1 s2.0 S1441352313000715 Main

G Model
SMR-243; No. of Pages 14

Sport Management Review xxx (2013) xxxxxx
Contents lists available at ScienceDirect
Sport Management Review

journal homepage: www.elsevier.com/locate/smr
The effect of familiarity on associated sponsor and event

brand attitudes following negative celebrity endorser
publicity
Jason P. Doyle a,*, Robin D. Pentecost b, Daniel C. Funk a,c
a
b
c
Department of Tourism, Sport & Hotel Management, Grifth University, Australia

Department of Marketing, Grifth University, Australia
School of Tourism & Hospitality Management, Temple University, United States
A R T I C L E I N F O
A B S T R A C T
Article history:
Received 28 January 2013
Received in revised form 21 October 2013
Accepted 21 October 2013
This research examined how familiarity with a brand inuences how negative publicity
related to a celebrity endorser is perceived. Specically, the current research determined if
familiarity with event and sponsor brands may temper any negative consequences of
being linked with negative celebrity endorser publicity. Two studies were conducted to
investigate this aim. Study 1 (n = 136) used unfamiliar brands and indicated attitudes
towards the selected sponsor and the event brands were signicantly reduced after
negative publicity surrounding an associated celebrity endorser emerged. In contrast,
individuals not exposed to negative publicity did not report reduced attitude scores. Study
2 (n = 272) used unfamiliar and familiar brands and found that negative publicity
surrounding celebrity endorsers has the capacity to weaken attitudes towards associated
event and sponsor brands. However, any negative impact was tempered by an individuals
familiarity with the respective brand, indicating brand familiarity has a moderating effect
on brand attitudes within the sports marketplace. This research furthers understandings
related to sport sponsorship theory by assessing the role of brand familiarity in tempering
information transference. Additionally, the paper provides insights to the events category
which has not been researched previously. Brand managers can use this information to
develop proactive and reactive strategies to employ to protect their brands when celebrity
endorsers attract negative publicity.
2013 Published by Elsevier Ltd on behalf of Sport Management Association of
Australia and New Zealand.
Keywords:
Negative publicity
Celebrity endorsers
Sport sponsorship
Event marketing
Attitude change
1. Introduction
Sponsorship is an important area of sport consumption with sponsors investing substantially to associate with sport
brands (Farrelly, 2010; Hughes & Shank, 2005). For example, in 2010 the Qatar Foundation spent US$200 million to establish
a ve-year sponsorship deal with Spanish football club Barcelona F.C. (BBC, 2010). Elsewhere, Molson Canadian recently
became the ofcial beer of the National Hockey League (NHL) as part of a seven-year sponsorship agreement valued at
approximately US$375 million (The Canadian Press, 2011). Such examples indicate sponsors see value in strategic alignment
* Corresponding author at: Grifth Business School, Building G27, Grifth University Gold Coast Campus, Parklands Drive, Southport, Gold Coast 4215,
Australia. Tel.: +61 7 555 27672; fax: +61 7 555 28507.
E-mail addresses: jason.doyle@grifth.edu.au, jasondoyle87@gmail.com (J.P. Doyle), r.pentecost@grifth.edu.au (R.D. Pentecost), dfunk@temple.edu
(D.C. Funk).
1441-3523/$ see front matter 2013 Published by Elsevier Ltd on behalf of Sport Management Association of Australia and New Zealand.
http://dx.doi.org/10.1016/j.smr.2013.10.003
Please cite this article in press as: Doyle, J.P., et al., The effect of familiarity on associated sponsor and event brand
attitudes following negative celebrity endorser publicity. Sport Management Review (2013), http://dx.doi.org/10.1016/
j.smr.2013.10.003
G Model

2
J.P. Doyle et al. / Sport Management Review xxx (2013) xxxxxx
with sport brands. From this perspective, sponsorships act as part of overall marketing strategies which help sponsors to
achieve their objectives through interorganisational linkages (e.g., Cousens, Babiak, & Bradish, 2006).
However, sponsorship is not without risks (Hughes & Shank, 2005; Till, Stanley, & Priluck, 2008). When entering into
sponsorship agreements, a primary objective for the sponsor is to create a favourable image and leverage its brand (Cornwell,
Weeks, & Roy, 2005). This leverage is achieved through positive brand perceptions created by associations with the
sponsored property (Gwinner & Eaton, 1999; Keller, 2003). However, the transference of positive associations to the sponsor
may not always occur because of factors outside of the control of the sponsoring brand (Fahy, Farrelly, & Quester, 2004; Till &
Shimp, 1998). One element of the sponsorship relationship largely outside the control of sponsors is the use of celebrity
endorsers, which are prominently used in advertising to leverage sponsorships (Agrawal & Kamakura, 1995).
Previous research indicates celebrity endorsement can result in an increase or decrease in favourable attitudes elicited
towards an endorsed brand based on the actions of the endorser (Till & Shimp, 1998; Till et al., 2008). To date, scholars have
focused on the range of positive outcomes emanating from successful sponsorship scenarios, such as links with successful
endorsers. Less attention has been paid to understanding possible negative outcomes which may result from instances
where endorsers are linked with unfavourable actions (e.g., Farrelly, 2010; Hughes & Shank, 2005; Zhou & Whitla, 2013). In
the face of rapid technological advancements and increased media dedicated to celebrity endorsers, a greater
understanding of the possible negative outcomes becomes a necessity (Zhou & Whitla, 2013). Although research detailing
the consequences of such negative publicity has commenced (e.g., Funk & Pritchard, 2006; Hughes & Shank, 2005; Till &
Shimp, 1998; White, Goddard, & Wilbur, 2009; Wilson, Stavros, & Westburg, 2008), opportunities to extend this line of
research exist.
Through this research we expand current sponsorship knowledge in two ways. First, based on calls in existing research
(e.g., Till & Shimp, 1998), we investigate how brand familiarity may mitigate attitude change in instances where negative
publicity related to a celebrity endorser arises. Second, we explore implications for a range of associated brands (e.g., event)
linked to the celebrity endorser, rather than the endorser or a directly promoted sponsor brand as per previous research
designs (White et al., 2009). Through this research it is hoped we can provide a better understanding of the potential effects
of negative celebrity endorser publicity for a greater range of stakeholders.
The manuscript consists of the following sections. First, a review of celebrity endorsement and the possible implications
which may arise from negative celebrity endorser publicity is provided. Second, theories of attitude formation which may
explain how negative publicity surrounding an endorser may impact associated brands are presented. Third, potential
factors which may mitigate the transference of negative attitudes onto associated brands are discussed. Fourth, an outline of
the method used to guide the current research is provided. Fifth, a presentation of results and a discussion of the ndings is
included. Sixth, implications for theory and practice are outlined. Finally, an acknowledgement of relevant research
limitations and directions for future research are offered.
2. Literature review
2.1. Celebrity endorsement
Celebrity endorsement involves linking a prominent person to a brand and is an effective component of sponsorship
(Agrawal & Kamakura, 1995; Erdogan, 1999). Sport sponsorships consist of exchanges between the sponsor and the
sponsored property which are inuenced by political, social and economical factors, as well as the actions and reactions
which take place within the specic environment (Cousens et al., 2006; Nickell, Cornwell, & Johnson, 2011). Thus,
sponsorship decision making must consider a broad range of stakeholders which have the potential to inuence sponsorship
effectiveness, including celebrity endorsers. Given the high prole and marketability of sport, many sponsors seek to
associate with sportspeople to act as celebrity endorsers and promote their brands.
Star athletes are often directly targeted and approached to promote sponsor brands (e.g., Fink, Parker, Cunningham, &
Cuneen, 2010); however celebrity endorsement may also take a less direct route. For instance, sponsors of Australias
National Rugby League (NRL) and Australian Football League (AFL) competitions are inextricably linked with star players
such as Greg Inglis and Gary Ablett given their prominent status within their respective codes. Similarly to sport leagues and
teams, sponsors of sport events such as Formula 1 racing or MotoGP have strong links with gureheads such as Mark Webber
and Casey Stoner. Consequently, sport sponsorship represents a complex exchange whereby sponsors enter into
relationships with not only the sport brand itself, but also its associated assets (e.g., Chanavat, Martinent, & Ferrand, 2009;
Cousens et al., 2006).
Brand managers are particularly wary of damaging associations, such as negative press surrounding associated athletes,
being linked to their brands (Till & Shimp, 1998). Such incidences may be seen as a risk to the brand because they may
negatively inuence stakeholder attitudes (Agyemang, 2011; Erdogan, 1999; Ferrand & Pages, 1999). When negative
publicity surrounding celebrity endorsers emerges, sponsors often detach themselves from sponsorship agreements due to
concerns of becoming tainted by association (Hughes & Shank, 2005). However, this is not always the case. For instance, the
Tiger Woods indelity scandal illustrated such a scenario where some sponsors (e.g., Tag Heuer, Accenture & Gillette)
terminated their sponsorship arrangement whilst others (e.g., Nike, EA Sports & Top Deck) did not. Similar discrepancies
were evident amongst sponsors associated with Lance Armstrong at the time he had his Tour de France titles stripped in
2012. The varied decision making reported in the above cases indicates sponsors know little about how their brands may be
j.smr.2013.10.003
G Model

