Академический Документы
Профессиональный Документы
Культура Документы
To cite this article: Kevin P. Gwinner & John Eaton (1999) Building Brand Image Through
Event Sponsorship: The Role of Image Transfer, Journal of Advertising, 28:4, 47-57, DOI:
10.1080/00913367.1999.10673595
Past sponsorship research has primarily focused on awareness building strategies, and has virtually ignored
brand image issues. As a result, little guidance is available for firms that seek to use sponsorship opportuni-
ties to aid in brand positioning. This study reports the results ofan experiment using undergraduate student
subjects, who assessed the degree to which a sporting event's image was transferred to a brand through event
sponsorship activity. Subjects in the sponsorship pairing treatment were more likely to report similarities on
brand-event personality components than subjects who were not exposed to the event-brand sponsorship link,
thus supporting the notion that sponsorship results in image transfer. Further, we found that when event and
brand are matched on either an image or functional basis the transfer process is enhanced. Management
implications for sporting event sponsorship and future research directions are discussed.
Kevin P. Gwinner (Ph.D., Arizona While firms enter into sponsorship arrangements for a variety of reasons,
State University) ill AMistant Profe8Blll'
of Marketing, Kansaa State University. two of the most common are: (1) to increase brand awareness, and (2) to
John Eaton <M.BA, University of establish, strengthen, or change brand image (Cornwell and Maignan 1998;
Toledo) is a doctDralcandidate in Crowley 1991; Gwinner 1997; Marshall and Cook 1992; Meenaghan 1991;
marketing at Arizona State University. Meerabeau et al. 1991). Typically, strategies aimed at increasing brand
The authors wish to thank Stephen awareness are implemented using a multitude of promotional media and
Nowlis (Arizona State University) for
guidance related to experimental are designed to have the sponsoring brand exposed to as many potential
design i88ues and the four anony- consumers as possible. Past research has examined the effectiveness of
mous Journal ofAduertising these awareness building strategies through a variety of methods (e.g., total
reviewers and former editor Les
Carlson (Clemson University) for event attendance, exit polls, sales following the event, and number of media
their helpful comments on earlier mentions). Regrettably, far less research attention has focused on brand
versions of this manullC1'ipt. image issues. Indeed, researchers have suggested that little is understood
about what makes sponsorship "work," particularly with regard to image
association (Javalgi et al, 1994; Lee, Sandler and Shani 1997).
The purpose of this paper is to gain insight into the brand image aspects of
sponsorship. Specifically, based on theoretical perspectives from the celeb-
rity endorsement, schema, and advertising literatures, we propose and test
several relationships involving the influence of sporting event sponsorship
on the sponsoring brand's image.
ciations held in consumers' memories regarding a ce- H1:A sporting event's image will transfer to
lebrity or sporting event become linked in memory with a sponsoring brand's image when they are
the endorsed or sponsoring brand In essence, the ce- linked through sponsorship.
lebrity or event image is transferred to the brand.
This transfer of associations is consistent with The Impact ofEvent and Brand Match-
McCracken's (1989) view of the celebrity endorsement up on Image Transfer
process. McCracken eschews the "credibility" and "at-
tractiveness" models of endorsement used to explain Many scholars have examined or commented on
the persuasive nature of endorsers. Instead he offers the importance of matching the characteristics of
a theory of meaning transfer, where "meaning" refers spokespersons with the characteristics of the prod-
to an overall assessment of what a celebrity "repre- ucts they endorse (see Lynch and Schuler [1994] for
sents" to the consumer. This meaning is built upon an excellent review of this literature). Generally, these
an individual's interpretation of the celebrity's public types of studies have found that a match between
image as demonstrated in "television, movies, mili- endorser and brand leads to a variety of positive out-
tary, athletics, and other careers" (McCracken 1989, comes for firms including enhanced spokesperson ex-
p. 315). According to this theory, the meaning attrib- pertise/credibility, a more positive attitude toward
uted to celebrities moves from the celebrity endorser the ad, a more positive attitude toward the brand and
to the product when the two are paired in an adver- higher brand recall.
tisement (McCracken 1989). The transfer process is Kahle and Homer (1985) were among the first to
completed when a consumer acquires/consumes the empirically examine and propose a "match-up" hy-
product, thus transferring the meaning to the user. pothesis in the context of celebrity endorsement. These
With regard to implications for brand image, sport- authors argued and found support for the notion that
ing events and celebrity endorsers are similar on two advertising effectiveness is increased when the im-
levels. First, consumers can associate both sporting age of the celebrity converges with the image of the
events and celebrities with particular meanings. While endorsed product. Specifically, they found physically
celebrities derive their meanings from consumer per- attractive celebrity endorsers of a beauty enhancing
ceptions of their various public activities (e.g., movies, product (i.e., endorser-brand match) to have a posi-
athletics, politics, etc.), the meanings associated with tive influence on consumer's brand attitudes, pur-
sporting events are derived from the type of event, the chase intentions, brand recall, and recall of adver-
event's characteristics (e.g., professional status, venue, tisement arguments. Building on Kahle and Homer,
size, etc.), and individual consumer factors such as one's many spokesperson-brand congruence studies have
past experiences with the event (Gwinner 1997). Sec- been conducted by manipulating some physical at-
ond, events may act in a manner analogous to celebrity tribute of the spokesperson to be in-congruence or
endorsers in the transfer of image to sponsoring brands. out-of-congruence with a given product. For example,
That is,just as consumers associate a celebrity's ''mean- in one experiment, Lynch and Schuler (1994) ma-
ings" with the brand they endorse, consumers may also nipulated muscularity of the spokesperson to be in or
associate a sporting event's "meanings" with a sponsor- out of congruence with products that either helped to
ing brand. Obviously, if this image transfer process is produce muscularity (e.g., exercise equipment) or prod-
occurring, then brand managers considering sponsor- ucts perceived to be masculine in nature (e.g., car
ship arrangements should not only consider exposure batteries). In another study, Kamins (1990) manipu-
issues (i.e., brand awareness) but should also take into lated spokesperson attractiveness to be in or out of
account the congruence between a sporting event's im- congruence with an attractiveness related product.
