountain Dew
and Coke3 formed a third group in the lower right quadrant. Finally, on the upper
left quadrant, Snapple, Starbucks Double Short, Sobe and Arizona RX formed the
fourth group. The contributions to the inertia and correlations between brands and
dimensions (cosine square) conrm that the diet brands and Dew Amp were the
main contributors to dimension I, and Mountain Dew, Coke, and Snapple were the
main contributors to dimension II.
It is important to understand that the analysis across cities may present differ-
ent results from the analysis by city because a different contingency table is used
(with new row and column totals). This implies that if a relationship between a
particular target and a brand may be higher (lower) than expected for one particu-
lar city, the relationship would not necessarily be higher (lower) than expected
when other cities are considered. While the relationship between brands and tar-
gets were not signicant for three cities when examined separately, many relation-
ships between brands and targets were higher than expected when considering the
entire data set. For example, females across different age groups in San Jose and
Portland were major contributors to dimension II and were positioned on the top
755
Figure 2 — Overall Preference Map
756 Ferreira, Hall, and Bennett
quadrants of the map along with niche drinks such as Snapple, Sobe, and Star-
bucks Double Shot. Males 24 years of age and younger in San Jose and Portland
were major contributors of dimension I and were positioned on the far left of the
map along with Amp, Adrenaline Rush, and Red Bull. Males 36 years of age and
younger in Louisville and older males 37+ years of age in Orlando were major
contributors of dimension II and positioned at the lower quadrants of the map
along with Mountain Dew and Coke. Finally, older males and females 25+ years
of age in Louisville were major contributors of dimension I and positioned on the
right quadrants of the map along with the diet drinks. When all cities are consid-
ered, Mountain Dew as the title sponsor was shown to be more positioned to male
targets especially in Louisville as evidenced by the attendees’ reported behavior.
Mountain Dew Amp was better positioned for young audiences especially in Port-
land and Diet Mountain Dew to older adults of both genders in Louisville.
Discussion
Sponsorship of sporting, entertainment, and art events has been used to accom-
plish many marketing objectives. Prominent marketing objectives that have
received considerable research attention include brand awareness, brand image
and image transfer, and ROI (Gardner & Shuman, 1987; Gwinner, 1997; Meen-
aghan, 1991; Pruitt, Cornwell, & Clark, 2004). One of the main advantages that
sponsorship offers over more traditional marketing tools is the ability to focus on
specic target markets (Gardner & Shuman, 1987), which are often comprised of
prospects who are emotionally invested in the sponsored event. After selecting
attractive target markets, corporations then select marketing communications
tools that will presumably best reach and potentially persuade the desired mar-
kets. Thus, it is important to gain an understanding of the role sponsorship can
play in positioning a brand.
This research has used correspondence analysis to discern Mountain Dew’s
positioning relative to competing brands in key target markets (young and male
audiences) of the Dew Action Sports Tour. As stated in the results section of this
study, correspondence analysis was shown to be effective in explaining self-
reported brand use based on latent benets of soft drinks. The resulting preference
map based on the entire data showed that Mountain Dew has successfully posi-
tioned itself to male segments only in Louisville, one of the four cities examined.
There are at least a few explanations for the ndings. First, the results suggest that
heterogeneous tastes and perceptions among target markets may exist in vastly
different geographical locations of the events. The data in Tables 2, 3, and 4 sup-
port at least differences in demographic proles by city. Therefore, the fact that
Mountain Dew tends to be more strongly associated with young and adult males
in Louisville does not necessarily mean that Mountain Dew tends to be more
strongly associated with the same targets in different locations of the country.