affected by links with negative celebrity endorser publicity, nor how to respond to negative endorser acts. Thus, further
research is needed to understand the potential impacts of negative celebrity endorser publicity.
2.2. Effects of negative information on attitudes
To examine how negative publicity may inuence attitudes, we must rst consider how attitudes are formed. Attitude
formation is organised around the processing of the relevant information (Fiske & Taylor, 1991). Seminal studies suggest
attitudes are formed using heuristic-systematic processing (Chaiken, 1980; Chaiken, Liberman, & Eagly, 1989), and
peripheral-central routes to persuasion (Petty & Cacioppo, 1986), based upon a persons motivation and opportunity (Fazio,
1990). A common point of these theories is that perceptions of an object are inuenced by the context in which it is found
(Kahneman, 2011). Such a context provides the bounded rationality (Simon, 1982) whereby people use available information
to compromise between the desire to make a correct judgement and the desire to minimise the effort required (Fazio, 1990;
Payne, Bettman, & Johnson, 1993).
Sponsorship is based on this bounded rationality in that sponsors wish to associate with entities that carry positive
associations, so that such associations can be attributed to their own brands (e.g., Gwinner & Eaton, 1999). Researchers posit
that attitudes can be shaped through the transference of information in this manner (Fazio, 1989; Fiske & Taylor, 1991) as
attitudes contain predetermined associations in the mind of the consumer which inuence judgement and behaviour
(Zimbardo, Ebbesen, & Maslach, 1977). From this perspective, associations attributed to sportspeople (e.g., Adam Scott) may
be transferrable to the events in which they participate (e.g., US Open Golf tournament) and also to sponsors linked with the
event (e.g., Nike). One key objective for sponsors is to associate with entities that elicit favourable reections onto their
brands. However, celebrity endorsers can do as much harm as good in terms of inuencing consumer attitudes (Erdogan,
1999; Miciak & Shanklin, 1994; Silvera & Austad, 2004).
Considering this perspective, both positive and negative publicity surrounding an endorser may inuence attitudinal
evaluations linked with associated event and sponsor brands. Previous research has examined what constitutes a sport
scandal and its severity (Hughes & Shank, 2005) and how organisations can deal with such transgressions (e.g., Fink, Parker,
Brett, & Higgins, 2009; Wilson et al., 2008). Such research has demonstrated that negative publicity surrounding a celebrity
endorser may result in a decrease in team identication (e.g., Fink et al., 2009) and in attitudes held towards related brands.
For example, Till and Shimps (1998) research using ctional brands, found that a celebrity sportspersons link with negative
publicity (in this case driving and drug offences) weakened consumer attitudes towards an endorsed bicycle brand. More
recent research corroborated this nding illustrating how attitudes towards a ctional athletic shoe were signicantly
weaker for individuals who were exposed to negative publicity concerning an endorser of the shoe brand compared to those
who were not provided with the negative information (White et al., 2009).
These studies provide a starting point to understand the inuence of negative celebrity endorser publicity on brand
attitudes and offer opportunities to further this line of research. Given White and colleagues (2009) call to assess how
information relating to celebrity endorsers may impact a variety of associated brands, opportunities exist to replicate and
extend this line of research into new areas. For example, no enquiry has yet investigated the impact of celebrity endorser
negative publicity on consumer attitudes towards an event, to date preferring to focus on directly promoted brands. As sport
events provide the context in which the celebrity is viewed (e.g., Kahneman, 2011; Simon, 1982), it is likely that negative
information surrounding celebrity endorsers may inuence sponsor and event attitudes. Thus, the following hypotheses are
presented to investigate the relationship between celebrity endorser negative publicity and attitudes towards associated
sponsor and event brands:
H1. Negative celebrity endorser publicity will signicantly weaken the attitudes consumers hold towards associated
sponsor brands.
H2. Negative celebrity endorser publicity will signicantly weaken the attitudes consumers hold towards associated event
brands.
2.2.1. Factors inuencing negative publicity processing
As sport sponsorships are inuenced by environmental inuences and actions of associated stakeholders (e.g., Cousens
et al., 2006) it is also important to consider factors which may temper any possible effects of celebrity endorser negative
publicity. Previous research suggests that well established brands are less likely to see signicant changes from
sponsorship; whereas those with less established brand attitudes are more likely to expect a greater impact from
sponsorships (Nickell et al., 2011). Similarly, Speed and Thompson (2000) found that pre-existing attitudes were also
related to attitudinal favourability measured after sponsorship actions had taken place. This suggests that brand awareness
is likely to inuence the processing of new information. As described above, individuals may encounter new information
which reects positively on the stakeholders within a sport sponsorship relationship and also that which carries a negative
valence.
Previous research indicates that an individuals prior perceptions towards a brand will impact upon how they
subsequently process negative information surrounding that brand (Ahluwalia, Burnkrant, & Unnava, 2000; Funk &
Pritchard, 2006). Ahluwalia and colleagues (2000) tested respondents reaction to negative information presented about
j.smr.2013.10.003
G Model

4
athletic shoes. Results from two experiments revealed that commitment (a form of pledge to a particular course of action;
see Gill, Grossbart, & Laczniak, 1988) towards the brand moderated how the negative publicity was processed, which in turn
inuenced whether or not brand attitudes were affected. The authors found that consumers with high commitment
dismissed any negative information presented about the brand, preferring instead to preserve their established attitudes.
Conversely, low committed consumers were less likely to argue with the information presented and instead based their
attitudes towards the brand in accordance with the valence of the stimulus material (Ahluwalia et al., 2000).
Extending this research to a sports context, Funk and Pritchard (2006) investigated the impact of commitment on how
consumers processed various forms of publicity (positive, neutral and negative) surrounding a Major League Baseball (MLB)
franchise. Findings revealed existing attitudes moderated the effects of these various forms of publicity. Specically,
individuals with low team commitment had their beliefs and feelings about the team positively or negatively inuenced
after reading a news release in accordance with the messages valence. Low committed respondents who read a neutral
release and highly committed individuals exposed to positively, negatively, or neutrally worded releases did not change
their attitudes towards the team after reading the stimulus material (Funk & Pritchard, 2006). This study illustrated how
highly committed individuals are more inclined to reject conictive messages that do not conform to their established
attitudes than individuals with low commitment, who based their attitudes more heavily on new information they
encountered (e.g., Eagly & Chaiken, 1995; Sherif, 1963). Extending this work, within the present research we propose to test
the effect of another related inuencing construct, familiarity (knowledge of a brand within a product class, including
product use contexts and attributes; see Gill et al., 1988; Raju & Reilly, 1979), and its effect on information processing
surrounding negative celebrity endorser publicity.
2.2.2. Familiarity
Previous scholarly work investigating the effects of negative publicity surrounding celebrity endorsers has called for
familiarity measures to be included in future research (Till & Shimp, 1998; White et al., 2009). Consumers can be separated
into novices and experts based upon their degree of product knowledge with such familiarity inuencing information
processing and brand evaluations (Alba & Hutchinson, 1987; Fazio, 1989). Familiarity thus impacts brand recall and
recognition, and assists with the formation of consideration sets (Samu, Krishnan, & Smith, 1999). Therefore, familiarity
plays a major part in attitude formation and in processes which impact subsequent decision making (Bettman & Sujan,
1987).
Research on attitude formation has shown attitudes towards less familiar brands are weaker in both strength and
accessibility than attitudes towards more familiar brands (Fazio, 1989). A persons familiarity with a brand has been linked
with positive brand attitudes (Bettman & Sujan, 1987) that are well established and likely to remain stable (Simonin & Ruth,
1998). Thus, for unfamiliar brands, consumer attitudes are less likely to be established making them more susceptible to
change (Simonin & Ruth, 1998). Familiarity applies to the sports marketplace as knowledge is the rst step in achieving
image transfer in sponsorship (Smith, 2004) and is a primary driver behind behaviour towards brands (Keller, 2003).
Although it is acknowledged information transference may be moderated by a consumers prior experience with brands (Till
& Shimp, 1998; White et al., 2009), opportunities exist to uncover the extent to which familiarity mitigates the effects of
negative celebrity endorser publicity. The following hypotheses test the inuence of familiarity on negative publicity
information transference within a sport sponsorship context:
H3. Increased familiarity with a sponsor brand will reduce the likelihood that attitudes towards the brand will be decreased
by negative publicity surrounding an associated celebrity endorser.
H4. Increased familiarity with an event brand will reduce the likelihood that attitudes towards the brand will be decreased
by negative publicity surrounding an associated celebrity endorser.
2.2.3. Co-variates of familiarity
Other factors related to familiarity also inuence consumer decision making. For instance, a study of attitude change
based on different advertising messages found when both objective and subjective messages are communicated (as may be
found in a press release) brand evaluations were directly related to the consumers involvement but not necessarily
familiarity with the brand (Gill et al., 1988). That is, involved consumers are inclined to note and make more analytic use of
subjective or objective information irrespective of familiarity. This can be especially signicant for information regarding
celebrity endorsers as involvement is important in distinguishing opinion leaders from non-opinion leaders (Chan & Misra,
1990). Thus, whilst familiarity has a signicant inuence on brand evaluations, other related factors can also signicantly
inuence brand attitudes, especially when it comes to advertising messages (Gill et al., 1988).
2.2.4. Involvement
Involvement may be conceptualised as a complex evaluation of an objects relevance according to an individuals
characteristics, his or her perception of the object and the situation within which it is found (Day, Stafford, & Camacho, 1995).
High involvement is associated with positive attitudes (Zaichkowsky, 1985) and has been found to also play a crucial role in
inuencing consumer comprehension and attention levels (Celsi & Olson, 1988). More recently, the utility of the construct to
explain attitudes or behaviours in sport and leisure settings has increased, with scholars suggesting involvement is crucial in
j.smr.2013.10.003
G Model