age and the image/positioning goals for their brands. The Kamins study found that when product and
While it is possible that the directionality of the image spokesperson physical characteristics were congru-
transfer may move from brand to event rather than ent, then spokesperson believability/knowledge was
event to brand, this is less likely to occur when the increased, while Lynch and Schuler (1994) found that
event has a strong established image relative to the congruency led to high perceived spokesperson knowl-
sponsoring brand. Further, since the primary focus of edge. One notable exception to the manipulation of
the spectator is typically on the activities of the sport- spokesperson physical characteristics is Misra and
ing event rather than on the sponsors, the event's im- Beatty (1990) who examined image congruence in a
age is likely to be more salient in their mind, suggest- holistic manner akin to McCracken's notion of mean-
ing the image transfer process would move from event ing. In their study, Misra and Beatty matched spokes-
to brand This discussion leads to the fll"St hypothesis: person characteristics with product characteristics.
Winter 1999 49
For example, in their pretest of celebrities, Clint In an event sponsorship context, McDaniel (1999)
Eastwood was associated with the characteristics of has explored an aspect of the match-up hypothesis by
"tough" and "rugged." In the congruent condition this matching event and brand in terms of involvement.
celebrity was paired with a fictitious brand of jeans He found that subjects rated attitude toward the ad
(Unitough jeans), while in the incongruent condition significantly more positively when a highly involving
he was paired with a fictitious board game called product (e.g., an automobile) was paired with a highly
Funnybone. Results of this study indicated that the involving sporting event (e.g., the Olympics) than
congruence condition resulted in higher recall and when the product was paired with a low involvement
more favorable brand attitudes than incongruent or sporting event (e.g., PBA Bowling). In his study, event-
neutral congruency pairings. Kamins and Gupta product involvement match was not found to have an
(1994) also manipulated spokesperson-product con- effect on attitude toward the brand or purchase in-
gruence in terms of image. They found increased con- tention. While providing insight into one match up
gruence resulted in perceptions of higher believabil- dimension, McDaniel's study did not attempt to match
ity and attractiveness of the spokesperson and a more the sporting event and the sponsoring brand on at-
favorable product attitude. tributes related to their respective "meanings" in the
So why does this match-up hypothesis seem to be in manner that McCracken (1989) discussed or that
effect? One of the more compelling arguments pro- Misra and Beatty (1990) explored.
posed is based on schema theory (Lynch and Schuler This begs the question, "on what basis might we
1994; Misra and Beatty 1990). A schema is a cogni- judge a sporting event to be similar or dissimilar to a
tive structure that represents knowledge about a type product?" McDonald (1991) discusses the importance
of stimulus, for example, a person, event, or object of product relevance to the sponsored event, suggest-
(Bartlett 1932; Lord and Foti 1986). Schema theory is ing that it might occur directly or indirectly. The
based on research which found that memory is not a direct method occurs when the sponsoring firm's prod-
verbatim account of past experiences, but rather a ucts are (or could be) used in the event. Indirectly,
blend of both specific memories as well as general relevance can be achieved if some aspect of the
abstractions about types of people, activities, and ob- sponsor's image corresponds with the event. Gwinner
jects (Bartlett 1932; Rumelhart and Ortony 1977). (1997) has used the terms "functional based" and
Schema represents a mechanism to allow individuals "image based similarity" to refer to the potential con-
to function in a complex environment. That is, in- gruence between events and the brands/companies
stead of having to recall from memory what behav- that act as sponsors. Consistent with McDonald (1991),
iors are appropriate in a specific situation (e.g., board- Gwinner (1997) has suggested that functional based
ing a United Airlines flight) or what evaluations have similarity can occur when the sponsored brand "is
been made of some specific person (e.g., Dr. Bernhard) actually used by the participants during the event..."
