Australasian Marketing Journal 19 (2011) 230–237

Contents lists available at ScienceDirect

Australasian Marketing Journal

journal homepage: www.elsevier.com/locate/amj

Effects of multiple endorsements and consumer–celebrity attachment on attitude

and purchase intention
Jasmina Ilicic ⇑, Cynthia M. Webster
Faculty of Business and Economics, Department of Marketing and Management, Macquarie University, NSW 2109, Australia

a r t i c l e i n f o a b s t r a c t

Article history: This study examines consumer–celebrity attachment within multiple endorsement situations. A 2  2
Received 1 March 2011 factorial design manipulates attachment strength to the celebrity and endorsement situation while con-
Revised 6 July 2011 trolling for celebrity attractiveness, familiarity and match-up. Results show strong attachment to a celeb-
Accepted 6 July 2011
rity encourages positive attitude towards the advertisement and brand regardless of the number of
Available online 2 August 2011
endorsements. A significant interaction effect is found for purchase intention, suggesting that both
attachment and the number of endorsements influence likelihood of purchase. When consumers are
Keywords:
more attached to a celebrity and view that celebrity endorsing multiple brands, this negatively impacts
Celebrity endorsement
Celebrity attachment
their purchase intention. Yet, when consumers have a weak attachment, their purchase intention is
Attitude increased with multiple endorsements.
Purchase intention Ó 2011 Australian and New Zealand Marketing Academy. Published by Elsevier Ltd. All rights reserved.

1. Introduction
Celebrities are a common feature in the contemporary market-
place, often becoming the face, or image, not only of consumer
products and brands, but of organisations themselves. In Australia,
for example, Hugh Jackman is the face of Lipton Ice Tea, Nicole Kid-
man features in ads for Nintendo DS and George Clooney teams up
with John Malkovich for Nespresso coffee. Brands make use of
well-known and likable celebrities by leveraging the celebrities’
equity (Keller, 2008). By pairing a brand with a celebrity, a brand
is able to leverage unique and positive secondary brand associa-
tions from a celebrity and gain consumer awareness, transfer posi-
tive associations tied to the celebrity onto the brand, build brand
image and ultimately enhance the endorsed brand’s equity (Keller,
2008). Celebrity endorsements lead to higher advertisement recall
rates (O’Mahony and Meenaghan, 1997/1998) and improve com-
munication with potential customers by establishing connections
between consumers and the advertised brand (Atkin and Block,
1983; Sherman, 1985). Additionally, celebrities aid in breaking
down cultural barriers (Mooij, 2004), assist in repositioning brand
and company images (Till and Shimp, 1998), and ultimately, im-
prove sales of the endorsed products (Kaikati, 1987). An announce-
ment of a celebrity endorser contract is found to positively affect
stock returns, making celebrities a worthwhile investment (Agra-
wal and Kamakura, 1995; Mathur et al., 1997).
⇑ Corresponding author. Tel.: +61 2 9850 4857; fax: +61 2 9850 6065.
E-mail addresses: jasmina.ilicic@mq.edu.au (J. Ilicic), cynthia.webster@mq.
edu.au (C.M. Webster).
Celebrities not only sell products and brands, they in fact exert
powerful influence across all facets of popular culture and public
life, influencing consumer attitudes and perceptions (Time Inc.,
2006). As such, celebrities increasingly are sought as endorsers
for several brands within different product categories. For exam-
ple, former Miss Universe, Jennifer Hawkins, endorses a depart-
ment store (Myer), lingerie (Lovable), a soft drink (Pepsi), soap
(Lux) and make-up (CoverGirl). While many celebrities endorse
multiple brands, overexposure is potentially detrimental (Till,
1998). Advertising professionals (Foote and Belding, 1978) suggest
selecting an appropriate celebrity endorser with care, so as not to
choose a celebrity tarnished by their associations with other prod-
ucts (Mowen and Brown, 1981). Research shows that consumers
perceive celebrities who endorse brands in multiple product cate-
gories as less credible than those who only endorse one (Mowen
and Brown, 1981; Tripp et al., 1994).
2. Multiple endorsements and consumer attachment
Much of the endorsement research focuses on what character-
istics constitute an effective celebrity endorser, namely in terms
of their personal attractiveness (McGuire, 1985), familiarity
(Kamins, 1990; Misra and Beatty, 1990), likeability (Erdogen,
1999), credibility, believability and expertise (Ohanian, 1990).
Other research examines whether the celebrity endorser’s image
is consistent with the image of the product, brand or organisation
(Kamins, 1990; Kamins and Gupta, 1994; Lynch and Schuler, 1994;
Till and Busler, 1998). Another stream of research assesses the im-
pact of celebrity endorsements for marketing purposes, including