determining outcomes such as brand loyalty (Bennett, Ferreira, Lee, & Polite, 2009; Havitz & Mannell, 2005). Therefore it is
reasonable to presume that, similar to products, attitudes towards relevant events and sponsors would be inuenced by the
level of consumer involvement with the event.
2.2.5. Fit
Another potential confound that inuences consumer decision making is the degree of congruency of the event and
sponsor brand. This congruency, or t, describes any associative judgement between the event and the sponsor (Johar &
Pham, 1999). Research contends that the t which exists between sponsors and sponsored brands impacts the effectiveness
of sponsorships (e.g., Gwinner & Bennett, 2008; Nickell et al., 2011; Speed & Thompson, 2000) and it is suggested these
associations increase brand knowledge and inuence perceptions (Keller, 2003). Highlighting the importance of t, recent
research indicates these perceptions inuence not just an organisations single sponsorship but also its portfolio of
sponsorships (Chien, Cornwell, & Pappu, 2011). Based upon these associations, consumer judgements may be founded on
anything from tangible physical attributes associated with the event (e.g., golf clubs and golf), to the image associated with
the event (e.g., prestige cars and golf). Associations can also be founded upon benets offered through usage of the product
(e.g., watching the event on a television made by a television company sponsor provides greater picture and sound clarity)
(Pentecost & Spence, 2009). With brand familiarity including product use contexts and attributes (e.g., Gill et al., 1988; Raju &
Reilly, 1979), such associations between the event and the sponsors brand are therefore likely to inuence brand attitudes,
again confounding the effect of familiarity.
2.3. The use of involvement and t as familiarity covariates
What is important from the above literature regarding involvement and t is their potential to impact the inuence of
familiarity. For example, a person may be familiar with a sports event but have low involvement with the event or its
associated properties. An individual who knows about the Australian Open tennis event but has never attended and does not
know any of the events sponsors would t this description. A person may also be familiar with an event and sponsor brand
relationship but not like their t. From this perspective, Harley Davidson sponsoring the Australian Open tennis would not
provide an easily digestible t. Therefore, not accounting for such variables in familiarity research may result in skewed
results due to their known inuence on decision making in the events marketplace (Cornwell & Maignan, 1998; Ferrand &
Pages, 1999).
To ignore the confounding inuence of involvement and t on familiarity is to increase the potential to provide a source of
variance in any experimental error calculation (Nunnally & Bernstein, 1994). One way to assess the inuence of involvement
and t is to treat them as covariates and to separate their individual effects, thus allowing for more sensitive tests in the
treatment of familiarity (e.g., Hair, Black, Babin, & Anderson, 2009; Nunnally & Bernstein, 1994). To keep the research centred
on familiarity and to also account for the confounding inuence of involvement and t, the following hypotheses are
developed:
H5. Increased familiarity (when the covariates of involvement and t are accounted for) with a sponsor brand will reduce the
likelihood that attitudes towards the brand will be decreased by negative publicity surrounding an associated celebrity
endorser.
H6. Increased familiarity (when the covariates of involvement and t are accounted for) with an event brand will reduce the
likelihood that attitudes towards the brand will be decreased by negative publicity surrounding an associated celebrity
endorser.
Fig. 1 depicts a model of our summarised hypotheses. The model has been derived from the above literature review and
conceptualises how negative publicity surrounding a celebrity endorser may inuence attitudes towards associated sponsor
and event brands, alongside the factors which may moderate such effects.
3. Method
Two studies framed within a motorsport event context were conducted to test our hypotheses. Study 1 was a pilot study
designed to test the manipulation effects of our negative stimulus material and to assess Hypotheses 1 and 2. The goal of
Study 1 was to examine if negative celebrity endorser publicity led to reductions in the favourability linked with associated
sponsor and event brands. Study 1 was conducted in a controlled classroom setting and followed a quasi-experimental
research design which collected attitudinal data before and after an experimental manipulation. To remove familiarity
effects, the target event and sponsor brand (in this case an energy drink brand) included in the study were ctional. The
celebrity endorser depicted in the manipulation was also invented for the same reason. The event and the sponsor brands
created were designed to represent typical brands within their respective categories, whilst the celebrity endorser was
described as a champion in the sport. These decisions were made to increase the believability of these manufactured
entities. To represent negative celebrity endorser publicity, a press release which explained that the celebrity endorser had
been arrested for speeding and driving under the inuence of alcohol was developed to act as the stimulus material
assessed.
j.smr.2013.10.003
G Model
[(Fig._1)TD$IG]
Attitudes
towards
Sponsor
H1
H3
Sponsor /
Event
Familiarity
Negative Celebrity
Endorser Publicity
Involvement
and Fit
Covariates
H5
H6
H4
Attitudes
towards Event
H2
Fig. 1. Conceptual overview and summary of hypotheses.
Study 2 used a full 2 2 (event familiarity high/low sponsor familiarity high/low) experimental design to assess the
potential impact of familiarity in tempering any possible effects of the negative manipulation. To maintain consistency
between the studies, the ctional event and sponsor used in Study 1 were retained to represent the unfamiliar brands in
Study 2. These ctional brands were accompanied by a real event and a well-known sponsor brand from the same categories
as the ctitious brands used. The use of these real brands enabled the inuence of familiarity to be introduced to the research
and extended the results to the real world, a point noted as lacking in previous research (e.g., Till & Shimp, 1998). To maintain
consistency between the studies, the celebrity endorser and the negative publicity stimulus material depicted in Study 1
were both used in Study 2. The addition of real brands in Study 2 enabled Hypotheses 36 to be tested.
3.1. Study 1 participants
Participants were undergraduate students from an Australian University located in a region which hosts a large annual
motorsport event. Invitations to participate in the research were distributed in class. In total, 136 individuals voluntarily
agreed to participate in the research. The experimental group was comprised of 77 individuals; whereas the control group
consisted of 59 individuals. The decision to recruit students was made due to accessibility and the time and cost benets
associated with student sampling (Ferber, 1977). Ferber (1977) suggests convenience sampling can be detrimental to
research results if the sample may provide a limited representation of the greater public. Recruitment of a student sample
was deemed appropriate for this study as the objective of Study 1 was to yield basic research results, as opposed to more in
depth causal or applied research. Demographics revealed the majority of participants were female (57.5%) and aged between
18 and 24 years (51.7%), which was representative of the student population. The experimental group and the control group
did not signicantly differ in terms of their demographic makeup.
3.2. Study 1 instrumentation and procedures
Data were collected using online surveys which collected demographic information and assessed respondent attitudes
before, and after, the introduction of the stimulus material. The surveys were made available for a period of two weeks and
required between 7 and 10 min to complete. Consistent with past research using press releases to elicit response (e.g., Fink
et al., 2009; Funk & Pritchard, 2006; Till & Shimp, 1998) all participants rst read a press release informing the reader of the
event, sponsor and celebrity endorser. To increase the believability of the ctional entities used, the sponsor and event were
described as new brands and the endorser was noted as being a champion in the sport. To ensure each respondent read the
press releases carefully, spelling and grammatical mistakes were entered into each release. Respondents were asked to
identify these errors as a means to increase internal validity and to reduce hypothesis guessing (e.g., Cook & Campbell,
1979).
After respondents had read the initial press release, their attitudes towards the event and sponsor brands were measured.
Three items from an established scale were adapted to capture these perceptions (Ruth & Simonin, 2003). All items were
measured using seven-point Likert scales anchored with strongly disagree (1) to strongly agree (7). Following the gathering of
these pre-manipulation scores, respondents were then asked to read a second press release which carried either a neutral
(control group) or a negative valence (experimental group). The negative press release described how the celebrity had been
arrested for driving offences whilst under the inuence of alcohol. The neutral press release paraphrased the rst release
indicating the event and the brand sponsor were looking forward to a successful event. After reading the second press
release, respondent attitudes were again assessed enabling the inuence of the stimulus material to be examined.
j.smr.2013.10.003
G Model