or specific object (Campbell's soup), one is able to (p.152). Examples of this type of similarity would
simply recall knowledge related to the general type of include Seiko being an official timer at the U.S. Open
situation (airline boarding), person (heart surgeon), Tennis Championships or Gatorade sponsoring the
or object (soup). Ironman Triathlon. In both cases, functional similar-
With regard to the use of schema theory in support of ity is present because of the use of the sponsoring
the match-up hypothesis, Misra and Beatty (1990)found brand in the event. Image based similarity has been
evidence ofa ''filtering model," which suggested spokes- described as occurring when the "image of the event
person characteristics that are incongruent with brand is related to the image of the brand..." (Gwinner 1997,
schema characteristics will be ''filtered out" and not p.152). For example, the Master's Golf Tournament
encoded as well as congruent information. They argued and Cadillac Automobiles may be similar in terms of
that the better recall demonstrated by subjects in their a prestige image. Drawing on the earlier schema
congruent condition is a result of better or more effec- theory discussion, it can be argued that congruent
tive encoding of information. Further, they proposed, event-brand information in the form of either func-
but did not test, that this encoded information associ- tional or image based similarity will lead to enhanced
ated with a congruent celebrity spokesperson schema image transfer. Thus, extending the notion of the
would become integrated with the product's schema. If match-up hypothesis found in the celebrity endorser
this were to hold true, then one would expect the context, we offer the following two-part hypothesis:
schemas of celebrities and the schemas of the products H2a: Similarity between brand and event will
they endorse to become more similar, assuming con- influence the image transfer such that
gruence of some salient characteristics. the image transfer will be stronger for
60 The Journal ofAdverlilling
brands having functional-based similar- was done because subjects in the main study needed
ity with the event they are sponsoring to have some prior image of the event in order to
than when brands have no similarity increase our confidence in the image transfer mea-
with the event. sure. That is, they needed to have an image of the
H2b: Similarity between brand and event will event in order to have it transfer to the brand.
influence the image transfer such that We developed scale items to measure functional
the image transfer will be stronger for and image based similarity based on the definition
brands having image-based similarity provided by Gwinner (1997). Functional based simi-
with the event they are sponsoring than larity was measured using three items assessed on
when brands have no similarity with seven-point strongly disagree/strongly agree scales.
the event. The three items were: (1) "It is likely that (partici-
pants) in the (event name) use (brand name) during
Method the (event name)," (2)''When I watch the (event name),
I often see (brand name) being (used)," and (3) "(Brand
Overview name) is not a product that (participants) in the (event
name) would consider (using)." The third item was
To test the hypotheses an experiment was conducted reverse coded. The parenthetical "participants" label
using one between groups factor (sponsorship: [yes in these questions was replaced by the appropriate
sponsorship and no sponsorship]) and one repeated participant title, depending upon the specific event
measures factor (level of event-brand similarity: [im- (e.g., player, rider, driver, etc.), Cronbach's alpha for
age based, functional based, and no similarity». While this scale is .89, thus demonstrating good reliability.
there may be many ways to examine image transfer, Image based similarity was also assessed with
our examination uses brand and event personality. seven-point strongly disagree/strongly agree anchored
Brand personality can be regarded as ''the set of hu- scales using the following three measures: (1) "The
man characteristics associated with a brand..." (Aaker (event name) and (brand name) have a similar im-
1997, p. 347). Brand personality has been described age," (2) 'The ideas I associate with (brand name) are
as an important aspect of brand image that is im- related to the ideas I associate with the (event name),"
pacted by one's expectations of the type of person who and (3) ''My image of the (event name) is very differ-
would use a particular product - user imagery - and ent from the image I have of (brand name)." The
in which situations a product might be used - usage third image based similarity measure was reverse
imagery (Aaker 1997; Keller 1993; Plummer 1985). coded in the analysis. Cronbach's alpha for this scale
Both user and usage imagery can be communicated is .90. The best "No Similarity" pairing was
in an event sponsorship context. The brand personal- operationalized as the event-product score receiving
ity concept is important because it serves as a mecha- the lowest score on a summed scale consisting of all
nism upon which producers can differentiate their six items listed above.
goods and services. This becomes especially critical One hundred and thirty-five undergraduate stu-
when other potentially differentiating features are dents enrolled in a marketing management course
perceived by consumers as equal across competing responded to the similarity pretest survey. In order
brands (plummer 1985). to reduce respondent fatigue, each subject was ran-
domly assigned to respond to questions regarding
Pretests three of the seven events. This resulted in a useable
sample of between 41 and 50 subjects per event.
The first pretest sought to find appropriate pair- As indicated in Table 1, for the image similarity
ings of sporting events and sponsoring products to condition, the pairing of the U.S. Open Golf Champi-
represent each of the three types of similarity (func- onship and Acura Automobiles was found to be the
tional based, image based, and no similarity) for use best pairing. For functional similarity, the pretest
in the main experiment. As illustrated in Table 1, showed that the Indianapolis 500 Auto Race and
each potential sporting event was paired with three Goodyear Tires was the best pairing. Finally, the best
different brands. Gwinner (1997) suggests that cer- illustration of the no similarity condition was World
tain event characteristics (e.g., size, history, venue, Cup Soccer and Camel Cigarettes. These pairings
etc.) will influence an event's image. Using this as a were assessed after removing those subjects who
guide, we selected seven sporting events based on scored below the scale mid-point on a brand familiar-
their national visibility and rich, long histories. This ity question in order to increase validity.