1441-3582/$ - see front matter Ó 2011 Australian and New Zealand Marketing Academy. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.ausmj.2011.07.005
 J. Ilicic, C.M. Webster / Australasian Marketing Journal 19 (2011) 230–237
consumer attitude towards the brand and/or the advertisement
and purchase intention (Misra and Beatty, 1990; Lynch and Schu-
ler, 1994; Tripp et al., 1994; Till and Shimp, 1998). Recent research
explores the deeper connections between consumers and celebri-
ties. Escalas and Bettman (2009) use McCracken’s (1989) meaning
transfer model, whereby the symbolic meanings associated with a
celebrity are transferred onto the brands the celebrity endorses,
and in turn, these meanings are then transferred from the celebrity
to consumers through selecting brands that communicate their
self-concept, ultimately forming a self-brand connection. Thomson
(2006) investigates consumer–celebrity (or human brand) connec-
tions through the varying levels of attachment consumers feel for
celebrities. Although, Thomson’s (2006) study identifies the exis-
tence of consumer–celebrity connections, research is yet to inves-
tigate whether these celebrity connections influence consumer
attitude and purchase intention in an endorsement context.
Although the business case for celebrity endorsement may ex-
plain why multiple endorsements by a celebrity are common fea-
tures in today’s marketing world, too little research is devoted to
examining the effectiveness of multiple brand endorsements (Mo-
wen and Brown, 1981; Tripp et al., 1994). In addition, the impact of
consumer attachment to celebrities on consumer brand evalua-
tions is not known. As research only identifies the existence of con-
sumer–celebrity attachment (Thomson, 2006), it is not known
whether consumer attachment to a celebrity affects their attitude
towards brands endorsed by that celebrity. While the implicit sug-
gestion of the meaning transfer process (McCracken, 1989) is that a
positive relationship exists between consumer attachment to a
celebrity and consumer attitude towards brands endorsed by that
celebrity, these relationships have not been empirically tested. The
influence of attachment within the brand endorsement context is
still unknown.
The purpose here is to contribute to the body of marketing re-
search on celebrity endorsements in three areas. Firstly, the multi-
ple endorsement situation is further examined to determine the
effectiveness of multiple endorsements by a celebrity. Secondly,
the present study extends the research on consumer–celebrity
relationships by investigating the degree to which attachment to
a celebrity impacts consumers’ evaluations of endorsed brands,
in terms of their advertisement and brand attitude, and purchase
intention. Thirdly, the present study examines the extent to which
consumer–celebrity attachment influences consumer attitude to-
wards the advertisements, their attitude towards the brands, and
their purchase intention for multiple brands endorsed by a single
celebrity.
2.1. Celebrity characteristics: attractiveness and congruency
Empirical studies in the area of celebrity endorsement focus on
how specific celebrity characteristics influence consumers’ evalua-
tions and recall of advertisements as well as the featured products
and brands (McGuire, 1985; Misra and Beatty, 1990; Ohanian,
1991; Lynch and Schuler, 1994). The Source Attractiveness Model
posits that the effectiveness of a message is dependent on the
similarity, familiarity, liking and attractiveness of an endorser
(McGui re, 1985). Experimental research demonstrates that attrac-
tive communicators are able to generate greater likeability, intent
to purchase and higher recall rates for the products and brands
endorsed (Kahle and Homer, 1985). Although the Source Attrac-
tiveness Model is valuable in researching celebrity endorsement
effectiveness, it has received criticism (McCracken, 1989). The
model proposes that if the attractiveness condition is satisfied,
‘‘any celebrity should serve as a persuasive source for any advertis-
ing message’’ (McCracken, 1989, p. 311). The model identifies the
importance of attractiveness but fails to explain the extent to
231
which a celebrity must be seen to be appropriate for the brand
endorsed.
The Match-Up Hypothesis (Kamins, 1990; Misra and Beatty,
1990; Kamins and Gupta, 1994; Lynch and Schuler, 1994; Till
and Busler, 1998) introduces the need for an appropriate fit be-
tween the celebrity and the brand being endorsed. The match-up
hypothesis occurs when ‘‘highly relevant characteristics of the
spokesperson are consistent with highly relevant attributes of
the brand’’ (Misra and Beatty, 1990, p. 160). In other words, the
spokesperson must be perceived as an appropriate match with
the brand and product category in which the brand is located.
For example, a muscular spokesperson is seen as congruent when
endorsing a stamina-associated product such as Nautilus weight-
lifting equipment (Lynch and Schuler, 1994). Results from experi-
mental work show that a spokesperson is perceived as more
knowledgeable, credible and persuasive when s/he matches the
product and brand. Congruency results in higher brand recall
(Misra and Beatty, 1990) and positive attitude towards the brand
endorsed, while incongruence leads to negative brand evaluations
(Kamins and Gupta, 1994).
Kamins’ (1990) work investigates the match-up between the
attractiveness of a celebrity and a product that is related to
enhancing personal appearance. The results reveal that for a prod-
uct that is attractiveness-related, such as perfume and cosmetics,
an attractive celebrity outperforms an unattractive celebrity on
likeability and familiarity. An attractive model also significantly
enhances spokesperson credibility and attitude toward the ad
when endorsing an attractiveness-enhancing product. Conversely,
when an attractive model endorses an attractiveness-unrelated
product, such as a computer, no significant effects are found. Till
and Busler (1998) further investigate the match-up hypothesis
and the importance of attractiveness contributing to the effective-
ness of fit between endorser and product. Results of their study
show that the combination of attractiveness and expertise results
in higher attitude toward the brand and purchase intention. Over-
all, the research on celebrity endorsement shows that it is impor-
tant to control specifically for product/brand and celebrity
associations, celebrity-product/brand congruency, as well as famil-
iarity and attractiveness of the celebrity, as these factors signifi-
cantly impact consumer evaluations of the brands endorsed by
celebrities.
2.2. Endorsement situation: single versus multiple
Limited research examines the impact of celebrities endorsing
multiple products. Mowen and Brown (1981) utilise Attribution
Theory (Kelley, 1967) in their study of multiple product endorse-
ments. According to Attribution Theory, consumers view advertise-
ments featuring celebrity endorsers and then seek ‘‘to determine