Second, it is also plausible to think that the effectiveness of Mountain Dew spon-
sorship activations may depend on regional differences. For example, signage,
symbols, and announcements may be more or less appealing depending on
regional differences (e.g., some cities may be more conservative than others,
Brand Positioning in a Sponsorship Context 757
hence the appropriateness of language, jargon, and accents may differ across
regions). Furthermore, as the events were held in different facilities, preexisting
signage in the venues may have created differing levels of clutter when activating
the brand at different locations (although the activations were similar across
events). Finally, since Chi-square analysis can be sensitive to sample size, these
results could also be attributed to differences in sample sizes between these
cities.
One theoretical implication of the ndings is that sponsorship effectiveness
can be dependent on the context. That is, the degree to which an intended relation-
ship is achieved may directly depend on the context within which the relationships
operate. In the context of sponsorship, the geographical location, vicinity, attend-
ees, culture, etc., may all inuence positioning. Keller (1993) suggests that brand
image is formed and inuenced by any contact a person has with a brand. Further-
more, whether a brand should position itself the same way in different regions is
also an important consideration. In this regard, the decision of standardization
across regions is a consideration similar (but perhaps to a lesser extent) to the
internationalization of the brand in different countries (Bagozzi et al., 1998).
The results from correspondence analysis were consistent with multiattribute
theory (Ajzen & Fishbein, 1980; Edwards & Barron, 1994; Keeney & Raiffa,
1976), which portrayed overall preference for soft drink brands as a function of
desired benets derived from consumption, including energy, diet, and conve-
nience. Therefore, self-reported consumption was explained by the attributes that
products possessed and the benets individuals derive from the attributes. In addi-
tion, the possibility of graphically portraying brands and target markets in the
same preference map provided a way to assess audience and brand t. Therefore,
correspondence analysis along with the resulting preference map was useful as a
form of sponsorship evaluation. With the analysis based on self-reported brand
use, results actually became more relevant and realistic since it was based on indi-
viduals’ self-reported behavior as opposed to individuals’ self-reported intentions.
Such mapping not only shows the consumer characteristics associated with con-
sumption of the sponsoring brand, but also includes the portrayal of competitor
positioning, which can be useful in planning future sponsorship and positioning
strategies.
Managerial Implications
The results of this study offer managerial implications in three key areas: sponsor-
ship renewals, new strategies, and brand management. Regarding sponsorship
renewals, as is witnessed by the sheer frequency of self-reported consumption, the
Mountain Dew brand has been relatively successful in accomplishing its position-
ing strategy in the venues of the Dew Action Sports Tour sponsorship activities
where data were collected. The analysis across cities shows that Mountain Dew is
most successfully positioned toward male consumers in Louisville. Based on
these results, managers must evaluate whether their sponsorship objectives have
been met, whether renewals should take place, and what new objectives should be
set. According to Hein (2007b), although the Dew Action Sports Tour will most
likely continue, Mountain Dew may change its “extreme” positioning strategy in
758 Ferreira, Hall, and Bennett
the near future to something new or not-so-extreme because of confusion in the
marketplace. Such consideration shows that Mountain Dew’s managers are not
completely satised with its current positioning strategy.
In addition to sponsorship renewals, managers may be concerned with brand
management and strategic issues for a couple of reasons. Firstly, it is reported that
in using action sports to reach target markets, Mountain Dew is aiming to attract
young consumers. The results show that the youngest age group of respondents
(17 years and younger) is more closely associated with energy drinks than with
the popular soft drinks. Mountain Dew managers may want to reconsider the
methods used to reach this target if it is indeed the objective. The fact that Moun-
tain Dew AMP is more strongly associated with the teen segment indicates that
Mountain Dew may prefer to position for older age groups to avoid cannibaliza-
tion. Secondly, it is requisite that the positions of competing soft drinks be consid-
ered. As seen on the aggregate data preference map, Mountain Dew shares similar
preference space to Coca-Cola, Powerade, and Gatorade. Acknowledging which
brands Mountain Dew is mostly competing with for this target market can aid
brand managers in formulating future sponsorship activities. Finally, managers
may also consider regional differences when planning activation if costs associ-
ated with customization are not prohibitive.