Winter 1999 61
Table 1
Pretest 1 Pairings of Sporting Events and Sponsoring Products
Similarity Type Event Alternatives Product Pairing Alternatives Mean Similarity (std dev)
Image Based Professional Beach Volleyball Cuervo Gold tequila 9.9 (5.8)
Dorito's tortilla chips 9.-7 (4.7)
Nintendo video games 9.0 (5.3)
Kentucky Derby American Express 10.3 (5.0)
Oldsmobile automobiles 9.2 (5.1)
John Hancock insurance 8.8 (4.0)
U.S. Open Golf Championship Acura automobiles 12.1 (5.5)
Sony camcorders 10.5 (5.1)
Michelob beer 9.4 (5.2)
Functional based Indianapolis SOO Auto Race Goodyear tires 19.2 (2.6)
Pennzoil motor oil 17.0 (3.6)
Shell gasoline 14.1 (4.8)
NCAA basketball tournament Champion brand uniforms 16.9 (3.1)
Reebok shoes 16.0 (4.4)
Powerade sports drink 15.5 (4.3)
No similarity Functional Image
World Cup Soccer Dell computers 6.9 (4.8) 7.8 (4.2)
Clorox bleach 6.4 (4.5) 8.4 (5.1)
Camel cigarettes 4.1 (2.0) 4.8 (3.1)
Functional Image
Rose Bowl Irish Spring soap 5.3 (3.7) 7.3 (5.1)
Renuzit air fresheners 4.7 (2.5) 5.2 (3.6)
Levi's jeans 4.1 (2.3) 7.5 (5.1)
Note: Bold faced event-product combinations were rated as the best representatives of their respective similarity type category (based on
high mean value for Image and functional similarity pairings and low mean value for no similarity pairing) and were subsequently used In
the main study.
A second pretest was conducted to create a set of credit. Because of the use of sporting events with
personality-oriented adjectives that could be used to long, rich histories, we believe a student sample will
describe image dimensions of each of the three events be familiar with the events and, therefore, the results
selected from the first pretest. Twenty adjectives were will be generalizable to a larger population. Subjects
generated by the authors for each of the three sporting were randomly assigned to one of the two sponsor-
events. In order to increase the saliency of the task, ship treatments. Randomization assures that the im-
only adjectives that could potentially describe the par- pact of individuals' prior product schemas will not
ticular event were included (Graeff 1996). Eighty-one bias the results in any given treatment. In each con-
undergraduate students enrolled in a marketing course dition, subjects assessed event-brand combinations
used a seven-point scale to rate the 60 adjectives on representing all three similarity conditions (i.e., im-
their usefulness in describing each of the three events age based, functional, no similarity). Cell sizes were
(20 adjectives per event). The ten adjectives rated as evenly distributed, resulting in a uniform 180 sub-
most useful in describing the event were selected for jects per cell assignment.
use in the experiment and are listed in Table 2.
Independent Variables
Experimental Procedure
Sponsorship condition. This condition consisted of
Three hundred and sixty undergraduate business two levels, one in which the event and brand were
students participated in the experiment for extra paired in a sponsor relationship (''yes sponsorship")
The Journal ofAdvertising
Table 2
Adjectives Used In Measurement of Image Transfer from Event to Brand
U.S. Open Golf Championship Indianapolis 500 Auto Race World Cup Soccer
(Image Based Similarity) (Functional Based Similarity) (No Similarity)
and one in which they were not ("no sponsorship"). In as a specific aspect of image. Recall that in pretest
the "yes sponsorship" level subjects viewed three black two, three groups of ten adjectives were selected as
and white photographs, each depicting one of the sport- being useful in describing each of the three events. In
ing events. Each picture was designed to resemble a the experiment each of the 10 adjectives were rated
magazine advertisement for the event and the actual logo as to how well it described the particular event (l=very
of the sponsoring brand was superimposed on the photo- well; 7=not at all) and then, separately, subjects re-
graph. In addition, ad copy typical of a sponsorship ar- sponded as to how well the same 10 adjectives de-
rangement was included on the photo (e.g., "Carnel ciga- scribed the brand. If an image transfer is occurring,
rettes is proud to sponsorWorld Cup Soccer '98'~. one would expect the image of the event and the
The cover page also added to the sponsorship ma- image of the brand to be more similar in the "yes
nipulation through the survey title, ''Product Spon- sponsorship" condition, as the event's image would be
sorship of Sporting Events Survey" and through the "transferring" to the brand. Accordingly, we calcu-
survey instructions which referred to the sponsorship lated a measure of congruence/similarity by taking
ties in the photographs (''Yes Sponsorship" instruc- the sum of the absolute differences between the cor-
tions: "We are conducting this study to better under- responding adjectives in the event and the brand rat-
stand how students feel about corporate sponsorship ings. For example, if a subject rated the adjective
of sporting events''). For the ''no sponsorship" level no ''mature'' as a "2" for the event and a "5" for the
photos were included; the survey was titled simply, brand, then the absolute difference for those corre-
"Image Survey," and mention of the sponsorship was sponding adjectives would be "3." We summed the
not included in the instructions ("No Sponsorship'" ten absolute difference scores for each event-brand
instructions: "We are conducting this study to better pair to create a congruence index. Smaller numbers
understand the images students have regarding dif- in the index indicate greater congruence (i.e., less
ferent events and products''). difference between event and brand). Hypothesis one
Similarity condition. As described in pretest 1, the would predict that this measure will be significantly
similarity condition was manipulated by pairing sport- smaller for those subjects in the ''yes sponsorship"
ing events and sponsoring brands to create three simi- level than in the ''no sponsorship" level-indicating
larity levels: image similarity (the U.S. Open Golf more similarity.