the causal reasons for a celebrity endorsing a product’’ (Mowen
and Brown 1981, p. 437). This suggests that consumers perceive
celebrity endorsers as believing in the brands they endorse, rather
than endorsing brands purely for monetary reasons. They find that
subjects react more favourably to a branded product and evaluate a
celebrity more positively when the celebrity endorses only a single
product compared to when the celebrity endorses multiple prod-
ucts. Findings also show that advertisements are perceived more
favourably and are more interesting when a celebrity endorses
only one product.
Tripp et al. (1994) use the Source Credibility Model to investi-
gate the impact of multiple product endorsements, whereby the
effectiveness of a message is dependent on the perceived level of
expertise (knowledge, experience and skills) and trustworthiness
(honesty, integrity and believability) of an endorser (Hovland and
Weiss, 1951; Hovland et al., 1953). An endorser who is perceived
to be an expert is found to be more persuasive (Aaker and Myers,
232 J. Ilicic, C.M. Webster / Australasian Marketing Journal 19 (2011) 230–237
1987) and to generate greater intentions to buy the brand (Oha-
nian, 1991). Tripp et al.’s (1994) results suggest that a celebrity is
seen as more trustworthy and more of an expert when only
endorsing one brand. Trustworthiness, expertise and liking of a
celebrity significantly lessen when the celebrity endorses multiple
products.
Taking into consideration past research on multiple product
endorsement, celebrity-product/brand congruency, celebrity-prod-
uct/brand associations and controlling for celebrity attractiveness
and familiarity, the following hypothesis is put forward:
H1. Consumers’ evaluations, in terms of (a) attitude towards the
advertisement, (b) attitude towards the brand and (c) purchase
intention are greater when celebrities endorse a single branded
product than when they endorse multiple brands.
2.3. Consumer–brand connections
Building a strong brand, and therefore consumer based brand
equity, partly relies on creating Consumer–brand connections.
Marketers now stress the importance of creating brand resonance
or brand relationships (Keller, 2008), due to their enduring nature.
Consumers’ attachment to brands is at the ‘‘core of all strong con-
sumer–brand relationships’’ (Fournier, 1998, p. 363), is a crucial
component of brand resonance (Keller, 2008) and, therefore a crit-
ical driver of brand equity (Park et al., 2007; Keller, 2008).
2.3.1. Consumer–celebrity brand attachment
Researchers argue that, through the marketing communication
process, brands take part in active relationship roles, qualifying
them as reciprocating relationship partners (Fournier, 1998;
Thomson et al., 2005). Thomson et al. (2005) investigate con-
sumer–brand relationships under attachment theory. Recognising
the basic human need to make strong emotional attachments to
particular others, attachment is defined as an emotion-laden tar-
get-specific bond between a person and a specific object, such as
another ordinary person, celebrity or brand (Bowlby, 1979,
1980). Attachment theory rests on the notion of ‘‘separation dis-
tress’’, which is the degree to which individuals express strong
emotional reactions when threatened with real or imagined sepa-
ration from an object of affection. This separation distress is a use-
ful indicator of the strength or intensity of an attachment bond
(Berman and Sperling, 1994). Thomson et al.’s (2005) study points
up the strong relationship between brand attachment (feelings of
connection, affection and passion) and psychological separation
distress.
Thomson (2006) further applies Self-Determination Theory (La
Guardia et al., 2000; Ryan and Deci, 2000) to investigate the exis-
tence of consumer–celebrity, or human brand, relationships.
According to Ryan and Deci (2000, p. 262), self-determination is
based on the basic psychological needs of Autonomy (freedom to
make choices), Relatedness (closeness to others) and Competence
(achievement) and these A–R–C needs are ‘‘crucial in predicting
feelings of attachment’’. Thomson (2006) examines the extent to
which individuals feel a celebrity satisfies their A–R–C needs.
Thomson’s (2006) study shows that participants have a stronger
attachment towards celebrities who make them feel they are cared
for (relatedness) but also free to do as they wish (autonomy). The
need for achievements to be recognised (competence) is not a fac-
tor in participants’ attachment to celebrities. As such, Thomson’s
(2006) results confirm the importance of autonomy and related-
ness components of self-determination, but not competency, for
the development of a consumer’s attachment for a celebrity.
Thomson (2006) also examines more positive outcomes of attach-
ment, such as satisfaction, commitment and trust. Once more,
results show attachment strength results in higher satisfaction,
trust and commitment with the celebrity brand relationship.
Although, Thomson (2006) identifies the existence of consumer–
celebrity attachment bonds, the importance of consumer–celebrity
relationships for marketing purposes in an endorsement situation
is unknown.
Research is yet to integrate the brand attachment literature
with the celebrity endorsement literature. The examination of
the attachment literature leads us to hypothesise that a relation-
ship exists between consumer–celebrity endorser attachment
and consumers’ evaluations of single and multiple endorsed
brands. Based on multiple product endorsement and consumer–
celebrity relationship research, the following hypotheses are put
forward:
H2. Consumers’ evaluations, in terms of (a) attitude towards the
advertisement, (b) attitude towards the brand and (c) purchase
intention, are greater in a strong celebrity endorser attachment
condition than in a weak celebrity endorser attachment condition.
H3. Consumer–celebrity endorser attachment has a greater influ-
ence than does celebrity endorsement on (a) attitude towards
the advertisement, (b) attitude towards the brand and (c) purchase
intention.
H4. Consumers’ evaluations, in terms of (a) attitude towards the
advertisement, (b) attitude towards the brand and (c) purchase
intention, are similarly more positive in a strong celebrity endorser
attachment condition than in a weak celebrity endorser attach-
ment condition, regardless of whether the celebrity is seen endors-
ing single or multiple brands.
Consistent with the literature, more positive evaluations will
occur when celebrities endorse a single brand as opposed to
endorsing multiple brands, regardless of the attachment strength.
Consumer–celebrity attachment strength, on the other hand, will
have a positive impact on consumer evaluations, regardless of
the number of brands endorsed by celebrities. The current research
extends the research on celebrity endorsement by examining the
extent to which attachment and endorsement situation influence
consumer attitude towards advertisements, attitude towards
brands and purchase intention.
3. Method