Limitations and Future Research
There are at least four limitations to this study that are important to highlight.
First, this study was susceptible to the problems associated with any eld study. In
eld studies, the ability to control for factors that could explain the differences
found between cities is limited. Hence, differences in taste, activation effective-
ness, facilities, geographic and economic conditions can all be associated and
provide partial explanations for the ndings. However, the analysis was useful to
generate possible hypotheses for the lack of consistency across cities that can be
further explored in future research endeavors. Second, the study was also limited
to a series of events within one sport category, action sports, and its relationship
with the soft drink category. Third, observations were obtained by asking respon-
dents to report their brand usage by picking any soft drinks they have consumed
for the past two weeks. Although self-reported data are useful and easy to collect,
they may be subjected to inaccuracies. Given the effortless task associated with
“picking” brands, we perceive that the advantages of this type of data overcome
its disadvantages. The alternative would be to obtain real purchase information
(secondary data). However, secondary data are not error-free and can be very
expensive, especially if additional primary research is needed to link action sports
attendance behavior to purchase data. Assuming the cost is not an issue and sec-
ondary data are available, secondary data can be used to complement eld studies.
Winer (1999) highlights the importance of combining data sources and offers
three examples of how scanner data can be used to externally validate the studies
conducted in laboratory settings. Finally, the analysis ignored the dynamic nature
of positioning. Future studies should attempt to capture changes of positioning
over time and correlate these changes to managerial and marketing variables. This
may also lead for example to the employment of preexperimental designs such as
Brand Positioning in a Sponsorship Context 759
pretest-posttest or quasi-experiments such as the time-series design (Campbell &
Stanley, 1963) to assess the impact of the sponsorship. Therefore, comparing the
positioning from the inaugural tours to the following years can show how posi-
tioning can evolve over time, and if expenditures and activations between this year
and the following year are tracked, a more sophisticated model can be built to
better isolate the effects of sponsorship.
Nevertheless, this study provided an exploration of the relationship between
an event sponsor, its competitors, and the effectiveness of acquiring consumers
from the desired target market through exposure to the sponsorship. Self-reported
brand use showed that the event sponsor was among the top three brands con-
sumed in each of the four cities where data were collected and the most consumed
brand overall. Correspondence analysis indicated that consumers differentiated 20
soft drink brands based on energy, diet, and convenience attributes. Based on
proximity of the target market to the sponsoring brand, it appears that Mountain
Dew has been relatively effective in using sponsorship to position the brand with
the intended target market in at least one of the cities examined.
Notes
1. In 2008, MTV Networks Music Group has also partnered with NBC to expand the tour,
and a winter tour has been added as an additional property. The tour has also rebranded as AST
(Action Sports Tour) Dew Tour.
2. The groups were not entirely equal in size because instead of using the age of 16 as the
cutoff for the rst quartile, we used 17 to particularly separate teenagers under age from young
adults. The cutoff ages for equal groups were 16, 24, and 36. Despite this cutoff, the analysis was
not substantially altered using either 16 or 17 years of age as the cutoff age.
3. Those interested can contact the authors to obtain preference maps of other individual
cities.
4. Although Powerade and Gatorade are also in the lower quadrants, their contribution to
dimensions I and II are small.
References
Aaker, D.A., & Shansby, J.G. (1982). Positioning your product. Business Horizons, 25(3),
36–62.
Arimond, G., & Elfessi, A. (2001). A clustering method for categorical data in tourism
market segementation research. Journal of Travel Research, 39(4), 391–397.
Ajzen, I., & Fishbein, M. (1980). Understanding attitudes and predicting social behavior.
Englewood Cliffs, NJ: Prentice-Hall Inc.
Bagozzi, R.P., Rosa, J.A., Celly, K.S., & Coronel, F. (1998). Marketing management.
Upper Saddle River, NJ: Prentice Hall.