Championship and Acura Automobile), functional Holistic image transfer measure. Although the dif-
similarity (the Indianapolis 500 Auto Race and ference score method discussed above has been used
Goodyear Tires), and no similarity (World Cup Soccer by most studies examining self-image congruence,
and Camel Cigarettes). Sirgy et a1. (1997) have argued for a more parsimoni-
ous measure of congruence. These authors suggest
Dependent Measures that a method which directly measures congruency
using the respondent's own image dimensions and
Adjective based image transfer measure. As stated employing a holistic evaluation is more appropriate
above, this study examines brand/event personality for examining image congruence between a brand
Winter 1999 63
and one's own image. We extended this advice by image similarity event-brand combinations. Although
adapting the Sirgy et al. (1997) measure to study a significant difference does exist for the no similar-
event-brand congruence. Specifically, we offered the ity event-brand pairing, the means are the opposite
following instructions (adapted from Sirgy et al. 1997) of the hypothesized direction. That is, there is greater
and asked subjects to rate the consistency between congruence between World Cup Soccer and Camel
the event image and the brand image: cigarettes in the "no sponsorship" condition. Thus,
-rake a moment to think about the (sporting event Hypothesis One is supported by two of the three ho-
name). Think about the various images and expe- listic congruence measures. This unexpected pattern
riences one would encounter when they attended of results in the no similarity pairing is considered in
or watched this event. Imagine this event in your more detail in the discussion section.
mind and then describe the event using several Hypotheses 2a and 2b state that image transfer
adjectives such as: exciting, traditional, young, will be stronger in sponsor relationships when there
conservative, sexy, or whatever adjectiveByou think. is either functional (H2a) or image (H2b) based simi-
describe the image of this sporting event."
larity than when there is no similarity. Accordingly,
Subsequent to this mental imagery task, consistency these hypotheses were tested using only data from
was scored on a seven point scale (l=Strongly Agree, those subjects exposed to the sponsorship relation-
7=Strongly Disagree) keyed to the following ques- ship (data in the top row of Table 3). Since each
tion: ''My image of the (sporting event name) is con- subject responded to all three event-brand sponsor-
sistent with my image of (brand name)." ships these hypotheses were analyzed using a one-
way analysis of variance, repeated measures design.
Results The test reveals a significant effect for sponsorship
similarity using the summed absolute difference scores
Hypothesis One was analyzed using a one-way as the dependent congruence measures, F
MANOVA, between groups design. With sponsorship (2,328)=231.85; p <.0001. Contrasts show that con-
treatment (yes, no) as the between groups factor and
gruence of event and brand image is significantly less
the three image congruence scores (based on summed in the no similarity event-brand combination than in
absolute differences) as the criterion, a significant the functional similarity, F(l,164)=192.91; p <.0001
multivariate effect was found for sponsorship treat- (H2a) and image based similarity, F(l,164)=362.11; p
ment (Wilk's lambda=.89, F [3,324]=12.33; p< .0001). <.0001) (H2b) combinations. Further, although not
The sample means are displayed in Table 3. Tukey's hypothesized, congruence is stronger in the image
HSD test reveals that subjects who were exposed to based similarity pairing than in the functional based
the sponsorship arrangement had significantly lower combination, F(l,164)=60.41; p <.0001. Results using
difference scores (i.e., higher image congruence) than the holistic congruence measures adapted from Sirgy
those subjects not exposed to the sponsorship tie for et al. (1997) provide consistent results. Under the
both the functional similarity (Indianapolis 500 and holistic congruence measure, both image and func-
Goodyear tires) and image similarity (U.S. Open Golf tional event brand pairings are more congruent than
Championship and Acura automobiles) event-brand the no similarity event-brand pairing. However,
combinations. There was not a significant difference counter to the absolute difference measure, the holistic
between subjects in the ''yes sponsorship" level and the measure indicates transfer is higher in the functional
"no sponsorship" level with regard to the no similarity similarity condition than in the image similarity condi-
event-brand combination (World Cup soccer and Camel tion. As such, hypotheses 2a and 2b are supported.
cigarettes). Therefore, two of the three "absolute differ-
ence" congruence measures support Hypothesis One.
Hypothesis One was also tested using the holistic Discussion
image congruence measures adopted from Sirgy et a1.
(1997). Again, using a one-way MANOVA, between The image transfer process. The test of Hypothesis
groups design, we found a significant multivariate One indicates a transfer of image as evidenced by the
effect (Wilk'slambda=.92, F [3,353]=10.48; p< .0001) majority of the personality congruence scores. Impli-
for sponsorship treatment. As with the adjective based cations related to the image transfer findings fall on
congruence measure, the Tukey HSD test showed both sides of the sponsorship arrangement. First, on
significantly higher congruence in the ''yes sponsor- the product side, brand managers should be cogni-
ship" treatment level than in the ''no sponsorship" zant of the event's image when deciding upon the
treatment level for both the functional similarity and allocation of sponsorship dollars. Our results suggest
The Journal of Advertising
Table 3
Summary of Image Congruence Means and Image Transfer Strength
that brand positioning goals may be partially accom- impacted by discrepant information than those with
plished through event sponsorship. However, the po- weak product schemas (Peracchio and Tybout 1996).