To test the hypotheses and ensure that celebrities and endorsed
brands are critically evaluated without any pre-existing brand asso-
ciations, a 2  2 between subjects factorial design is employed to
manipulate attachment strength to the celebrity (strong versus
weak) and endorsement situation (single versus multiple), while
controlling for the extraneous variables, through pretesting of celeb-
rity attractiveness, celebrity-product match, and pre-existing brand
associations. Attachment strength is manipulated by using two
celebrities, one celebrity for the strong attachment condition and
one celebrity for the weak attachment condition. Hypothetical
brands in generic product categories are used for the single and mul-
tiple endorsement conditions (see Fig. 1). Two pre-tests are ar-
ranged, the first of which determines the celebrities to be used
within the main experiment and the second assists in the selection
of the products and brand names to be used in the main experiment.
3.1. Pre-test 1: celebrity selection
Pre-test 1 is undertaken to select two celebrities who are not
known for endorsing any brands, are perceived to be equal on
attractiveness and familiarity, and yet are seen to differ on
 J. Ilicic, C.M. Webster / Australasian Marketing Journal 19 (2011) 230–237
Fig. 1. 2  2 Factorial research design.
attachment. While sports, music, and movie celebrities are appro-
priate, TV celebrities are chosen, as ample opportunity exists for
frequent exposure, increasing the likelihood of varying levels of
attachment. The Hottest on TV website (2007) is used to identify
attractive and familiar Australian TV personalities. The website al-
lows the public to vote for whom they believe is the hottest Austra-
lian celebrity on television for the month of July. In order to be an
eligible candidate for the website, the television personalities have
to be a regular presenter, panelist, actor or reporter on a television
show produced in Australia. Celebrities are evaluated on various
characteristics including looks, charisma, personality, hype, fresh-
ness and popularity. In total, 16 celebrities meet the criteria and
eight males and eight females are selected for pre testing. An
extensive web search shows that none of the 16 celebrities have
previously endorsed any products or brands.
A convenience sample of 25 students from an undergraduate
marketing course is asked to rate their familiarity with the 16
celebrities and evaluate the attractiveness of each. Familiarity is
assessed with a single item 7-point likert scale. Five items of
attractiveness are measured on a 7-point likert scale based on Oha-
nian (1990) including attractive, handsome or beautiful, classy,
elegant, and sexy. The 25 informants then evaluate their attach-
ment to each of the 16 celebrities. A series of rank order questions
are used to measure six different items and outcomes of attach-
ment based on Thomson (2006) including relatedness, autonomy,
separation distress, satisfaction, trust and commitment.
Following McGuire (1985), a combination of the familiarity and
attractiveness guides the choice of two male celebrities: Rove
McManus and Eddie McGuire. Both Rove and Eddie receive high
mean scores for familiarity (5.6 and 5.5, respectively), are per-
ceived to be similarly attractive (4.64 and 4.89, respectively) and
neither is thought to have endorsed brands previously. Rove re-
ceives very high mean rank scores on all levels of attachment,
whereas Eddie is ranked lower by respondents on all levels of
attachment (4.99 and 2.55, respectively). Accordingly, Rove is used
in the Strong Attachment condition and Eddie is used in the Weak
Attachment condition.
3.2. Pre-test 2: product and brand name selection
Pre-test 2 is undertaken to ensure that product categories cho-
sen are seen as attractiveness-unrelated and seen as neither con-
gruent nor incongruent with the image of the two chosen
celebrities (Rove and Eddie). Previous studies utilise products such
as a pen (Mowen and Brown, 1981; Till and Busler, 1998), home
computer (Kamins, 1990; Kamins and Gupta, 1994; Biswas et al.,
2006), coffee (Misra and Beatty, 1990), and camera (Tripp et al.,
1994). Translating the product name into another language creates
fictitious brand names for the products. If a subject understands
the word, no attributes are associated as the translation is purely
233
the name of the product. The brand names for the test products
are: Gafae Coffee, Marc Pen, Ordinateur Home Computer and
Photographica Camera.
A convenience sample of 19 students in an undergraduate mar-
keting course completes the survey. Informants are provided with
a list of 10 products (six attractiveness-related and four attractive-
ness-unrelated) and asked to indicate the degree to which each of
the products is attractiveness-related (1 = attractiveness-related
and 7 = attractiveness-unrelated) and to evaluate whether the
products and brands are congruent with or match the image of
the celebrity (1 = match and 7 = mismatch). Only those products
that are seen as attractiveness-unrelated and neither congruent
nor incongruent with the celebrity are chosen. Although the cam-
era (attractiveness mean = 5.21, Rove congruence mean = 3.53, Ed-
die congruence mean = 3.79), pen (attractiveness mean = 6.16,
Rove congruence mean = 4.47, Eddie congruence mean = 3.73), cof-
fee (attractiveness mean = 5.37, Rove congruence mean = 4.73, Ed-
die congruence mean = 4.57) and home computer (attractiveness
mean = 5.37, Rove congruence mean = 5.84, Eddie congruence
mean = 5.05) are all perceived as attractiveness-unrelated prod-
ucts, the home computer is seen as incongruent with the images
of both celebrities. As such, Photographica Camera, Marc Pen and
Gafae Coffee product categories and brands are selected for use
in the main experiment, with Photographica Camera chosen as
the test brand.
3.3. Main experiment
In total, 237 undergraduate students participate in the experi-
ment, all of whom are enrolled in marketing courses at a regional
university in Australia. Undergraduate students are specifically
chosen as suitable subjects for this particular study as research
indicates that young adults are more susceptible to celebrity en-
dorser influence (Kaikati, 1987) and consumers’ aged between 16
and 34 are twice as likely to purchase products from a celebrity
they like (The Future Data Foundation, 2004 cited in Pringle and
Binet, 2005). In addition, Thomson’s (2006) study investigates the
existence of consumer–celebrity attachment bonds specifically in
undergraduate students. This study uses the same sampling frame
in order to replicate and extend Thomson’s (2006) research to the
endorsement context. Valid responses are obtained from 198 sub-
jects with roughly equal numbers in each condition. A total of
50.5% of subjects view Eddie McGuire the Weak Attachment condi-
tion and 49.5% view the Strong Attachment condition of Rove
McManus. Females comprise 48.5% of the subjects, with 51.5%
male. Females and males report similar mean attachment levels
for Rove McManus (females mean = 3.09, SD = 1.13, n = 48, males