Bendixen, M. (1996). A practical guide to the use of correspondence analysis in marketing
research. Marketing Bulletin Online, 14, Technical Note 2, 16-38.
Bennett, G., & Lachowetz, T. (2004). Marketing to Lifestyles: Action sports and Genera-
tion Y. Sport Marketing Quarterly, 13, 239–243.
Calantone, R.J., Benedetto, C.A., Hakam, A., & Bojanic, D.C. (1989). Multiple multina-
tional tourism positioning using correspondence analysis. Journal of Travel Research,
28(2), 25–32.
760 Ferreira, Hall, and Bennett
Campbell, D.T., & Stanley, J.C. (1963). Experimental and quasi-experimental designs for
research. Hopewell, NJ: Houghton Mifin Company.
Carroll, J.D., Green, P.E., & Schaffer, C.M. (1987). Comparing interpoint distances in cor-
respondence analysis: A clarication. JMR, Journal of Marketing Research, 24(4),
445–450.
Cliffe, S.J., & Motion, J. (2005). Building contemporary brands: A sponsorship-based
strategy. Journal of Business Research, 58(8), 1068–1077.
Cornwell, T.B., Weeks, C.S., & Roy, D.P. (2005). Sponsorship-linked marketing: Opening
the black box. Journal of Advertising, 34(2), 21–42.
Edwards, W., & Barron, F.H. (1994). SMARTS and SMARTER: Improved simple methods
for multiattribute utility measurement. Organizational Behavior and Decision Pro-
cesses, 60(3), 306–325.
Elrod, T. (1991). Internal analysis of market structure: Recent developments and future
prospects. Marketing Letters, 2(3), 253–266.
Elrod, T., & Winer, R. (1992, August). An Empirical comparison of mapping methods
based on panel data. Unpublished Manuscript, Faculty of Business, University of
Alberta, Edmonton.
Fennell, G., & Allenby, G.M. (2004). An integrated approach: Market denition, market
segmentation, and brand positioning create a powerful combination. Marketing
Research, 16(4), 28–34.
Fennell, G., Allenby, G., Yang, S., & Edwards, Y. (2003). The effectiveness of demographic
and psychographic variables for explaining brand and product category use. Quantita-
tive Marketing and Economics, 1, 223–244.
Gardner, M.P., & Shuman, P.J. (1987). Sponsorship: An important component of the pro-
motions mix. Journal of Advertising, 16(1), 11–17.
Gladden, J., & McDonald, M. (2006). Alternative global avenues: Action sports and events.
In J. Amis & T.B. Cornwell (Eds.), Global Sport Sponsorship (pp. 191–206). Oxford,
NY: Berg.
Green, C., & Muller, T. (2002). Positioning a youth sport camp: A brand-mapping exercise.
Sport Management Review, 5, 179–200.
Greenacre, M.J. (1984). Theory and Application of Correspondence Analysis. London:
Academic Press.
Greenacre, M.J. (1989). The Geometric interpretation of correspondence approach. Jour-
nal of the American Statistical Association, 82, 437–447.
Gwinner, K. (1997). A model of image creation and image transfer in event sponsorship.
International Marketing Review, 14(3), 145–158.
Hair, J.F., Anderson, R.E., Tatham, R.L., & Black, W.C. (1998). Multivariate data analysis.
Upper Saddle River, NJ: Prentice Hall.
Halliday, J. (2004). Honda does a Volvo, plays on safety. Advertising Age, 75(36), 3–4.
Hein, K. (2007a). PepsiCo puts the pop in pop culture. Brandweek, 48(17), 4.
Hein, K. (2007b). Not so extreme: Dew retooling image. Brandweek, 47(39), 6.
Hoffman, D.L., & Franke, G.R. (1986). Correspondence Analysis: Graphical representa-
tion of categorical data in marketing research. JMR, Journal of Marketing Research,
23(3), 213–227.
Holt, D. (2003). What becomes an icon most? Harvard Business Review, 81(3), 43–49.