3.3. Study 1 analysis

Data analysis consisted of the following three steps. First, the data were subjected to descriptive analyses to ascertain the
demographic makeup of the sample. Second, reliability tests were run to ensure the items included were dependable
measures of their intended constructs. The attitudinal scale used displayed acceptable reliability scores (a > .70) across the
experimental and the control group for both the event and the sponsor brand (Nunnally & Bernstein, 1994). The related items
were subsequently aggregated to generate a single mean score which represented respondent attitudes towards the event
and sponsor brands. Third, paired samples t-tests were used to test Hypotheses 1 and 2. The use of paired samples t-tests is
appropriate within quasi-experimental designs and in instances where pre and post manipulation scores are compared
(Field, 2005; Field & Hole, 2003). A discussion of the Participants, Instrumentation and Analysis used in Study 2 is now
included.
3.4. Study 2 participants
Participants in Study 2 were sourced through an external market research company with links to Australias national
postal service. The decision to recruit respondents from the general public was made to increase the generalisability of the
results and based on recommendations outlined by previous scholars (e.g., Till & Shimp, 1998). Invitations to participate in
the research were emailed to 1219 individuals identied as residents of the geographic region in which Study 1 took place.
Two-hundred and seventy-two individuals agreed to participate in the research, representing a response rate of 22.3%. All
respondents were randomly assigned to one of four experimental conditions. Condition 1 (n = 68) included the ctitious
event and ctitious sponsor (low event and sponsor familiarity), Condition 2 (n = 66) included the ctitious event and real
sponsor (low event, high sponsor familiarity), Condition 3 (n = 69) included the real event and real sponsor (high event and
sponsor familiarity), and Condition 4 (n = 69) included the real event and ctitious sponsor (high event, low sponsor
familiarity).
Respondent demographics revealed participants were predominantly female (59.6%) and primarily aged between 25 and
64 years old (76%). Although this sample may not reect the typical motorsports fan, such a demographic provides a sample
drawn from the real world (Aronson, Brewer, & Carlsmith, 1985). Whilst the use of realism in research versus the articiality
of laboratory settings may be disputed, such an argument is based upon the given situation, the specic questions asked and
the kind of answers sought (Pentecost & Andrews, 2010). Given that the research sought to describe what happens in the real
world, rather than does it happen at all, the use of such a sample was deemed to provide benets such as increased
generalisability of ndings (e.g., Pedhazur & Schmelkin, 1991).
3.5. Study 2 instrumentation and procedures
Building upon Study 1, Study 2 employed a 2 2 factorial research design to represent differing scenarios of familiarity
using ctitious and real sponsor and event brands. The surveys were active for two weeks and required between 7 and
10 min to complete. The same celebrity endorser and press releases from Study 1 were retained for use in Study 2. To test
familiarity effects, the ctitious brands used in Study 1 were retained to represent low familiarity brands, whilst an authentic
national motorsport event and a leading energy drink brand were included to represent high familiarity brands. The use of
authentic brands introduced a higher level of generalisability and applicability to the research, a point lacking in previous
studies (Till & Shimp, 1998).
Three items adapted from Ruth and Simonin (2003) were again used to measure attitudes towards the target brands.
Study 2 also included items to measure brand familiarity, t and involvement, which enabled the remaining hypotheses to
be tested. Familiarity was measured using two items from Simonin and Ruths (1998) scale; whereas respondent perceptions
of t between the event and the sponsor brand were examined by including three items previously used by Gwinner and
Eaton (1999). Finally, four items adapted from Speed and Thompson (2000) were included to measure involvement with the
event. All items were operationalised on seven-point Likert scales anchored with strongly disagree (1) to strongly agree (7). For
a list of the items used and the respective reliability scores for the items used to measure each construct, please see
Appendix A.
Involvement with the sponsor brand was not measured as the sponsor brands used (both real and ctional)
operate in the fast moving consumer goods categories, which are known for their low involvement in decision
making (e.g., Kotler, Burton, Deans, Brown, & Armstrong, 2013). Thus, we made the decision to measure only
involvement with the event, a marketplace using more exclusively targeted communications. The t between
the celebrity endorser and the event brands depicted were not measured as the press release used in both studies
described the endorser as a champion in the sport. Hence, a t between the event and the endorser was implied within
the studies.
3.6. Study 2 analysis
The analyses used in Study 2 involved the following ve steps. First, demographic information was assessed to determine
the samples characteristics. Second, reliability tests were run to verify that the items included were reliable measures of
j.smr.2013.10.003
G Model

their intended constructs (Nunnally & Bernstein, 1994). The items used to measure attitudes, familiarity, involvement and t
all displayed suitable reliability in each condition group and thus were aggregated to generate a mean score for each
construct. Third, the mean score of familiarity for each brand was inspected to ensure that the ctional brands were low in
familiarity and the authentic brands were high in familiarity. Fourth, the relationship between familiarity, involvement and
t was inspected through the use of bivariate correlation analyses. Finally, Multivariate Analysis of Covariates (MANCOVA)
tests were run to test Hypotheses 36.
MANCOVA analysis was chosen to account for the inuence of potential covariates and to isolate the effects of the
main construct of interest (e.g., Hair et al., 2009; Nunnally & Bernstein, 1994). In this case, the researchers were
interested in the inuence of familiarity, and therefore involvement and t were treated as covariates. As the
experimental design did not include covariates as treatment factors, MANCOVA allowed any variance attributable to
these variables to be accounted for prior to the calculation of the inuence of the dependent variables (Hair et al., 2009).
Partial eta squared (Z2) was used to test for strength of association between the independent variables (IV) and the
dependent variables (DV) and to determine the portion of the variance in the DV which could be explained by each IV
(Tabachnick & Fidell, 2012).
4. Results
4.1. Study 1 results
Results provided support for Hypotheses 1 and 2 (see Table 1). For the experimental group, attitudes towards both the
sponsor (pre-test M = 4.05, SD = 1.20; post-test M = 3.47, SD = 1.20) t(163) = 3.18, p < .01 and the event (pre-test M = 4.02,
SD = 1.19; post-test M = 3.51, SD = 1.28) t(163) = 2.74, p < .01 decreased signicantly after participants were exposed to the
negative publicity. In contrast, no signicant differences (p > .05) in attitudes towards the event or the sponsor were found in
the control group. These results indicate the negative celebrity endorser publicity weakened brand attitudes and suggested
that the negative stimulus was indeed perceived negatively and was working as hypothesised.
Table 1
Summary of Study 1 paired samples t-tests.
Attitudes
Pre-test
Post-test
Sig.
SD
SD
Experimental
Sponsor
Event
77
4.05
4.02
1.20
1.19
3.47
3.51
1.20
1.28
Control
Sponsor
Event
59
4.17
4.10
.79
.90
4.11
4.09
.83
.86
.002*
.007*
.415
.927
Note: M = mean score; SD = standard deviation.