mean = 3.04, SD = 1.14, n = 50). Males, however, are slightly more
attached to Eddie McGuire than females (males mean = 2.53,
SD = 0.84, n = 52; females mean = 2.10, SD = 0.77, n = 48).
234 J. Ilicic, C.M. Webster / Australasian Marketing Journal 19 (2011) 230–237
3.3.1. Materials: stimuli and measures
Ad booklets containing 12 ads each are constructed for the sin-
gle and multiple endorsement conditions. In the single endorse-
ment condition, the ad booklets contain one test ad featuring
either Rove McManus (strong) or Eddie McGuire (weak) endors-
ing Photographica Camera. Another five Australian TV personali-
ties, who in the pre testing process rank highly on familiarity,
attractiveness and neutral on attachment, feature in five filler
ads endorsing other fictitious brands. Six additional filler ads fea-
ture attractiveness-unrelated products with fictitious brand
names, but no celebrity endorser. In the multiple endorsement
condition, subjects view either Rove (Strong) or Eddie (Weak)
endorsing one fictitious test brand (Photographica) and two func-
tional brands (Marc and Gafae). There are nine filler ads with only
three of the five Australian TV celebrities from the single ad book-
let featuring with multiple attractiveness-unrelated products (e.g.
rice, milk and water) and fictitious brands. All test ads, functional
ads and filler ads within the booklets follow a similar layout. The
celebrity is featured only from the shoulders up so as not to en-
hance attractiveness (as attractiveness is not a component inves-
tigated within this particular study). The two celebrities within
the test ads and functional ads are both wearing black suits and
are featured from mid chest upwards, with a grey studio
background.
Measures used to quantify attachment are taken directly from
Thomson (2006), who adapts four items from La Guardia et al.
(2000) based on self-determination and four items from Hazan
and Shaver (1994) based on attachment. Two items for Autonomy
and two for Relatedness are adapted from La Guardia et al. (2000)
and measured on 7-point likert scales ended 1 = not at all, 7 = very
much. Separation distress is measured with four items taken from
Hazan and Shaver (1994). Satisfaction (two items), Trust (three
items) and Commitment (two items) are sourced from Fletcher
et al.’s (2000) Perceived Relationship Quality Scale and measured
on 7-point likert scales ending 1 = not at all, 7 = very much. Atti-
tude towards the Ad and attitude towards the brand are measured
using four items Mitchell and Olson’s (1981) 5-point semantic dif-
ferential scales. Purchase intention is measured using one item
from Kamins and Gupta (1994) and two items from Choi (2002).
3.3.2. Procedure
With course coordinator approval, the researcher greets stu-
dents within classes and distributes project information state-
ments. Students who agree to take part in the study sign consent
forms and are randomly assigned a questionnaire plus ad booklet.
Subjects are given a brief time to look over the ad booklet. First,
subjects provide information on their TV viewing and on their
familiarity with the celebrity endorser (either Rove or Eddie). They
are then asked to view a photograph of either Rove or Eddie and
evaluate their attachment to the celebrity. Subjects then are direc-
Table 1
Means for attachment, ad attitude, brand attitude and purchase intention.
Variable Mean weak attachment Mean strong attachment
Attachment 2.1502 3.0022
Ad attitude 2.4948 2.9867
Brand attitude 2.9651 3.1413
Purchase intention 3.0067 3.4082
Table 2
Means for attachment, ad attitude, brand attitude and purchase intention by attachment strength and endorsement situation.
Dependent variable Strong attach/single end mean Strong attach/multiple end mean
Ad attitude 2.9583 3.0163
Brand attitude 3.1719 3.1080
Purchase intention 3.5933 3.2153
ted to view a specific ad and to presume that the celebrity had
signed a contract with one fictitious brand (Photographica Camera)
or all three fictitious brands (Gafae Coffee, Marc Pen and Photo-
graphica Camera). In the multiple brand endorsement conditions
subjects are asked to view three ads (functional ads for Marc and
Gafae and test ad for Photographica). Finally, subjects evaluate
their attitude towards the ad, attitude towards the brand, and their
purchase intention of the test brand Photographica Camera.
4. Results
An independent samples t-test manipulation check finds a sig-
nificant difference (t = 6.812, p < 0.001) in attachment, regardless
of endorsement condition, between Rove and Eddie, with subjects
reporting a stronger attachment to Rove (mean = 3.00, SD = 0.61)
and a weaker attachment to Eddie (mean = 2.15, SD = 0.96). Table 1
presents means for attachment and outcome evaluations for the
test brand Photographica. On average, subjects report positive lev-
els of attachment with significantly higher levels of attachment for
Rove (strong attachment). Subjects report a neutral attitude to-
wards the brand and neutral attitude towards the ad in which
the test brand Photographica featured, with a slightly more posi-
tive attitude in the strong attachment condition. Purchase inten-
tion is low, with greater intent to purchase in the strong
attachment condition.
Table 2 presents means for the outcome evaluations of the four
experimental conditions on the test brand Photographica. Subjects
report stronger attitude towards the brand in the single endorse-
ment situations in both the strong and weak attachment condi-
tions, providing support for Hypothesis 1b. Attitude towards the
ad and purchase intention are inconsistent with the current expla-
nations of the effect of multiple endorsements on consumer evalu-
ations. Attitude towards the ad is stronger when the celebrity is
seen endorsing multiple brands, rejecting Hypothesis 1a. Purchase
intention is more likely in the single endorsement situation only
for the strong attachment celebrity, showing partial support for
Hypothesis 1c. Table 3 also shows subjects report stronger ad
and brand attitude and greater likelihood of purchase in both of
the strong attachment conditions, compared to those in the weak
attachment conditions, providing support for Hypothesis 2a, 2b
and 2c.
Table 3 shows the results of the main effects of all the depen-
dent variables, attitude towards the ad, attitude towards the brand
and purchase intention. Results indicate that the main effect for
attachment is significant (p < 0.001), with the endorsement situa-
tion not significant. In other words, attitude towards the ad, atti-
tude towards the brand and purchase intention all are influenced
by attachment but are not significantly influenced by the endorse-
ment situation, showing support for Hypothesis 3a, 3b and 3c.
Mean single endorsement Mean multiple endorsement
2.5812 2.5614
2.6684 2.8091
3.0842 3.0194
3.1987 3.2146
Weak attach/single end mean Weak attach/multiple end mean
2.3900 2.6064
2.9947 2.9384
2.7959 3.2133
 J. Ilicic, C.M. Webster / Australasian Marketing Journal 19 (2011) 230–237 235