tential to transfer an image inconsistent with posi- That is, for those with strong product schemas, new
tioning goals also exists. This implies that before en- information that is incongruent with existing
tering into sponsorship arrangements firms would be knowledge is distorted and discounted so as not to
advised to take a proactive role in measuring their upset the existing schema (Crocker, Fiske and Taylor
target consumers' event image perceptions in order 1984). Perhaps a similar phenomenon is at work in
to confirm that the event's image is consistent with our pairing of Camel cigarettes and the World Cup
the firm's brand positioning goals. On the event side Soccer Championship. In their work examining the
of the sponsorship arrangement, the image transfer effect of new information on attitude change, Sherif
findings suggest that an opportunity exists for event and Hovland (1961) found that in the case ofextremely
managers to differentiate their events based on the discrepant information, attitude changes often
image they convey. As such, an event management occurred in the opposite direction as that advocated
team looking to attract new product sponsors or en- by the discrepant communication (e.g., someone with
hance the value for existing sponsors might provide a positive attitude toward a product is exposed to
empirical evidence showing the various dimensions extremely negative information about the product and
of their event's image that could be used to enhance their product attitude becomes even more positive).
or maintain the image of the potential sponsor. In- In these cases the extremely discrepant product
deed, cultivating a particular image may allow the event information may be viewed as highly biased, and thus
to demand an added premium from sponsoring firms. individuals are apt to overcompensate with regard to
Of particular interest are the results found in con- their attitude toward that product. While the research
junction with the no similarity event-brand pairing in this area examines attitude change as opposed to
in both the adjective based and holistic measures. image transfer, their findings are similar to the pattern
Although only significantly different in the holistic displayed by our holistic congruence measure. That is,
measure, the means in both cases were the reverse of with respect to the no similarity condition (perhaps
the hypothesized direction for this condition. That is, more appropriately named dis-similarity), not only is
there is a higher congruence mean in the "no sponsor- event-brand congruence not higher in the sponsorship
ship" treatment than in the ''yes sponsorship" treat- condition, it is actually significantly lower. Perhaps
ment for the advertisement depicting Camel cigarettes those exposed to the cigarette-soccer pairing in the
sponsoring the World Cup Soccer Championship. sponsorship condition found the notion so incongruent
We speculate that this is an artifact of the type of that they pushed the images apart in their minds.
product used to operationalize the no similarity Event and brand match-up. Our results extend the
condition. Our manipulation may have inadvertently match-up hypothesis popularized in the celebrity en-
represented dis-similarity (i.e., the healthy image of dorsement literature to an event sponsorship con-
sport contrasted with the unhealthy image of text. The support of hypotheses 2a and 2b suggests
cigarettes) rather than simply a neutral/no similarity that firms can leverage their sponsorship dollars, in
condition. There is evidence to suggest that consumers terms of image transfer, by selecting events based on
with strong product schemas (i.e., elaborate amounts image or functional similarity. While there may be
of interconnected product knowledge) will be less times when other promotional objectives (e.g., brand
Winter 1999 6S
awareness) are the overriding goal, recent research tified sports fans. A third issue is our use of different
has begun to document the importance of image re- sets of instructions in the "yes" and "no" sponsorship
lated objectives to corporate sponsors (Abratt, Clayton conditions, as well as the use of photos only in the
and Pitt 1987; Hoek, Gendall and Sanders 1993; Irwin ''yes'' sponsorship condition. Given our research de-
and Sutton 1994). Indeed, in their framework for sign, we are unable to assess the influence, if any,
evaluating the attractiveness of sponsorship opportu- that these differences between conditions may have
nities, Irwin and Asimakopoulos (1992) describe im- had on the dependent variables above and beyond
age association as one of the six primary sport spon- establishing the event-brand linkage. Finally, we ex-
sorship objectives. The point is that if image transfer amined image congruence after the exposure to spon-
is of concern, then event selection should be made sorship stimuli. Thus, we assumed-rather than
with degree of similarity in mind. To enhance the tested-the direction of the image transfer. Accord-
strength of image transfer in cases of event sponsor- ingly, it is possible that the image transfer occurs
ship, it seems plausible that the marketing manager from the brand to the event rather than from the
may wish to alter the communication regarding the event to the brand. Examining the direction of the
product prior to the event, to be more congruent (on image flow and under what conditions it might be
either a functional or image basis) with the image of switched represents a fruitful future research direc-
the event. For example, a flurry of advertising linking tion. The use of known versus unknown brands might
the attributes of the event with the advertised product. be instrumental in understanding these effects (Tripp,
Our findings indicate that if the match between the Jensen and Carlson 1994).