* p < .01.
4.2. Study 2 results

Respondent familiarity with the authentic and ctitious brands used was assessed to verify the study accurately used
brands that were high and low in familiarity. As reported in Table 2, familiarity with the ctional event and the ctional
sponsor was low (below the midpoint level of 4.0); whereas familiarity with the real event and sponsor was high (above the
mid-point of 4.0). Such differences between the ctitious and real brands indicated that the experimental design was
appropriately constructed for the purposes of the research. Therefore, the design allowed the researchers to assess the
inuence of familiarity and its covariates on potentially tempering the inuence of negative celebrity endorser publicity on
brand attitudes.
Table 2
Summary of Study 2 familiarity mean scores.
Condition 1
Fictional event/ctional sponsor
Condition 2
Fictional event/real sponsor
Condition 3
Real event/real sponsor
Condition 4
Real event/ctional sponsor
Familiarity
SD
Event
Sponsor
Event
Sponsor
Event
Sponsor
Event
Sponsor
68
68
66
66
69
69
69
69
2.26
2.47
2.76
6.30
4.71
5.40
5.00
2.36
1.54
1.64
1.60
0.81
1.60
1.41
1.50
1.18
Note: M = mean score; SD = standard deviation.
j.smr.2013.10.003
G Model

Bivariate correlation analyses provided support that involvement and t could be treated as covariates of familiarity
(see Table 3). Involvement and t were shown to have a signicantly positive correlation (p < .05) with both attitudes
towards the event (r = .84 and .29), and towards the sponsor brand (r = .46 and .38). As both covariates were also shown to
have a signicant correlation to familiarity to the event (r = .65 and .24) and familiarity to the sponsor brand (r = .39 and .15)
the use of MANCOVA to separate these potential confounds was deemed appropriate in an effort to test for familiarity
effects.
Table 3
Study 2 inter-construct correlations.
Atte
Atte
Attb
Fit
Invol
Fame
Famb
1
.43**
.29**
.84**
.61**
.31**
Attb
Fit
Invol
Fame
Famb
1
.28**
.24**
.15*
1
.65**
.39**
1
.32**
1
.38**
.46**
.25**
.44**
Note: Values below the diagonal are correlation estimates.

* p < .05.
** p < .01.
Results of the between subjects MANCOVA tests supported Hypotheses 3 and 4. In particular, familiarity with the event
displayed a signicant positive effect (p < .05) on tempering the impact of negative celebrity endorser publicity on attitudes
to both the event (F = 5.62; Z2 = .021) and the sponsor brand (F = 11.15; Z2 = .040). Further, familiarity with the brand had a
signicant positive effect (p < .05) on tempering changes in sponsor brand attitudes (F = 3.99; Z2 = .015) but no effect on
attitudes towards the event (p > .05). No interaction effect was found (see Table 4). The MANCOVA results also revealed that
involvement and t had a signicant effect on the impact of the negative stimulus on attitude change.
The greatest inuence on attitudes was found in the involvement construct. Involvement with the event was found to
have a signicant positive effect (p < .01) on both attitudes towards the event (F = 528.49; Z2 = .665) and the sponsor
(F = 55.86; Z2 = .174). Similarly, t was shown to have a signicant effect (p < .05) on attitudes towards the sponsor
(F = 26.18; Z2 = .090) and towards the event (F = 4.05; Z2 = .015). Therefore, Hypotheses 5 and 6 were supported.
Table 4
Summary of Study 2 between subjects MANCOVA.
Independent variables
Event familiarity
Sponsor familiarity
Involvement
Fit
Event familiarity sponsor familiarity
Dependent variables
a
Event attitudes
Sponsor attitudesb
Event attitudes
Sponsor attitudes
Event attitudes
Sponsor attitudes
Event attitudes
Sponsor attitudes
Event attitudes
Sponsor attitudes
F
5.623
11.152
.174
3.990
528.498
55.865
4.059
26.188
1.010
.651
Z2
Sig.
*
.018
.001**
.677
.047*
.000**
.000**
.045*
.000**
.316
.420
.021
.040
.001
.015
.665
.174
.015
.090
.004
.002
Note: F = F-value. Z2 = Partial eta squared.

a
Denotes R-squared = .71 (adjusted R = .70).
b
Denotes R-squared = .32 (adjusted R = .30).
* p < .05.
** p < .01.
5. Discussion
Results from the two studies provide support that celebrity endorser negative publicity has the potential to inuence
consumer attitudes towards associated event and sponsor brands. Furthermore, this inuence is most pronounced when
attitudes towards each have not been previously formed, such as in the case of new or unfamiliar brands. This nding
concurs with previous sponsorship research which has linked the actions of celebrity endorsers to changes in attitudes held
towards associated brands (Hughes & Shank, 2005; Till & Shimp, 1998; White et al., 2009; Wilson et al., 2008) with such
evaluations being tempered by previous commitment to the brand (Ahluwalia et al., 2000; Funk & Pritchard, 2006).
Study 1 assessed the effect of negative celebrity endorser publicity on unfamiliar sponsor and event brands. Results
suggested the negative stimulus provided was indeed perceived negatively. Furthermore, inspection of the mean scores
indicated that negative perceptions were transferrable to, and impacted upon, the event and sponsor brand attitudes for
individuals in the experimental group. No such effects were observed within the control group. This discrepancy between
the groups provides support that negative publicity attributed to celebrity endorsers has the capacity to weaken sponsor
j.smr.2013.10.003
G Model

10
brand attitudes. Extending past research, this research illustrated that associated brands, in this case the event brand, also
may be adversely affected by negative publicity surrounding associated celebrity endorsers.
Study 2 extended Study 1 by including real and therefore familiar brands in the experimental design. Results supported
the hypothesis that any weakening of attitudes towards the associated brands would be tempered by familiarity with the
respective brands. After accounting for the variance explained by the covariates of involvement and t, familiarity was found
to signicantly inuence the degree to which the negative information impacted attitudes towards the sponsor, and to a
lesser extent, the event. This indicated that attitudes towards known brands are less likely to change when negative
information which can be linked with the brand is encountered. Such a nding extends marketing research into the sport
management discipline by revealing attitudes are stronger and less susceptible to change for more familiar brands than
those held towards less familiar brands (e.g., Bettman & Sujan, 1987; Fazio, 1989). Relating this nding to attitude formation
within the sponsorship arena, the less familiar an individual is with a sponsors brand, the more likely they will be to rely on
external links to form an attitude towards that brand. Such a nding aligns with research which posits individuals seek to
minimise the effort required to form a valid attitude (Fazio, 1990; Payne et al., 1993).
5.1. Theoretical implications
This research contributes to sport sponsorship knowledge by extending understandings of how negative celebrity
endorser publicity may impact associated brand attitudes. Overall, results conrm previous sponsorship and attitude
formation literature which posits associated brands can be linked with the actions of celebrity endorsers. Additionally, this
research extends this understanding by showing effects are tempered by brand knowledge in the form of familiarity,
involvement and t. Three key theoretical contributions were made based on this research.
First, the current research answered calls to assess how negative celebrity endorser publicity may impact a range of
associated brand categories (White et al., 2009). Findings revealed that respondent attitudes towards sponsors and event
brands can be adversely impacted when negative publicity surrounding an associated celebrity endorser emerges. This is a
point especially relevant to brands which are not market leaders who may use sponsorship as a way to promote their brands.
New brands and events, such as start-up entities or those new to a particular market, do not possess tangible association sets
(e.g., Ahluwalia et al., 2000; Funk & Pritchard, 2006; Nickell et al., 2011) and thus attitudes towards these brands are more
susceptible to be inuenced by the information which surrounds them (e.g., Eagly & Chaiken, 1995; Sherif, 1963). It is for this
reason such brands may benet more from sponsorships than established competitors. However, for the same reason, such
brands are more susceptible to harmful consequences during instances where negative publicity emerges. In the current
research, negative consequences were apparent for both sponsor and the event brands, providing an improved
understanding of how negative publicity may affect associated brands and sponsorship portfolios (e.g., Chien et al., 2011;
Cousens et al., 2006).
Second, the use of familiarity measures within Study 2 answered calls from researchers who suggested higher familiarity
would limit the degree of negative information transference (Till & Shimp, 1998; White et al., 2009). Results conrmed that
although brands are susceptible to negative information surrounding celebrity endorsers, familiarity and its covariates (e.g.,
involvement and t) determine to what extent this may occur. This nding may be explained by understanding that
individuals with less knowledge about a brand are more likely to be inuenced by any information that may be linked to that
brand (Fazio, 1990; Payne et al., 1993). The prominent inuence of involvement and t within the current research extends
this knowledge and suggests that these constructs are more important variables which determine how attitudes may be
impacted by negative information.
Involvement with the event accounted for 67% of the variance in event attitudes and 17% of the variance in sponsor brand
attitudes, far exceeding the inuence of familiarity. One reason for such a nding may be the inuence of involvement on
information processing and the high correlation between event involvement and respondent attitudes reported in our
studies. Eisend (2013) found highly involved consumers were more inuenced by the degree of negativity included in
stimulus materials than low involvement consumers, who held attitudes which were inuenced by the amount of
information in the message. Other research has found enduring personal factors such as product class involvement are
correlated with information processing tendencies (e.g., Celsi & Olson, 1988; Tsiotsou, 2013). Thus, whilst familiarity with
the event does have a signicant inuence on attitudes, familiarity only describes the degree of brand knowledge held;
whereas involvement describes a deeper psychological connection. From this perspective, in deciphering the messages
linked with brands, involvement is likely the primary driver of information processing, with familiarity providing only a
modest effect. Such a nding is consistent with previous theorising which suggests that involvement is key to message
elaboration and inuences how information is processed (e.g., Petty & Cacioppo, 1986).
The importance of the perceived degree of t between the event and the sponsor brand in tempering negative attitudinal
consequences also provides an interesting nding. Fit accounted for a comparable amount of variance in event attitudes
(1.5%) and more of the variance in sponsor brand attitudes (9%) than did familiarity. Previous research indicates that
attitudes towards sponsor brands are likely to be positive when consumers hold positive event attitudes and perceive that
the event and sponsor are congruent with one another (e.g., Becker-Olsen, 2003; Becker-Olsen & Simmons, 2002; Gwinner &
Bennett, 2008). Findings from the current research suggest that in instances where negative information can be linked with
associated brands, perceptions of t temper any negative consequences more for sponsors than event brands. It is possible
that within the sport sponsorship category utilised in this research, the celebrity endorser was more strongly linked to the
j.smr.2013.10.003
G Model