Table 3
Main effects and interaction effects.

Dependent variable Source Sum of squares Df Mean square F Sig.

Attitude towards the ad photographica Attachment 11.413 1 11.413 23.401 .000
Endorsement .898 1 .898 1.841 .177
Attachment  endorsement .299 1 .299 .614 .434
Attitude towards the brand photographica Attachment 1.418 1 1.418 6.269 .013
Endorsement .177 1 .177 .783 .377
Attachment  endorsement .000 1 .000 .001 .977
Purchase intention for photographica Attachment 7.865 1 7.865 4.309 .039
Endorsement .019 1 .019 .010 .919
Attachment  endorsement 7.789 1 7.789 4.267 .040

Figs. 2 and 3 graphically show those subjects in the strong attach-

ment condition report a significantly and similarly higher attitude
towards the ad and brand for both the multiple and single endorse-
ment situations than those subjects in the weak attachment condi-
tion, providing support for Hypothesis 4a and 4b. This finding
indicates that when consumers are more attached to a celebrity,
their attitude towards the ads and brands in which the celebrity is
featured do not decline when the celebrity endorses multiple
brands. In fact, attitude towards the ad is higher in the multiple
endorsement situations in both the strong and weak attachment
conditions, again highlighting inconsistencies in the current multi-
ple endorsement literature and rejection of Hypothesis 1a.
Purchase intention is a more complex behaviour. A significant
interaction effect between attachment and endorsement situation
exists for purchase intention. Fig. 4 shows that purchase intention

Fig. 4. Interaction effect of the dependent variable purchase intention.

is most likely in the strong attachment/single endorsement situa-

tion and least likely in the weak attachment/single endorsement sit-
uation. Endorsement situation is found to play a crucial role in
purchase intent, rejecting Hypothesis 4c. When consumers are more
attached to a celebrity and view that celebrity endorsing multiple
brands, this negatively impacts their purchase intention, a finding
consistent with the research undertaken on multiple endorsements
(Mowen and Brown 1981). Yet, when consumers have a weak
attachment to a celebrity, their purchase intention is increased if
the celebrity is seen to endorse multiple brands.
Overall, consumer attachment to a celebrity does impact con-
Fig. 2. Attitude towards the ad. sumer attitude towards the ad, brand and purchase intentions.
Attachment to a celebrity has a significant role in explaining con-
sumer attitude. Consumer attachment to a celebrity results in high-
er attitude towards the ad and attitude towards the brand the
celebrity endorses, regardless of whether the celebrity is seen
endorsing a single brand or multiple brands. Interestingly, attach-
ment also has a positive influence on consumer purchase intention,
but only when the celebrity endorses a single brand. Consumer pur-
chase intention in fact decrease when they see a celebrity to whom
they are strongly attached endorsing multiple brands. On the con-
trary, consumer purchase intention increase when they are weakly
attached to a celebrity and view this celebrity endorsing multiple
brands. The findings suggest that it is a combination of the number
of endorsements made by a celebrity in addition to the degree of
attachment strength that influences consumer purchase intent.