event and product can be made stronger, then the re- Another interesting issue surrounding image trans-
sulting image transfer will be more pronounced. fer and related to degree of similarity is that of con-
flicting images among multiple sponsorships. It is
Limitations and Directions for not uncommon for firms to engage in multiple spon-
Future Research sorship relationships over the course of any given
year (Farrelly, Quester and Burton 1997). It may be
The results of this study should be considered in insightful to consider the effect to a consumer's brand
light of several constraints. First, due to concerns schema when a brand sponsors multiple events with
over our subjects' available time, we conducted this conflicting images. Would the image transfer cancel
experiment using one event-brand pairing per simi- out? Would the most recent image have the larger
larity type and a single exposure to the sponsorship impact? Here schema theory may prove useful for
stimuli. Other studies may wish to develop research providing insight. As discussed, discrepant informa-
designs allowing for a more robust treatment of the tion is less likely to lead to schema change because it
similarity condition utilizing multiple sponsorship is often discounted by the individual. However, a brand
ties. Although some experimental control may be lost, associated with multiple events, each having a differ-
a field experiment at an actual sporting event would ent and discrepant meaning, might lend credence to
provide a context in which multiple brands coupled the discrepant information, or at least call into ques-
with the dynamic environment of a realistic event tion the existing schema. Alternatively, social adap-
could be studied. Further, a field study would be able tation theory would suggest that discrepant event
to incorporate other elements that are difficult to associations would, presumably, not be effective
create in a "lab" setting (e.g., the influence of other sources of information and, therefore, would not fa-
spectators and the "secondary" nature of the sponsor- cilitate adaptation to environmental conditions (Kahle
ship association versus the "primary" concern of the and Homer 1985). In contrast, it may be possible for a
event itself). Second, our use of a student sample firm to positively leverage the image transfer by spon-
should be expanded to include non-student respon- soring multiple events with consistent images. Of
dents. For example, future studies might explore how course, the sponsorship of two (or more) events with
different ''types" of fans experience the image trans- conflicting images only becomes an issue when a given
fer process. Given the increasing amount of research targeted consumer group is aware of both sponsor-
on sports-fan team identification (e.g., Fisher and ship ties. Still, within a single event it may be possible
Wakefield 1998; Mael and Ashforth 1992; Wann and to have conflicting similarity types (i.e., functional and
Branscombe 1995; Wann and Dolan 1994), it would image based). For example, an interesting future re-
be interesting to test for differences on sponsorship search question could address how image transfer might
issues (e.g., image transfer, sponsor recall, sponsor be affected when an event-product pairing is high in
patronage, etc.) between high identified and low iden- functional similarity, but low in image similarity (and
68 The Journal ofAdverliBing
vice versa). Additionally, will image transfer be en- Hoek, Janet A, Philip J. Gendall and John Sanden (l993l, "SponBar-
ship Management and Evaluation: Are ManagoerB' A_unptionB
hanced when both types of similarity are present (and
Justified?," JOWTIDl ofPromotion M~, 1 (4),53-66.
consistent) in the sponsorship arrangement? Irwin, Richard and William A. Sutton (1994), -Sport Sponsorehip
A related issue is the impact of multiple sponsors of Objectives: An Analysis of their Relative Importance for Ma-
a given event. Gwinner (1997) suggests that the trans- jor Corporate Sponsors," European Journal for Sport Man-
fer of an event's image to a sponsoring brand will be CJIlement, 1 (2l, 93-101.
and Makis K. Asimakopoulos (1992), "An Approach
moderated by the exclusiveness of the sponsorship to the Evaluation and Selection of Sport Sponsorship Propel&-
arrangement as measured by the number of other als," Sport Marketing Quarterly, 1 (2), 43-51.
sponsors and the level of the sponsorship. He sug- Javalgi, ~shekhar G., Mark B. Traylor, Andrew C. Gross and
Edward Lampman (1994l, "AwareneBB of Sponsorship and
gests transfer will be less in instances of multiple Corporate Image: An Empirical Investigation," Journal of
sponsors and lower sponsorship levels (e.g., title spon- Aduertising, 23 (4l, 47-58.
sor versus perimeter fence signage). This is consis- Kahle, Lynn R. and Pamela M. Homer (1985l, "Physical Attrae-
tent with recent research in the celebrity endorser tiveneBB of the Celebrity Endorser: A Social Adaptation Per-
spective," Journal of Consumer Research, 11 (4l, 954-961.
literature which has found that the number of prod- Kamins, Michael A (1990l, "An Investigation into the 'Match-up'
ucts endorsed by a celebrity is negatively related to Hypothesis in Celebrity Advertising: When Beauty May be
endorser credibility, likability and attitude toward Only Skin Deep," Journal ofAduertising, 19 (1),4-13.
the ad (Tripp, Jensen and Carlson 1994). As such, and Kamal Gupta (1994l, "Congruence Between
Spokesperson and Product Type: A Matchup Hypothesis Per-
studies might explore how image transfer and brand spective," Psyclwlogy and Marketing, 11 (6l, 569-586.
attitudes are impacted by the number of sponsors Keller, Kevin Lane (1993l, "Conceptualizing, Measuring, and Man-
and the level of sponsorship. Further, although our aging Customer-Based Brand Equity," Journal ofMarketing,
results are supportive of the image transfer hypoth- 57 (1), 1-22.
Lee, Myung-Soo, Dennis M. Sandler and David Shani (1997l, "Atti-
esis, they do not provide insight into the enduring tudinal Constructs Toward Sponsorship: Scale Development
nature of this phenomenon. Additional studies are Using Three Global Sporting Events," International Market-
needed to assess the long term influence of event ing Reuieui, 14 (3l, 159-169.
sponsorship as it relates to image transfer. Lord, Robert G. and Roseanne J. Foti (1986l, "Schema Theoriell,
Informational Processing, and Organizational Behavior," in
The Thinking Organization: Dynamics of Organizational S0-
Reference cial Cognition, Henry P. Sims, Jr., Dennis A Gioia, and A8IlO-
dates, eds., San Francisco, CA: JOll88y-BaBB Inc.