11
event than the sponsor. Such an explanation may be valid given that the event provides the context in which the celebrity
competes; whereas the sponsor brand is further removed from the celebrity and thus, may be somewhat protected from the
transference of negative information (e.g., Fazio, 1990; Kahneman, 2011; Payne et al., 1993; Simon, 1982).
Third, the majority of previous research (Ahluwalia et al., 2000; Funk & Pritchard, 2006; Till & Shimp, 1998; White et al.,
2009; Zhou & Whitla, 2013) has used student samples, raising issues concerning the generalisability of results. Study 2 was
relevant to the real world as it incorporated both a general public sample and authentic brands into its design. Thus, results
from this study provide a replication and extension to the real world, increasing the validity of ndings and the applicability
to brand and event managers.
5.2. Managerial implications
This research may be used by brand managers to better understand how to direct their brands and offers insights into the
potential consequences linked with negative celebrity endorser publicity. Given that current research ndings indicated
attitudes towards the event and the sponsor were adversely affected, these results are relevant to a broad range of
stakeholders and provide an insight into consumer perceptions of sport sponsorships. Results suggest that negative publicity
linked with celebrity endorsers has the potential to negatively inuence the attitudes held towards associated brands. Given
the widespread media attention dedicated to sport and the instantaneous dissemination of news via social media, it is
particularly important that sport organisations are proactive in protecting their brands from the potential negative
consequences identied by the current research. The proactive and reactive strategies which sport brand managers may
implement based on this research are now discussed.
First, brand managers should carefully consider the risks and rewards of entering into agreements with other entities.
Whereas sponsorship and celebrity endorsements can result in increased brand awareness and the transference of positive
associations, this research also highlights the possible negative consequences which sponsors and events must consider (e.g.,
Till & Shimp, 1998; Till et al., 2008). Findings from this research suggest that it is important for sponsor brands to carefully
select the events and endorsers they are associated with to protect against such undesirable consequences. Similarly, to
safeguard their brands, event managers should work collaboratively with sponsors to develop an internal code of behaviour
that include punishments for possible breaches of that code (Farrelly, 2010). Event managers should also take steps to ensure
that individual athletes or competitors are not placed in situations which are likely to result in negative publicity, given that
such attention can impact negatively on the event itself, and other stakeholders. Particular attention and resources should be
invested by new or less known brands and events in light of this research, given they are most susceptible to the effects
negative information transference.
The current research ndings may inform and augment existing protocols and assist brand managers who have existing
sponsorships to best manage negative celebrity endorser publicity. Whilst it is important that brands should conduct due
diligence and implement proactive strategies to prevent negative celebrity endorser publicity from occurring, unforeseen
events can and will transpire. Thus, event and brand managers need to develop strategic communication protocols when
undesirable events inevitably occur. Although this may seem like a logical suggestion, up to 70% of organisations lack crisis
management plans (Burnett, 1998). Research highlights the importance of crisis management plans which have the
potential to inuence how severe negative effects may be felt following instances of negative publicity. For instance, Fink and
colleagues (2009) found that leadership responses which are perceived as weak (e.g., slowly dealing with the matter
internally) can further harm consumer attitudes. Therefore, should an incident occur, management must have detailed plans
in place to swiftly denounce the behaviour and distance themselves from the offender. Based on the current research
ndings, by not denouncing the act, brand managers risk further damaging consumer attitudes, especially from individuals
who have low brand familiarity.
Perhaps more importantly, this research highlighted how the negative consequences are more prominent for individuals
low in involvement with the event and those who perceived a lack of congruence to exist between the event and the sponsor.
Thus, brand managers may benet from targeted strategies which aim to increase consumer involvement and reinforce the
synergies which exist between the event and the sponsor brand. Hosting pre-event welcome functions and using social
media channels to communicate frequently with consumers may be effective in fostering a sense of involvement with the
event. Additionally, these initiatives could be hosted and promoted by both the event and the sponsor as a means to enhance
the perceived level of t between these stakeholders.
6. Limitations and future research
This research has ve main limitations which should be acknowledged. First, the research was conned to an Australia
motorsport context and used a limited selection of events and brands in the research design. Although the research used real
brands within Study 2, these brands only represented specic product categories and may not be representative of brands
operating in different markets. Therefore, future research spanning different research settings is needed to support or refute
our ndings. Research in different countries, sport event settings, and utilising brands from different product categories will
be benecial in testing the generalisability of results. Another related limitation in this regard stems from the use of a
ctional celebrity endorser, who was preferred over the use of a real motorsport celebrity to remove familiarity biases. The
endorser was described as a champion in the sport to increase his credibility; however it is possible some individuals with
j.smr.2013.10.003
G Model

12
motorsport knowledge could have guessed he was ctional. It is likely that the use of real celebrity endorsers would produce
different results and introduce another level of complexity to research of this kind. Thus, future research should control for
manipulations in familiarity levels of not only sponsors and the event, but also the endorser.
Second, although the negative publicity used in our research reported an illegal act denounced by society, the specic
behaviour and presentation of the press release may have inuenced the results. It is possible that different types of
transgressions would evoke different reactions (e.g., Fink et al., 2009; Louie & Obermiller, 2002). Thus, a wide range of
negative behaviours, ranging from minor (e.g., littering) to major (e.g., murder) transgressions should be investigated in
future research to explore potential effects across a spectrum of negative publicity (e.g., Fiske, 1980; Fournier, 1998). Such
research may provide additional insights into the specic acts deemed serious enough to damage consumer attitudes and
those which may be largely inconsequential in terms of attitude change.
The third limitation stems from the use of the press release, which may have produced different results than other forms
of media. Thus, future research should consider disseminating negative information via other sources (e.g., Ahluwalia et al.,
2000). It may be that negative information conveyed by word of mouth or via social media may be more damaging than the
same information presented by a less credible source, such as advertising (Sternthal, Dholakia, & Leavitt, 1978). Negative
framing in the form of comparative advertising may, under certain conditions, generate specic related cognitions, thereby
diminishing the impact of information (Shiv, Edell, & Payne, 1997) as sponsorship ad manipulations have been found to
impact subjects perceptions of ad schema congruence and their subsequent attitudes towards the sponsorship ads
(McDaniel & Heald, 2000). Thus, in more typical viewing conditions, perceptions formed through involvement, t, and
familiarity may cause greater differences in message processing and subsequently inuence attitudinal responses.
Fourth, to further validate our ndings, research which looks at brands and events relying on rogue images should also be
conducted. Certain events (e.g., Ultimate Fighting Championship, Crusty Demons), athletes (e.g., Dennis Rodman, John
McEnroe), teams (e.g., Oakland Raiders, Detroit Pistons) and potential sponsors (e.g., Rockstar Energy Drink, Harley Davidson
Motorcycles) thrive on extreme images. In these situations, an endorser engaging in what the general public would
traditionally denounce as negative behaviour may in fact align to certain rogue brand ethos and reinforce the desired image.
Thus, future research should aim to assess if such negative publicity weakens or in fact increases consumer attitudes in such
a scenario.
Finally, this research did not take into account the notion of multiple sponsorship effects (e.g., Chien et al., 2011; Cousens
et al., 2006). Consumers develop brand attitudes based on multiple sources of information they encounter, including the
links that brands have with other entities. Sponsor brands typically diversify their interests by developing sponsorship
portfolios that include sponsoring a varied array of events which can be dissimilar to one another. Similarly, events align
themselves with sponsors which sometimes portray vastly different images from one another. From this perspective, a
broader consideration of the interactions between brands should be adopted by future research in this area. Although Study
1 utilised ctional brands which do not have sponsorship portfolios, the results reported in Study 2 may have been
inuenced by the specic brands chosen. It is likely that attitudes towards the brands selected were inuenced by
consumers knowledge and interaction with these brands across their multiple sponsorship portfolios (Chien et al., 2011).
Future research should build upon the current manuscript by designing experiments which can account for multiple
sponsorship scenarios. Research of this nature is likely to both extend understandings of the role of familiarity in tempering
attitude change, and provide insights into the potential consequences for a broader range of brands which could conceivably
be linked with negative celebrity endorser publicity.
Acknowledgements
The authors would like to thank Grifth Universitys Centre for Tourism Sport and Services Research for assistance in
completing this paper. We would also like to thank Professor Chad Perry for his comments on the manuscript.
Appendix A. Survey constructs, items and mean scores Study 1 and Study 2
Construct
Items used
Study 1
Study 2
Event attitudes
Overall, my attitude towards the ____ event is positive