5. Managerial and theoretical implications

Practitioners continually face the challenge of identifying

Fig. 3. Attitude towards the brand. appropriate and effective celebrity endorsers for their brands
236 J. Ilicic, C.M. Webster / Australasian Marketing Journal 19 (2011) 230–237
who have the ability to enhance their brand equity through the
transference of awareness and image associations (Keller, 2008),
and symbolic meanings linked to the celebrity (McCracken,
1989). Previous criteria are based purely on the characteristics of
the celebrity such as their attractiveness (McGuire 1985), familiar-
ity (Kamins, 1990; Misra and Beatty, 1990), likeability (Erdogen,
1999), credibility, believability and expertise (Ohanian, 1990).
However, findings from this study indicate that brand managers
can significantly benefit from selecting celebrity endorsers with
whom consumers have a connection, in terms of an attachment
bond. Results of this study suggest that celebrities to whom con-
sumers are attached may be more effective endorsers of brands
than those celebrities to whom consumers are less attached. When
consumers are attached to a celebrity, they possess significantly
more positive attitude towards the advertisement and brand with
which the celebrity features and a greater likelihood to purchase a
brand when the celebrity is seen only to endorse one brand. Find-
ings from this study suggest that managers can significantly en-
hance the equity of their brands by leveraging the associations of
celebrities to whom consumers of their target market are attached.
Much of the theoretical research within celebrity endorsement
focuses on trustworthiness as a significant factor affecting source
credibility and consequently consumer evaluations of the effec-
tiveness of a celebrity as an endorser (Hovland and Weiss, 1951;
Hovland et al., 1953). Trust is an important positive outcome of
strong celebrity attachment. Although previous research identifies
trustworthiness of the celebrity as an important factor predicting
consumer evaluation of their effectiveness as endorsers, research
to date does not identify how trust can be formed or influenced.
This study confirms Thomson’s (2006) discovery of the exis-
tence of attachment bonds between consumers and celebrities
using both Attachment and Self-Determination theories. The re-
sults of this study suggest that the constructs that explain the de-
gree of relationship strength can be used in the creation and
formation of an attachment towards a celebrity endorser. Brand
managers can benefit from incorporating the dimensions that fos-
ter attachment bonds into marketing communication strategies to
encourage consumer–celebrity connections with those celebrities
who endorse their brands. Strategies to create stronger connec-
tions between consumers and celebrity endorsers should focus
on emphasising the two key basic human needs of autonomy
and relatedness, as these needs are highly related to consumer
attachment to celebrities.
Communication executional frameworks can portray celebrity
endorsers with brands in a way that empowers consumers to ex-
press themselves. If celebrities are perceived as people who value
and appreciate consumers’ expressions of their individuality, then
possibly consumers will develop a sense of freedom to value their
choices with regards to endorsed brands. This will encourage con-
sumers to feel that the celebrity enables them a degree of auton-
omy. Brand managers also can aim to create a connection or
closeness between the endorser and consumers. If celebrity
endorsers are represented as individuals who genuinely care about
consumers, this can encourage a sense of belonging and recogni-
tion, facilitating consumers’ needs for relatedness.
Furthermore, increasing the frequency of interactions between
consumers and celebrity endorsers can perhaps reduce consumer
uncertainty and foster the creation of attachment bonds (Berman
and Sperling 1994). By carefully crafting communication messages
and designing media strategies with appropriate levels of reach
and frequency, the development of attachments to celebrity
endorsers can be fostered and, in turn, alter consumer attitude to
the brands endorsed. By fostering such attachments, brand manag-
ers can then perhaps be able to influence consumers’ degree of sat-
isfaction, trust and commitment, leading to purchase. In effect, the
brand will benefit by encouraging positive consumer attitude and
purchase intention. These strategies may also be effective for cre-
ating an attachment to celebrities who will in the future endorse
a brand.
Although previous studies identify negative effects for brands
that attempt to leverage equity from celebrities who endorse mul-
tiple brands (Mowen and Brown, 1981; Tripp et al., 1994), the find-
ings of this study suggest that consumer attitude towards the
advertisement and brand are positive even when a celebrity is seen
endorsing multiple brands. Purchase intention, on the other hand,
are found to decrease when a strongly attached celebrity is seen to
endorse more than one brand. An increase in consumer purchase
intent is found only when a weak attachment celebrity is seen to
endorse multiple brands. If practitioners are only interested in con-
sumer attitude towards their ads and brands, selecting celebrities
to whom consumers are attached and who have been seen endors-
ing multiple other brands are of no concern. However, if practitio-
ners focus on purchase intention, then care should be taken with
regards to utilising celebrities to whom consumers are attached
who also have multiple brand endorsements.
6. Future research directions
Results from this study show an interesting relationship be-
tween strong consumer attachment to a celebrity and purchase
intention. Although purchase intention is high when a strong
attachment celebrity is seen endorsing a single brand, purchase in-
tent decreases when that celebrity is seen endorsing multiple
brands. On the other hand, purchase intent is found to significantly
increase when a weak attachment celebrity is seen to endorse mul-
tiple brands. Since an interaction effect has occurred, the findings
for purchase intention require further investigation. Future re-
search also is needed to explore the relationship between con-
sumer strong and weak attachment to celebrities and their
purchase intentions for the multiple brands they endorse. For
example, experimental work incorporating purchase simulations
will aid in clarifying the conditions under which multiple endorse-
ments have either positive or negative ramifications for brands.
Although the use of fictitious brands provides a strictly con-
trolled situation that restricts consumer associations in order to
identify the purest assessment of the brand, the use of these brands
also limits the generalisability of the results to new or unknown
brands. Practitioners will benefit from understanding the impact
that consumer–celebrity attachment has on existing brands. Fu-
ture research can investigate the impact of consumer attachments
to celebrities using genuine endorsement situations, with real ads
and real brands. In addition, celebrity associations are not identi-
fied in this study and have the ability to significantly affect con-
sumer brand evaluations. As such, future research can explore
consumer associations tied to celebrities and the influence these
associations have on their evaluations of endorsed brands. It is
likely that attachment dimensions are strongly linked in consumer