Aaker, Jennifer L. (1997), "Dimensions of Brand Personality," Jour- Lynch, James and Drue Schuler (1994l, "The Matchup Effect of
nal ofMarketing Research, 34 (B), 347-356. Spokesperson and Product Congruency: A Schema Theory
Abratt, RUll88ll, Brian C. Clayton and Leyland F. Pitt (1987l, "Cor- Interpretation," Psyclwlogy and Marketing, 11 (5), 417-445.
porate ObjectiveB in Sporta Sponsorship," International Jour- Mael, Fred and Blake E. Ashforth (1992l, "Alumni and Their Alma
nal ofAduertiBing, 6, 299-311. Matter: A Partial Test of the Reformulated Model of Organi-
Bartlett. Frederic C. (1932l, Remembering: A Study in Experimen- zational Identification," Journal of Organizational Behaaior,
tal and Social Psyclwlogy, Cambridge, MA: Cambridge Uni- 13 (2l, 103-123.
vereity PreBB. Marshall, D.W. and G. Cook (1992l, "The Corporate (Sporta) Spon-
Cornwell, T. Bettina and Isabelle Maignan (1998l, "An Interna- sor," International Journal ofAduertiBing, 11,307-24.
tional Review of Sponsorship Research," Journal ofAduertis- McCracken, Grant (1989l, "Who is the Celebrity EndorlJllr? Cul-
ing, 27 (1), 1-21. tural Foundations of the Endorsement Procese," Journal of
Crocker, J., Susan T. Fiske and Shelley E. Taylor (1984l, "Sche- Consumer Research, 16 (3l, 310-321.
matic Bases of Belief Change," in Attitudinal Judgment, J. McDanie~ Stephen R. (l999l, "An Investigation of Match-up Eft'ec:tII in
Richard EilJllr,ed., New York: Springer-Verlag, 197-226. Sport Sponsorship Advertising: The Implications of Consumer
Crowley, Martin G. (1991l, "Prioritising the Sponsorship Audi- Advertising Schemas," Psyclwlogy & Marketing, 16 (2l, 163-184.
ence," European Journal ofMarketing, 25 (Ill, 11-21. McDonald, Colin (1991l, "Sponsorship and the Image of the Spon-
Farrelly, Francis John, PllIICA1e G. Quester and Richard Burton eor," European Journal ofMarketing, 25 (11), 31-8.
(1997l, "Integrating Sporta Sponsorship into the Corporate Meenaghan, Tony (1991l, -Sponsorship-Legitimising the Medium,"
Marketing Function: An International Comparative Study," European Journal ofMarketing, 25 (11l, 5-10.
International Marketing Review, 14 (3l, 170-182. Meerabeau, Elizabeth, Roy Gillett, Michael Kennedy, Johnson
FiBher, Robert J. and Kirk Wakefield (1998l, "Factors Leading to Adeoba, Michael Byas8 and Kingsley Tabi (1991l, -Sponsor-
Group Identification: A Field Study of Winners and Losers," ship and the Drinks Industry in the 1990s," European Jour-
Psychology and Marketing, 15 (1), 23-40. nal ofMarketing, 25 (Ill, 39-56.
Graeff, Timothy R. (1996l, "Image Congruence Effects on Product Misra, Shekhar and Sharon E. Beatty (1990l, "Celebrity Spokes-
Evaluations: The Role of Self-Monitoring and Public/Private person and Brand Congruence: An ABBeBBment of Recall and
Conaumption," Psyclwlogy &I Marketing, 13 (5l, 481-499. Affect," Journal ofBusiness Research, 21 (2l, 159-173.
Gwinner, Kevin P. (1997l, "A Model of Image Creation and Image Peracchio, Laura A and Alice M. Tybout (1996l, "The Moderating
Transfer in Event Sponsorship," International Marketing Re- Role of Prior Knowledge in Schema-Based Product Evalua-
uiew, 14 (3l, 145-158. tion," Journal of Consumer Retlearch, 23 (3l, 177-192.
Winter 1999 67
Plummer, JOllllph T. (1985), "How Penonality Makea a Differ- Tripp, Carolyn, Thomu D. Jenlllln and Lee Carleen (1994), ~
ence," JourruJ. ofAdvertising ReBearch, 24 (6), 27-31. Effects of Multiple Product EndorlM!mentll by Celebrities on
Rumelhart, David E. and Andrew Ortony (1977), '"Ihe Representation CoDBUJDera' Attitudea and Intentione," Journal of CoRllumer
« Knowledg8 in Memory," in Schooling and the Acquisition of Research; 20 (4), 535-547.
KnDwletJile, RidJard C. AnderBon, Rand J. Spiro, and William E. Wann, Daniel L. and Thomu J. Dolan (1994), -AttributioDII of
Montague, eds., HilladaIe, NJ: Lawrence Erlbaum Associates. Highly Identified Sparta Spectatora, n The Journal of Social
Sherif, Muzafer and Carl I. Hovland (1961), Social Judgerrum.t: Psychology, 134 (6), 783-792.
AAimilation and ContrCJI/t Effectll in Communication and At- - - - - and Nyla R. Branscembe (1995), -Influence ofIden-
titude Change, New Haven: Yale Univeraity PreB8. tification with a Sparta Team On Objective Knowledge and
Sirgy, M. JOIIIlPh, Dhruv Grewal, Tamara F. Mangleburg, Jae-ok Subjective Beliefs," International Journal of Sportll PII;yChol-
Park, Kye-Sung Chon, C.B. Claiborne, J.S. Johar and Harold ogy, 26 (Oct.lDee.), 551-567.
Berkman (1997), -AueB8ing the Predictive Validity of Two
Methode of Mll&IIUI'ing Self-Image Congruence," Journal of
the Academy ofMar1ceting Science, 25 (3), 229-241.