My overall attitude towards the ____ event is unfavourable (reverse coded)
Generally, I have a good attitude towards the ____ event
.76.90
.97.99
Sponsor attitudes
Overall, my attitude towards the ____ brand is positive

My overall attitude towards the ____ brand is unfavourable (reverse coded)
Generally, I have a good attitude towards the ____ brand
.77.91
.99.99
Event familiarity
I am very familiar with the ____ event

I can easily recognise the ____ event
N/A
.86.97
Sponsor familiarity
I am very familiar with the ____ brand

I can easily recognise the ____ brand
N/A
.87.97
Event involvement
Supporting the ____ event is important to me

I do not want to attend the ____ event (reverse coded)
N/A
.91.97
j.smr.2013.10.003
G Model

13
Appendix A (Continued )
Construct
Items used
Study 1
Study 2
N/A
.85.97
I would enjoy following coverage of the ____ event

Attending the ____ event is important to me
Image t
The ____ brand and the ____ event have a similar image
The ideas I associate with the ____ brand are different to those I associate with the ____ event
(reverse coded)
My image of the ____ brand is the same as the image I have of the ____ event
References
Agrawal, J., & Kamakura, W. (1995). The economic worth of celebrity endorsers: An event study analysis. Journal of Marketing, 59(3), 5662.
Agyemang, K. J. (2011). Athlete brand revitalisation after a transgression. Journal of Sponsorship, 4(2), 137146.
Ahluwalia, R., Burnkrant, R. E., & Unnava, H. R. (2000). Consumer response to negative publicity: The moderating role of commitment. Journal of Marketing
Research, 37(2), 203214.
Alba, J., & Hutchinson, J. (1987). Dimensions of consumer expertise. Journal of Consumer Research, 13(4), 411454.
Aronson, E., Brewer, M., & Carlsmith, J. M. (1985). Experimentation in social psychology. In Lindsay, G., & Aronson, E. (Eds.), Handbook of social psychology pp. 441
486). (Vol. 1New York, USA: Random House
BBC (2010). Barcelona & Qatar Foundation agree 125m shirt deal. Retrieved from http://news.bbc.co.uk/sport2/hi/football/europe/9276343.stm.
Becker-Olsen, K. L. (2003). And now, a word from our sponsor: A look at the effects of sponsored content and banner advertising. Journal of Advertising, 32(2), 17
32.
Becker-Olsen, K. L., & Simmons, C. (2002). When do social sponsorships enhance or dilute equity? Fit, message source, and the persistence of effects. Advances in
Consumer Research, 29(1), 287289.
Bennett, G., Ferreira, M., Lee, J., & Polite, F. (2009). The role of involvement in sports and sport spectatorship in sponsors brand use: The case of Mountain Dew and
Action Sports sponsorship. Sport Marketing Quarterly, 18(1), 1424.
Bettman, J., & Sujan, M. (1987). Impacts of framing on evaluation of comparable and non-comparable alternatives by expert and novice consumers. Journal of
Consumer Research, 14(2), 141154.
Burnett, J. (1998). A strategic approach to managing crises. Public Relations Review, 24(4), 475483.
Celsi, R., & Olson, J. (1988). The role of involvement in attention and comprehension processes. Journal of Consumer Research, 15(2), 210224.
Chaiken, S. (1980). Heuristic versus systematic information processing and the use of source versus message cues in persuasion. Journal of Personality and Social
Psychology, 39(5), 752766.
Chaiken, S., Liberman, A., & Eagly, A. (1989). Heuristic and systematic information processing within and beyond the persuasion context. In J. Uleman & J. Bargh
(Eds.), Unintended thought (pp. 212252). New York, USA: Guilford Press.
Chan, K. K., & Misra, S. (1990). Characteristics of the opinion leader: A new dimension. Journal of Advertising, 19(3), 5360.
Chanavat, N., Martinent, G., & Ferrand, A. (2009). Sponsor and sponsees interactions: Effects on consumers perceptions of brand image, brand attachment, and
purchasing intention. Journal of Sport Management, 23(5), 644670.
Chien, P. M., Cornwell, T. B., & Pappu, R. (2011). Sponsorship portfolio as a brand-image creation strategy. Journal of Business Research, 64(2), 142149.
Cook, T., & Campbell, D. (1979). Quasi-experimentation: Design and analysis for eld settings. Boston, USA: Houghton Mifin.
Cornwell, T., & Maignan, I. (1998). An international review of sponsorship research. Journal of Advertising, 27(1), 121.
Cornwell, T., Weeks, C., & Roy, D. (2005). Sponsorship-linked marketing: Opening the black box. Journal of Advertising, 34(2), 2142.
Cousens, L., Babiak, K., & Bradish, C. L. (2006). Beyond sponsorship: Re-framing corporate sport relationships. Sport Management Review, 9(1), 123.
Day, E., Stafford, M., & Camacho, A. (1995). Opportunities for involvement research: A scale-development approach. Journal of Advertising, 24(3), 6975.
Eagly, A. H., & Chaiken, S. (1995). Attitude strength, attitude structure, and resistance to change. In R. Petty & J. Krosnick (Eds.), Attitude strength: Antecedents and
consequences. New Jersey, USA: Lawrence Erlbaum and Associates.
Eisend, M. (2013). The moderating inuence of involvement on two-sided advertising effects. Psychology and Marketing, 30(7), 566575.
Erdogan, B. Z. (1999). Celebrity endorsement: A literature review. Journal of Marketing Management, 15(4), 291314.
Fahy, J., Farrelly, F., & Quester, P. (2004). Competitive advantage through sponsorship: A conceptual model and research propositions. European Journal of
Marketing, 38(8), 10131030.
Farrelly, F. (2010). Not playing the game: Why sport sponsorship relations break down. Journal of Sport Management, 24(3), 319337.
Fazio, R. H. (1989). On the power and functionality of attitude accessibility. In A. Pratkanis, S. Breckler, & A. Greenwald (Eds.), Attitude structure and function. New
Jersey, USA: Lawrence Erlbaum and Associates.
Fazio, R. H. (1990). Multiple processes by which attitudes guide behavior: The MODE model as an integrative framework. In Zanna, M. (Ed.). Advances in
experimental social psychology (Vol. 23). New York, USA: Academic Press.
Ferber, R. (1977). Research by convenience. Journal of Consumer Research, 4(1), 5758.
Ferrand, A., & Pages, M. (1999). Image management in sport organizations: The creation of value. European Journal of Marketing, 33(3/4), 387401.
Field, A. (2005). Discovering statistics using SPSS (2nd ed.). Thousand, Oaks, USA: Sage Publications.
Field, A., & Hole, G. (2003). How to design and report experiments. London, UK: Sage Publications.
Fink, J. S., Parker, H. M., Brett, M., & Higgins, J. (2009). Off-eld behavior of athletes and team identication: Using social identity theory and balance theory to
explain fan reactions. Journal of Sport Management, 23(2), 142155.
Fink, J. S., Parker, H. M., Cunningham, G. B., & Cuneen, J. (2010). Female athlete endorsers: Determinants of effectiveness. Sport Management Review, 15(1), 1322.
Fiske, S. (1980). Attention and weight in person perception: The impact of negative and extreme behavior. Journal of Personality and Social Psychology, 38(6), 889
906.
Fiske, S., & Taylor, S. (1991). Social cognition (2nd ed.). New York, USA: McGraw-Hill.
Fournier, S. (1998). Consumers and their brands: Developing relationship theory in consumer research. Journal of Consumer Research, 24(4), 343373.
Funk, D. C., & Pritchard, M. (2006). Sport publicity: Commitments moderation of message effects. Journal of Business Research, 59(1), 613621.
Gill, J. D., Grossbart, S., & Laczniak, R. N. (1988). Inuence of involvement, commitment and familiarity on brand beliefs and attitudes of viewers exposed to
alternative ad claim strategies. Journal of Advertising, 17(2), 3343.
Gwinner, K., & Bennett, G. (2008). The impact of brand cohesiveness and sport identication on brand t in a sponsorship context. Journal of Sport Management,
22(4), 410426.
Gwinner, K., & Eaton, J. (1999). Building brand image through event sponsorship: The role of image transfer. Journal of Advertising, 28(4), 4757.
Hair, J. F., Black, W. C., Babin, B. J., & Anderson, R. E. (2009). Multivariate data analysis (7th ed.). New Jersey, USA: Prentice Hall.
Havitz, M., & Mannell, R. (2005). Enduring involvement, situational involvement, and ow in leisure and non-leisure activities. Journal of Leisure Research, 37(2),
152177.
Hughes, S., & Shank, M. (2005). Dening scandal in sports: Media and corporate sponsor perspectives. Sport Marketing Quarterly, 14(4), 207216.
Johar, G., & Pham, M. (1999). Relatedness, prominence, and constructive sponsor identication. Journal of Marketing Research, 36(3), 299312.
Kahneman, D. (2011). Thinking fast and slow. New York, USA: Farrar, Strauss, Giroux.
j.smr.2013.10.003
G Model