knowledge structures for particular celebrities and can be identi-
fied through mapping techniques.
References
Aaker, D.A., Myers, J.G., 1987. Advertising Management, 3rd ed. Prentice-Hall, New
Jersey.
Agrawal, J., Kamakura, W., 1995. The economic worth of celebrity endorsers: an
event study analysis. Journal of Marketing 59 (3), 56–62.
Atkin, C., Block, M., 1983. Effectiveness of celebrity endorsers. Journal of Advertising
Research 23 (1), 57–61.
Berman, W.H., Sperling, M.B., 1994. The structure and function of adult attachment.
In: Sperling, M.B., Berman, W.H. (Eds.), Attachment in Adults: Clinical and
Developmental Perspectives. Guilford Press, New York, pp. 3–28.
Biswas, D., Biswas, A., Das, N., 2006. The differential effects of celebrity and expert
endorsements on consumer risk perceptions: the role of consumer knowledge,
 J. Ilicic, C.M. Webster / Australasian Marketing Journal 19 (2011) 230–237
perceived congruency, and product technology orientation. Journal of
Advertising 35 (2), 17–31.
Bowlby, J., 1979. The Making and Breaking of Affectional Bonds. Tavistock
Publications, London.
Bowlby, J., 1980. Loss: Sadness and Depression. Basic Books, New York.
Choi, S. M. 2002. Attributional Approach to Understanding Celebrity/Product
Congruence Effects: Role of Perceived Expertise. PhD Dissertation, Michigan
State University.
Erdogen, B.Z., 1999. Celebrity endorsement: a literature review. Journal of
Marketing Management 15, 291–314.
Escalas, J.E., Bettman, J.R. 2009. Connecting with celebrities: Celebrity endorsement,
brand meaning, and self-brand connections. Working Paper. <http://elab.
vanderbilt.edu/research/papers/Connecting%20with%20Celebrities%20-%20
Celebrity%20Endorsement,%20Brand%20Meaning,%20and%20Self-Brand%20
Connections%20%5BEscalas,%20Bettman%5D.pdf> (Retrieved 4.09.09).
Fletcher, G., Simpson, J.A., Thomas, G., 2000. The measurement of perceived
relationship quality components: a confirmatory factor analytic approach.
Personality and Social Psychology Bulletin 26 (3), 340–354.
Fournier, S., 1998. Consumers and their brands: developing relationship theory in
consumer research. Journal of Consumer Research 24 (March), 343–373.
Foote, Cone, Belding, 1978. Profiting from Celebrity Endorsers. Foote, Cone, and
Belding Communications Inc., New York.
Hazan, C., Shaver, P.R., 1994. Attachment as an organizational framework for
research on close relationships. Psychological Inquiry 5 (1), 1–22.
Hottest On TV. 2007. Hottest on TV Official Rankings. <www.hottestontv.com.au>
(Retrieved 21.06.07).
Hovland, C.I., Janis, I.L., Kelley, H.H., 1953. Communication and Persuasion:
Psychological Studies of Opinion Change. Greenwood Publishing Group, New
Haven.
Hovland, C.I., Weiss, W., 1951. The influence of source credibility on communication
effectiveness. Public Opinion Quarterly 15 (4), 635–650.
Kahle, L.R., Homer, P.M., 1985. Physical attractiveness of the celebrity endorser: a
social adaptation perspective. Journal of Consumer Research 11 (4), 954–961.
Kaikati, J.G., 1987. Celebrity advertising: a review and synthesis. International
Journal of Advertising 6 (2), 93–105.
Kamins, M.A., 1990. An investigation of the ‘match-up hypothesis’ in celebrity
advertising: when beauty may be only skin deep. Journal of Advertising 19 (1),
4–13.
Kamins, M.A., Gupta, K., 1994. Congruence between spokesperson and product
type: a match-up hypothesis perspective. Psychology and Marketing 11 (1), 4–
13.
Keller, K.L., 2008. Strategic Brand Management: Building, Measuring and Managing
Brand Equity. Prentice Hall, New Jersey.
Kelley, H.H., 1967. Attribution theory in social psychology. In: Levine, D. (Ed.),
Nebraska Symposium on Motivation. University of Nebraska Press, Lincoln, pp.
192–238.
La Guardia, J.G., Ryan, R.M., Couchman, C.E., Deci, E.L., 2000. Within-person
variation in security of attachment: a self-determination theory perspective
on attachment, need fulfilment, and well-being. Journal of Personality and
Social Psychology 79 (3), 367–384.
Lynch, J., Schuler, D., 1994. The match-up effect of spokesperson and product
congruency: a schema theory interpretation. Psychology and Marketing 11 (5),
417–445.
237
Mathur, L.K., Mathur, I., Rangan, N., 1997. The wealth effects associated with a
celebrity endorser: the Michael Jordan phenomenon. Journal of Advertising
Research 37 (May/Jun), 67–79.
McCracken, G., 1989. Who is the celebrity endorser? Cultural foundations of the
endorsement process. Journal of Consumer Research 16 (December), 310–321.
McGuire, W.J., 1985. Attitudes and attitude change. In: Gardner, L., Aronson, E.
(Eds.), Handbook of Social Psychology, vol. 2. Random House, New York, pp.
233–346.
Misra, S., Beatty, S.E., 1990. Celebrity spokesperson and brand congruence: An
assessment of recall and affect. Journal of Business Research 21, 159–173.
Mitchell, A.A., Olson, J.C., 1981. Are product attribute beliefs the only mediator of
advertising effects on brand attitude? Journal of Marketing Research 18
(August), 318–332.
Mooij, M., 2004. Consumer Behavior and Culture, Consequences for Global
Marketing and Advertising. Sage Publications, London.
Mowen, J.C., Brown, S.W., 1981. On explaining and predicting the effectiveness of
celebrity endorsers. Advances in Consumer Research 8, 437–441.
O’Mahony, S., Meenaghan, T., 1997/1998. The impact of celebrity endorsements on
consumers. Irish Marketing Review 10 (2), 15–24.
Ohanian, R., 1990. Construction and validation of a scale to measure celebrity
endorsers perceived expertise, trustworthiness and attractiveness. Journal of
Advertising 19 (3), 39–52.
Ohanian, R., 1991. The impact of celebrity spokespersons’ perceived image on
consumers’ intention to purchase. Journal of Advertising Research 31, 46–54.
Park, C.W., MacInnis, D.J., Priester, J., 2007. Brand attachment and management of a
strategic brand exemplar. In: Schmitt, B.H., Rogers, D.L. (Eds.), Handbook of
Brand and Experience Management. Edward Elgar Publishing, Northampton,
Massachusetts, pp. 3–17.
Pringle, H., Binet, L., 2005. How Marketers can use celebrities to sell more
effectively. Journal of Consumer Research 4 (3), 201–214.
Ryan, R.M., Deci, E.L., 2000. Self-determination theory and the facilitation of
intrinsic motivation, social development, and well-being. American
Psychologist 55 (1), 66–78.
Sherman, S.P., 1985. When you wish upon a star. Fortune (August 9), 66–73.
Thomson, M., MacInnis, D., Park, C.W., 2005. The ties that bind: measuring the
strength of consumers’ emotional attachments to brands. Journal of Consumer
Psychology 15 (1), 77–91.
Thomson, M., 2006. Human brands: Investigating antecedents to consumers’ strong
attachments to celebrities. Journal of Marketing 70 (July), 104–119.
Till, B.D., 1998. Using celebrity endorsers effectively: lessons from associative
learning. Journal of Product and Brand Management 7 (5), 400–409.
Till, B.D., Busler, M., 1998. Matching products with endorsers: attractiveness versus
expertise. Journal of Consumer Marketing 15 (6), 576–586.
Till, B.D., Shimp, T.A., 1998. Endorsers in advertising: the case of negative celebrity
information. Journal of Advertising 27 (1), 67–82.
Time Inc. 2006. The power of celebrity, South Pacific, Australia. <http://
www.timeinc.com.au/news/Corporate_News/24_7_2006_Time_Inc._releases_
groundbreaking_research_study_on_The_Power_of_Celebrity> (Retrieved 21.
02.07)
Tripp, C., Jensen, T.D., Carlson, L., 1994. The effects of multiple product
endorsements by celebrities on consumers attitudes and intentions. Journal of
Consumer Research 20 (March), 535–547.