Howard, D.R., & Crompton, J.L. (2004). Financing Sport. Morgantown, WV: Fitness
Information Technology, Inc.
IEG (International Event Group). (1999). IEG’s Complete Guide to Sponsorship: Every-
thing You Need to Know about Sports, Arts, Event, Entertainment, and Cause Market-
ing. Chicago, IL: IEG, Inc.
IEG (International Event Group). (2004). IEG Sponsorship Report. Chicago: IEG, Inc.
Kaciak, E., & Louviere, J. (1990). Multiple correspondence analysis of multiple choice
experiment data. JMR, Journal of Marketing Research, 27, 455–465.
Brand Positioning in a Sponsorship Context 761
Kardes, F. (1998). Consumer Behavior & Managerial Decision Making. Reading, MA:
Addison-Wesley Educational Publishers, Inc.
Kates, S.M., & Goh, C. (2003). Brand morphing. Journal of Advertising, 32(1), 59–68.
Keller, K.L. (1993). Conceptualizing, measuring, and managing customer-based brand
equity. Journal of Marketing, 57, 1–22.
Keller, K.L., & Aaker, D.A. (1992). The effects of sequential introduction of brand exten-
sions. JMR, Journal of Marketing Research, 29(1), 35–50.
Keeney, R.L., & Raiffa, H. (1976). Decisions with multiple objectives: preferences and
value tradeoffs. New York: Wiley.
Lippert, B. (2003). Child’s play. Adweek, 44(17), 36.
Martin, J.H. (1994). Using a perceptual map of the consumer’s sport schema to help make
sponsorship decisions. Sport Marketing Quarterly, 3(3), 27–33.
McCarthy, M. (2001, August 14). ESPN’s promotion of X Games goes to extremes.
USA Today. Retrieved December 11, 2007, from http://www.usatoday.com/money/
advertising/2001-08-14-xgames.htm
Meenaghan, T. (1991). The role of sponsorship in the marketing communications mix.
International Journal of Advertising, 10, 35–47.
Meenaghan, T. (2001). Understanding sponsorship effects. Psychology and Marketing,
18(2), 95–122.
Milne, G.R., McDonald, M.A., Sutton, W.A., & Kashyap, R. (1996). A niche-based evalua-
tion of sport participation patterns. Journal of Sport Management, 10(4), 417–434.
Mullman, J. (2006). Miller puts the gruff back into High Life. Advertising Age, 77(42),
3–4.
Nicholls, J.A.F., Roslow, S., & Dublish, S. (1999). Brand recall and brand preference at
sponsored golf and tennis tournaments. European Journal of Marketing, 33(3), 365–
387.
Pracejus, J.W. (2004). Seven psychological mechanisms through which sponsorship can
inuence consumers. In L.R. Kahle & C. Riley (Eds.), Sports marketing and the psy-
chology of marketing communication (pp. 175–189). NJ: Lawrence Erlbaum Associ-
ates, Inc.
Pracejus, J.W., & Olsen, G.D. (2004). The role of brand/cause t in the effectiveness of
cause-related marketing campaigns. Journal of Business Research, 57(6), 635–640.
Pruitt, S.W., Cornwell, T.B., & Clark, J.M. (2004). The NASCAR phenomenon: Auto
racing sponsorships and shareholder wealth. Journal of Advertising Research, 44(3),
281–296.
Ries, A., & Trout, J. (1986). Positioning: The battle for your mind. NY: McGraw-Hill.
Speed, R., & Thompson, P. (2000). Determinants of sports sponsorship response. Journal
of the Academy of Marketing Science, 28, 226–238.
Wild, C.J., & Seber, G.A. (1993). Comparing two proportions from the same survey. The
American Statistician, 47(3), 178–181.
Winer, R.S. (1999). Experimentation in the 21st century: The importance of external valid-
ity. Journal of the Academy of Marketing Science, 27(3), 349–358.