14
Keller, K. (2003). Strategic brand management: Building, measuring, and managing brand equity (2nd ed.). New Jersey, USA: Prentice Hall.
Kotler, P., Burton, S., Deans, K., Brown, L., & Armstrong, G. (2013). Marketing (9th ed.). Frenchs Forest, NSW: Pearson Australia.
Louie, T. A., & Obermiller, C. (2002). Consumer responses to a rms endorser (dis)association decisions. Journal of Advertising, 31(4), 4152.
McDaniel, S. R., & Heald, G. R. (2000). Young consumers responses to event sponsorship advertising of unhealthy products: Implications of schema triggered affect
theory. Sport Management Review, 3(2), 163184.
Miciak, A. R., & Shanklin, W. L. (1994). Choosing celebrity endorsers. Marketing Management, 3(3), 5159.
Nickell, D., Cornwell, T. B., & Johnson, W. J. (2011). Sponsorship-linked marketing: A set of research propositions. Journal of Business and Industrial Marketing, 26(8),
577589.
Nunnally, J., & Bernstein, I. (1994). Psychometric theory (3rd ed.). New York, USA: McGraw-Hill.
Payne, J., Bettman, J., & Johnson, E. (1993). The adaptive decision maker. New York, USA: Cambridge University Press.
Pedhazur, E. J., & Schmelkin, L. P. (1991). Measurement, design, and analysis: An integrated approach. New York, USA: Lawrence Erlbaum and Associates.
Pentecost, R. D., & Andrews, L. (2010). Fashion retailing and the bottom line: The effect of generational cohorts, gender, fashion consciousness, attitudes and
impulsiveness on fashion expenditure. Journal of Retailing and Consumer Services, 17(1), 4352.
Pentecost, R. D., & Spence, M. T. (2009). Spill-over effects in multi-sponsored events: Theoretical insights and empirical evidence. World Journal of Management,
1(1), 95117.
Petty, R., & Cacioppo, J. (1986). The elaboration likelihood model of persuasion. In L. Berkowitz (Ed.), Advances in experimental social psychology (pp. 123205). New
York, USA: Academic Press.
Raju, P. S., & Reilly, M. D. (1979). Product familiarity and information processing strategies: An exploratory investigation. Journal of Business Research, 8(2), 187
212.
Ruth, J., & Simonin, B. (2003). Brought to you by brand A and brand B: Investigating multiple sponsors inuence in consumers attitudes toward sponsored events.
Journal of Advertising, 32(3), 1930.
Samu, S., Krishnan, H., & Smith, R. (1999). Using advertising alliances for new product introduction: Interactions between complimentary and promotional
strategies. Journal of Marketing, 63(1), 5774.
Sherif, C. W. (1963). Social categorization as a function of latitude of acceptance and series range. The Journal of Abnormal and Social Psychology, 67(2), 148156.
Shiv, B., Edell, J. A., & Payne, J. W. (1997). Factors affecting the impact of negatively and positively framed ad messages. Journal of Consumer Research, 24(3), 285
294.
Silvera, D. H., & Austad, B. (2004). Factors predicting the effectiveness of celebrity endorsement advertisements. European Journal of Marketing, 38(11/12), 1509
1526.
Simon, H. (1982). Models of bounded rationality. Massachusetts, USA: MIT Press.
Simonin, B., & Ruth, J. (1998). Is a company known by the company it keeps? Assessing the spillover impacts of brand alliances on consumer brand attitudes.
Journal of Marketing Research, 35(1), 3042.
Smith, G. (2004). Brand image transference through sponsorship: A consumer learning perspective. Journal of Marketing Management, 20(3/4), 457474.
Speed, R., & Thompson, P. (2000). Determinants of sports sponsorship response. Journal of the Academy of Marketing Science, 28(2), 226238.
Sternthal, B., Dholakia, R., & Leavitt, C. (1978). The persuasive effect of source credibility: Tests of cognitive response. Journal of Consumer Research, 4(4), 252260.
Tabachnick, B., & Fidell, L. (2012). Using multivariate statistics (6th ed.). New York, USA: Pearson.
The Canadian Press (2011). Labatt loses legal faceoff with Molson Coors in NHL sponsorship spat. Retrieved from: http://www.nhl.com/ice/news.htm?id=594809.
Till, B. D., & Shimp, T. (1998). Endorsers in advertising: The case of negative celebrity information. Journal of Advertising, 27(1), 6782.
Till, B. D., Stanley, S. M., & Priluck, R. (2008). Classical conditioning and celebrity endorsers: An examination of belongingness and resistance to extinction.
Psychology & Marketing, 25(2), 179196.
Tsiotsou, R. (2013). Investigating the role of enduring and situational involvement with the program context on advertising effectiveness. Journal of Marketing
Communications, 19(2), 114135.
White, D., Goddard, L., & Wilbur, N. (2009). The effects of negative information transference in the celebrity endorsement relationship. International Journal of
Retail and Distribution Management, 37(4), 322335.
Wilson, B., Stavros, C., & Westburg, K. (2008). An examination of the impact of player transgressions on sponsorship B2B relationships. Public Relations Review,
34(2), 99107.
Zaichkowsky, J. (1985). Measuring the involvement construct. Journal of Consumer Research, 12(3), 341352.
Zhou, L., & Whitla, P. (2013). How negative celebrity publicity inuences consumer attitudes: The mediating role of moral reputation. Journal of Business Research,
66(8), 10131020.
Zimbardo, P., Ebbesen, E., & Maslach, C. (1977). Inuencing attitudes and changing behavior. Massachusetts, USA: Addison-Wesley.
j.smr.2013.10.003

1 s2.0 S1441352313000715 Main

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

1 s2.0 S1441352313000715 Main

Uploaded by

Copyright:

Available Formats

G Model

SMR-243; No. of Pages 14

Contents lists available at ScienceDirect

Sport Management Review

The effect of familiarity on associated sponsor and event

Department of Tourism, Sport & Hotel Management, Grifth University, Australia

SMR-243; No. of Pages 14

J.P. Doyle et al. / Sport Management Review xxx (2013) xxxxxx

SMR-243; No. of Pages 14

SMR-243; No. of Pages 14

J.P. Doyle et al. / Sport Management Review xxx (2013) xxxxxx

SMR-243; No. of Pages 14

SMR-243; No. of Pages 14

J.P. Doyle et al. / Sport Management Review xxx (2013) xxxxxx

Fig. 1. Conceptual overview and summary of hypotheses.

SMR-243; No. of Pages 14

3.3. Study 1 analysis

SMR-243; No. of Pages 14

Note: M = mean score; SD = standard deviation.

4.2. Study 2 results

Note: M = mean score; SD = standard deviation.

SMR-243; No. of Pages 14

Note: Values below the diagonal are correlation estimates.

Note: F = F-value. Z2 = Partial eta squared.

SMR-243; No. of Pages 14

J.P. Doyle et al. / Sport Management Review xxx (2013) xxxxxx

SMR-243; No. of Pages 14

SMR-243; No. of Pages 14

Overall, my attitude towards the ____ event is positive

Overall, my attitude towards the ____ brand is positive

I am very familiar with the ____ event

I am very familiar with the ____ brand

Supporting the ____ event is important to me

SMR-243; No. of Pages 14

I would enjoy following coverage of the ____ event

SMR-243; No. of Pages 14

J.P. Doyle et al. / Sport Management Review xxx (2013) xxxxxx

You might also like